Antibody-Based Biotherapeutics in Cancer

Abstract

In this chapter, the different classes and characteristics of mAbs used as cancer therapeutics are described. Safety aspects of selected antibodies are discussed as well.

Introduction

More than 80 years ago, in the 1940 ies the first chemotherapeutic drugs were used to support surgery and/or radiation therapy against cancer. In the following decades, chemotherapy developed to a mainstay of therapy rather than being supporting in character. In the last two decades, monoclonal antibodies (mAbs) have faced a growing interest as antitumor therapeutics. In November 1997, rituximab was the first-in-class therapeutic mAb approved by the Food and Drug Administration (FDA). In June 1998 followed the approval in the European Union. At that time, even the manufacturer believed the approval as monotherapy for relapsed/refractory, CD20-positive, low-grade, or follicular B-cell non-Hodgkin’s lymphoma would be a niche indication. The following rapid development of many different mAbs in multiple different cancer indications shows how incorrect this assessment was. Nowadays, a large number of mAbs have been approved for use in numerous cancer indications with many more in different stages of clinical development.

Classification of Monoclonal Antibodies

Structure, functionality, classes, and possible modifications of mAbs and their mode of action are described in Chap. 8. mAbs can be roughly divided into unconjugated antibodies and conjugated antibodies. The latter ones can be subdivided into toxin-coupled antibodies, the so-called chemo-immunoconjugates now designated as antibody–drug conjugates (ADC), and radionuclide-coupled antibodies, the radio immune conjugates (RIC). These “naked,” uncoupled toxins given alone would be too toxic for the patient. The antibody works as a tugboat, tugging the toxin to accumulate selectively nearby or in the tumor. A RIC can be considered “inner radiation therapy.” Figure 23.1 gives a rough systematic overview about the different mAbs used in cancer therapy.

Fig. 23.1

Schematic classification of different mAbs used in cancer therapy; for details, see text

Pharmacological Targets of Oncologic Antibodies

mAbs can be directed against surface structures of malignant cells. These structures comprise receptors or proteins, such as the epidermal growth factor receptor (EGFR) or the cluster of differentiation (CD) antigens. Besides that, some mAbs intercept soluble (growth) factors in the blood and peripheral tissue fluids, which could lead to a tumor survival benefit. One example in oncology is the vascular endothelial growth factor (VEGF), which is intercepted by bevacizumab or aflibercept (“VEGF-Trap”). The most recent and rapidly growing class of mAbs are immune agonistic antibodies, also called immune checkpoint inhibitors (ICIs), currently directed against the cytotoxic T-lymphocyte antigen 4 (CTLA4) and the programmed cell death receptor1 (PD 1) or against its ligand (PD-L1), lymphocyte activation gene-3 (LAG-3) blocking antibodies, and T-cell immunoreceptor with Ig and ITIM domains (TIGIT), which is an immunoreceptor tyrosine-based inhibitory motif. The main difference in the mode of action compared to all the other classes of mAbs and even chemotherapeutics is that not the tumor itself is attacked, but the immune system of the patient is held “under fire” to re-recognize the tumor. Table 23.1 gives an overview over the present mAbs with an oncology indication, ordered by the target structure.

Table 23.1 Therapeutic mAbs used in cancer therapy clustered by their targets (in alphabetical/numerical order)

All mentioned targets can also be found on nonmalignant tissues, which explains the observed toxicity. At present time, none of the approved mAbs is directed against a tumor-specific molecular driver, as they are known, for instance, in Philadelphia chromosome-positive CML (the fusion protein BCR/Abl), in anaplastic lymphoma kinase (ALK)-positive non-small-cell lung cancer (NSCLC), in EGFR-mutated NSCLC (mutated EGFR), or in the FLT3-positive AML. Her2 (= ErbB2), for example, can also be found in normal tissues, but in Her2-positive breast cancers, it is overexpressed.

Pharmacokinetics of mAbs: General Remarks

Unlike small, defined molecules, the pharmacokinetic properties of mAbs differ markedly. Distribution into tissue is slow (internalization) because of the molecular size of mAbs. Volumes of distribution are generally low. Metabolism is similar to catabolic degradation of endogenous and dietetic peptides and proteins. The half-life is usually relatively long, which allows for long dosing intervals during maintenance therapy. Rituximab, for instance, is given every 2 months in the first line setting and every 3 months in the relapsed or refractory situation. Possible factors influencing the pharmacokinetics include the amount of the target antigen, which corresponds to the total tumor mass, immune reactions to the antibody (anti-drug antibodies), and patient demographics such as gender or age (see section on rituximab). Population pharmacokinetic analyses have been applied in assessing covariates in the disposition of mAbs. Both linear and nonlinear eliminations have been reported for mAbs, which is probably caused by target-mediated disposition (overview (Ryman and Meibohm 2017)). If a tumor responds to a mAb therapy, the amount of the remaining target antigen diminishes, and the half-life can be prolonged. However, mAb dosing is based on different pharmacokinetic models, leading to approved dosing strategies of body surface area-based (rituximab), body weight-based (trastuzumab, pertuzumab), or flat dosing (atezolizumab, obinutuzumab).

Anti-CD Antibodies

The cluster of differentiation (also known as cluster of designation or classification determinant and often abbreviated as CD) denotes groups of immune phenotyping surface properties for the identification and investigation of cell surface molecules providing targets on cells in order to classify these cells according to their biochemical or functional characteristics. CD molecules are membrane-bound proteins, often glycosylated, in part cell specific. In terms of physiology, CD molecules can act in numerous ways, often acting as receptors important to cell functioning and/or survival. Others have enzymatic activity. A signaling cascade is usually initiated or modulated, altering the behavior of the cell. Additionally to the mechanisms of ADCC and CDC (Chap. 8), mAbs can alter or even block signal transduction pathways, leading to cell death via apoptosis (overview (Ludwig et al. 2003)).

Anti-CD20 Antibodies

In the 1980s, CD20 was identified as a B-cell marker by Stashenko and coworkers (Stashenko et al. 1980). CD20 is expressed on early pre-B cells, it remains through B-cell development and is then lost from plasma cells (Tedder and Engel 1994). It is not found on hematopoietic stem cells. Therefore, anti-CD20 antibodies are also designated as anti-lymphocytic antibodies. The antigen is neither shed nor internalized. CD20 is expressed on the majority of B-cell lymphomas (Stashenko et al. 1980; Nadler et al. 1981; Tedder and Engel 1994). It is a non-glycosylated member of the membrane-spanning 4-A (MS4A) family (Ishibashi et al. 2001). The antigen consists of three hydrophobic regions forming a tetraspan transmembrane molecule with a single extracellular loop and intracellular N- and C-terminal regions (Tedder et al. 1988). CD20 has been shown to be present on the cell surface as a homo-multimer, in tetramer complexes (Polyak and Deans 2002; Polyak et al. 2008). Despite decades of research, no natural ligand for CD20 could be detected (Cragg et al. 2005). However, subsequent data revealed that CD20 is resident in lipid raft domains of the plasma membrane where it probably functions as a store-operated calcium (SOC) channel following ligation of the B-cell receptor (BCR) with the antigen, implying that it plays a role in regulating cytoplasmic calcium levels after antigen engagement (Bubien et al. 1993; Kanzaki et al. 1995; Li et al. 2003).

Pharmacology. Anti-CD20 antibodies can be classified into type I and type II antibodies with different pharmacodynamics (Cragg and Glennie 2004). Type I (rituximab-like) mAbs induce CD20 to redistribute into large detergent-resistant microdomains (lipid rafts), whereas type II (tositumomab-like) mAbs do not (Deans et al. 1998). These lipid microenvironments on the cell surface—known as lipid rafts—also take part in the process of signal transduction. Lipid rafts containing a given set of proteins can change their size and composition in response to intra- or extracellular stimuli. This favors specific protein–protein interactions, resulting in the activation of signaling cascades. For details, see Simons and Toomre (2000). Rituximab and ofatumumab are type I, whereas obinutuzumab is the first glyco-engineered type II mAb. Type I antibodies display a remarkable ability to activate complement and elicit complement-dependent cytotoxicity (CDC, Chap. 8). This is a result of enhanced recruitment of the complement system component C1q. This ability, however, appears to be directly linked to their (CD20-bound) translocation into lipid rafts, which cluster the antibody Fc regions, thus enabling improved C1q binding (Cragg et al. 2003). Type II mAbs do not show these characteristics: They do not change CD20 distribution after binding; no concomitant clustering is observed, and they are relatively ineffective in CDC. Interestingly, they evoke far more homotypic adhesion and direct killing of target cells in a caspase-independent manner (Chan et al. 2003; Ivanov et al. 2008). In summary, type I mAbs have to cross-link the homo-tetrameric CD20 antigen for a pharmacodynamic effect. In diseases like CLL, with a lower density of CD20 antigens compared with other B-cell lymphomas, the tetramers have to be bridged with more than one molecule of rituximab. That is the explanation why rituximab in CLL is given at a higher dose of 500 mg/m² in subsequent cycles after the usual starting dose of 375 mg/m². For illustration, see Fig. 23.2.

Fig. 23.2

Anti-CD20 mAbs can be divided into two distinct subtypes, termed types I and II. Type I mAbs induce CD20 to redistribute into large detergent-resistant microdomains (rafts), whereas type II mAbs do not. This differential ability of anti-CD20 mAbs to redistribute CD20 in the cell membrane impacts on many of the binding properties and effector functions that control the therapeutic success of anti-CD20 mAbs. Type I and II mAbs have the ability to evoke different effects: type I mAbs engage ADCC and cause CD20 modulation but do not elicit direct cell death, whereas type II mAbs mediate direct cell death and engage ADCC but do not promote CD20 modulation (Beers et al. 2010). In diseases with a low CD20 density, more than one rituximab molecule must bridge the CD20 antigens

Type II mAbs bind within a CD20 tetramer, without lipid raft redistribution. They induce a “closed conformation” (Beers et al. 2010; Niederfellner et al. 2011) (Fig. 23.3). The antigens don’t have to be bridged. The type II mAb-induced cell death is dependent on homotypic adhesion, requires cholesterol, and is energy dependent, involving the relocalization of mitochondria to the vicinity of the cell–cell contact (Beers et al. 2010).

Fig. 23.3

Type II mAbs (tositumomab-like) bind within a CD20 tetramer, and the antigens do not have to be bridged. No redistribution into lipid rafts occurs (Beers et al. 2010, Niederfellner et al. 2011)

The pharmacodynamic mechanism of action and the resulting effects of rituximab are illustrated in Fig. 23.4 (from Stolz and Schuler (2009)).

Fig. 23.4

Overview of rituximab-inducible indirect (ADCC und CDC) effector mechanisms and intracellular, CD20-triggered signal transduction cascades after CD20 cross-linking. FCR is Fc receptor (Stolz and Schuler 2009)

Rituximab

Rituximab is used as a monotherapy as well as in combination with chemotherapy. It has become a mainstay in the treatment of B-cell malignancies such as aggressive and indolent non-Hodgkin’s lymphomas (NHLs) and chronic lymphocytic leukemia (CLL), as well as other B-cell-triggered diseases. It is also widely utilized in an off-label fashion for numerous clinical conditions like immune thrombocytopenia (ITP). Histology subtype has been described as the main tumor parameter that influences rituximab efficacy. In an early trial, when rituximab was given for four doses as monotherapy, small lymphocytic lymphoma histology negatively influenced the objective response rate compared with other low-grade lymphomas (McLaughlin et al. 1998). Several studies have described a low objective response rate in patients with CLL treated with the standard dose of 375 mg/m² rituximab monotherapy used in other NHL (Nguyen et al. 1999; Huhn et al. 2001; O’Brien et al. 2001). As mentioned above, the level of CD20 expression (density) may differ depending on tumor histology (Manches et al. 2003). This was initially used to explain the variation in rituximab efficacy among different histological subtypes. Despite conflicting in vitro data, the dosing in CLL was adjusted to 500 mg/m² for subsequent doses after an initial dosing of the well-known 375 mg/m² finally. The reduced dose for the first infusion is necessary because of a usually high tumor load in need for therapy in CLL blast crisis. Responding to rituximab therapy bears a high risk of tumor lysis and cytokine release syndrome as well as other infusion-related reactions (IRR); Rituximab Safety, section below. For induction therapy as well as for maintenance therapy (used in follicular lymphoma (FL)), a minimum serum concentration has to be achieved for tumor response (Berinstein et al. 1998; Gordan et al. 2005).

Responses to therapeutic mAbs have also been reported to correlate with specific polymorphisms in the Fcγ receptor (FcγR), especially when ADCC is a major part in the mechanism of action (Mellor et al. 2013). These polymorphisms are associated with differential affinity of the receptors to mAbs. At present, these functional FcγR polymorphisms are not suitable to be used as pharmacogenetic biomarkers that could be used to better target the use of mAbs in cancer patients. The available studies do not describe a consistent effect of FcγR genotype on the clinical antitumor activity of therapeutic IgG1 mAbs. Inconsistencies observed in studies are likely related to differences in tumor type, the cytotoxic agents used in combination, the clinical settings, like metastatic vs. adjuvant situation, the clinical benefit parameters measured, the therapeutic antibody used, as well as the magnitude and even the direction of the effect. However, even patients with an “unfavorable” FcγR genotype do benefit from an antibody-containing regimen (overview (Mellor et al. 2013)).

To minimize IRR, the infusion rate is increased slowly from the initial infusion rate of 50 mg/h every 30 min by doubling the subsequent infusion rate if no reactions occur up to 400 mg/h. Shortly before approval of rituximab biosimilars, a fixed dose, subcutaneous (s.c.) formulation was launched in the EU. Rituximab is concentrated up to 120 mg/mL in recombinant hyaluronidase, a penetration enhancer (“spreading factor”). The dose for follicular lymphoma is 1400 mg (12 mL) and for CLL 1600 mg (16 mL). The first dose has still to be given by the i.v. route due to tolerability concerns. The subsequent s.c. doses can be given over 6 min—a substantial advantage from the patient’s point of view.

As shown by the SABRINA study, the pharmacokinetic profile of s.c. rituximab was non-inferior to i.v. rituximab. I.v. and s.c. rituximab had similar efficacy and safety profiles, and no new safety concerns were noted. S.c. administration does not compromise the anti-lymphoma activity of rituximab when given with chemotherapy.

Rituximab Safety

From today’s point of view, dose finding and scheduling of rituximab administration, the first approved therapeutic anticancer antibody, were more empirical than rational or biologic target based. As no dose-limiting toxicity or clear dose–response relationship was found during a phase I study with single doses of 10, 50, 100, 250, or 500 mg/m² (Maloney et al. 1994), the 375 mg/m² dose level with a 4-weekly dosing scheme was chosen for phase II studies (Maloney et al. 1997; McLaughlin et al. 1998) without an apparent reasoning. A dose escalation trial was conducted in CLL, and even at 2250 mg/m² (monotherapy), the maximum tolerated dose was not reached (O’Brien et al. 2001). Rituximab has significant activity in patients with CLL at the higher dose levels. However, the dose–response curve is not steep. Myelosuppression and infections are uncommon.

Rituximab is generally well tolerated in patients with both malignant and nonmalignant diseases, including children (Quartier et al. 2001) and even pregnant women (Herold et al. 2001). Severe adverse events are rare but possible. The most common adverse events are infusion-/foreign protein-related and occur most frequently during or shortly after the first infusion. However, extremely rare, hypersensitivity reactions can occur even after years of maintenance therapy, leading to permanent discontinuation of its use. This syndrome consists of chills, fever, headache, rhinitis, pruritus, vasodilation, asthenia, and angioedema. A cytokine release syndrome is possible. Less often reported are hypotension, rash, bronchospasm, pain at tumor sites, and rash or severe skin toxicity inclusive Stevens–Johnson syndrome and toxic epidermal necrolysis (= Lyell syndrome). Routine premedication consists of corticosteroids, if they are not already a component of the chemotherapy- or antiemetic scheme. The same drug classes are used for intervention in case of hypersensitivity reactions (HSR) despite correct premedication.

Rituximab induces a rapid depletion of CD20⁺ B cells in the peripheral blood. B cells remain at low levels for at least 2–6 months with recovery to pretreatment values occurring within 12 months (Maloney et al. 1997). Other hematological toxicities comprise temporary reduction of platelets or neutrophils and occasionally reduced immunoglobulin levels, which also can be caused by the underlying disease and its bone marrow involvement. Despite the fact that the drug is generally well tolerated, a small number of patients experience unexpected and severe toxicities. It is speculated that these patients have a rituximab serum level dramatically above the average. However, so far, this could not be proven in clinical routine.

Also, gender and age seem to influence rituximab kinetics and clinical outcome. The German High-Grade Non-Hodgkin Lymphoma Study Group first detected in elderly women with diffuse large B-cell lymphoma (DLBCL) a slower elimination of rituximab compared with men, which translated in a better outcome for these women (Pfreundschuh et al. 2014). Consequently, the Study Group conducted a phase II trial increasing the number of rituximab infusions to achieve high rituximab levels early during treatment. In this DENSE-R-CHOP-14 trial, 100 elderly patients with aggressive CD20⁺ B-cell lymphoma received sic cycles of biweekly CHOP-14 combined with 12 × rituximab (375 mg/m²) on days 0, 1, 4, 8, 15, 22, 29, 43, 57, 71, 85, and 99 (CHOP-14 treatment regimen: cyclophosphamide 600 mg/m² day 1; doxorubicin (= hydroxydaunorubicin) 50 mg/m² day 1; vincristine (Oncovin®) 1.4 mg/m² max. 2 mg day 1; predniso(lo)ne 100 mg days 1–5 every 14 days).

This intensification of rituximab administration achieved higher rituximab serum levels and resulted in higher complete remission and event-free survival rates in elderly patients with poor-prognosis DLBCL (Pfreundschuh et al. 2007). The negative impact of male gender in DLBCL was confirmed by a meta-analysis (Yildirim et al. 2015). While body weight contributes to a faster elimination in males according to Muller et al. (2012), others found a significant longer 5-year OS in a high BMI group (>22.55 kg/m²) when compared to that of the low BMI group (Weiss et al. 2017).

With the increasing use of Rituximab, a growing number of rare adverse effects have been recognized. Late-onset neutropenia (LON) is, according to the National Cancer Institute Common Toxicity Criteria, defined as grade 3–4 neutropenia, in which absolute neutrophil count is less than 1.0 × 10³/L, occurring 4 weeks after the last rituximab administration. LON has been reported in 5–27% of rituximab-treated lymphoma patients. Similar figures apply for autoimmune patients, but those appear to have more infections during the neutropenic period (Tesfa et al. 2011; Tesfa and Palmblad 2011).

The mechanisms of LON after rituximab treatment still have not been elucidated. One hypothesis is that LON may result from hematopoietic lineage competition due to an excessive B-cell recovery in the bone marrow by promotion of B-cell lymphopoiesis over granulopoiesis (Anolik et al. 2003). Others see a correlation between a high-affinity FCGR3 158 V allele and LON in lymphoma patients (Weng et al. 2010). A clinically relevant question is if it is safe to re-treat patients with rituximab who previously developed LON. Recurrence of LON episodes upon reexposed patients has been reported. Nevertheless, a close clinical follow-up or complete blood count monitoring is considered adequate in most LON cases (Tesfa and Palmblad 2011).

Following rituximab administration, reactivation of hepatitis B, in some cases resulting in fulminant hepatitis, hepatic failure, and death, has been reported. This led to testing of active replicating hepatitis as well as to screening of all patients for hepatitis B viral infection by measuring HbsAg and anti-HBc before initiating treatment with rituximab as a therapeutic precondition. Monitoring and/or reactivation of antiviral prophylaxis should be considered and is advised by the FDA’s Full Prescribing Information (FPI) and the European Medicines Agency’s Summary of Product Characteristics (SmPC). HBV reactivation can occur up to 24 months following completion of rituximab therapy.

The immune-suppressing antibody with or without chemotherapy can lead to reactivation of latent JC polyoma virus, leading to potentially fatal progressive multifocal leukoencephalopathy (PML). Patients presenting with onset of neurologic manifestations should consult a neurologist with this suspected diagnosis. For natalizumab, an anti-alpha4-integrin antibody used against multiple sclerosis, which induced PML in some patients, removal of the drug via plasma exchange was an important part of the therapy, as it reconstituted the immune system. In a case report, the first successful removal of rituximab with plasma exchange therapy (rituximab apheresis) and accompanying complement-dependent cytotoxicity (CDC) test assay for monitoring has been described. Neurologic symptoms of the patient improved within the first 2 weeks (Burchardt et al. 2012).

Ofatumumab

Ofatumumab is a type I human immunoglobulin (Ig) G1k antibody. Ofatumumab binds with greater avidity than rituximab to another epitope of CD20, which encompasses the small extracellular loop (residues 74–80) and the N-terminal region of the second large extracellular loop (Fig. 23.5). Several crystal structure-based analyses of the Fab fragment of the antibody suggest that ofatumumab can bind closer and tighter to the cell membrane than rituximab, leading to more effective CDC. Induction of CDC appears to be greater than with rituximab and occurs even at a lower density of CD20 on the cell surface (Cheson 2010). After binding to CD20 on malignant B cells, ofatumumab induces clustering of CD20 into lipid rafts, similar to other type I antibodies as described above. In contrast to rituximab, ofatumumab does not induce apoptosis of B-cell lines (Cheson 2010).

Fig. 23.5

Distinct CD20-binding epitopes of rituximab, ofatumumab, and obinutuzumab (designated as GA101). Ofatumumab binds to the so-called small loop. (From Klein et al. (2013)).

Initially licensed for double refractory CLL, that is, those resistant or refractory to fludarabine and alemtuzumab, ofatumumab can be used in previously untreated as well as relapsed patients after conventional chemotherapy. As already mentioned, CLL in need for therapy (blast crisis) bears a risk of rapid cell breakdown in the sense of a tumor lysis syndrome. Therefore, a reduced initial flat dose of 300 mg is given. The subsequent flat dose is 1000 mg or 2000 mg, depending on the clinical situation (for details, see FPI and SmPC).

Ofatumumab Safety

Qualitatively, the toxicity profile of ofatumumab is similar to that of rituximab. This includes hepatitis B virus reactivation and the potential for PML. No additional toxicity or safety signals have been observed so far.

Obinutuzumab

Obinutuzumab is the first glycoengineered, type II humanized anti-CD20 mAb. It is posttranslational defucosylated, resulting in the absence of a fucose sugar residue from IgG oligosaccharides in the Fc region of the mAb molecule (Tobinai et al. 2017). Obinutuzumab binds within the CD20 tetramer, as depicted in Fig. 23.3. CDC, probably, does not contribute to the overall activity of obinutuzumab. The limited capacity of obinutuzumab to fix complement via its Fc portion may further enhance its capability to bind to FcγRIII and mediate ADCC (summarized in Tobinai et al. (2017)). Obinutuzumab was significantly more effective than rituximab in depleting B cells in whole blood samples from healthy donors (n = 10) and from an individual with CLL (Mossner et al. 2010). In CLL, the dosing of obinutuzumab is 100 mg on day 1, followed by 900 mg on day 2 in cycle 1. If tolerated and no IRR occur, the second part of the split dose can be given on day 1. In subsequent cycles, there is no dose splitting on day 1. As mentioned before, starting therapy of CLL may lead to the tumor lysis syndrome. Therefore, the split dose is a safety measure and not necessary in other diseases, e.g., indolent NHL. According to the FPI and SmPC, no dose splitting is advised for follicular lymphoma.

In contrast to rituximab, obinutuzumab is ramped up in cycle 1 with a flat dose of 1000 mg on days 1, 8, and 15. Consequently, patients in head-to-head comparisons of obinutuzumab with rituximab received 3000 mg obinutuzumab in cycle 1, whereas rituximab patients received 375 mg/m². This was often seen as a dosing imbalance. However, the manufacturer did not study a second experimental arm with the same flat dose for rituximab, which would have been considered a real head-to-head study. Different dosing regimens were tested for obinutuzumab. For achieving the desired serum concentration, 1600 mg on days 1 and 8 in cycle 1, followed by 800 mg on day 1 in subsequent cycles, was determined. 1600 mg might need very long infusion times due to possible IRR, especially in CLL. Results in the GAUGUIN study (Morschhauser et al. 2013) and pharmacokinetic modeling showed that obinutuzumab 1000 mg per cycle with additional 1000 mg doses on days 8 and 15 of cycle 1 can achieve exposures similar to the 1600/800 mg regimen. This simplified flat-dose 1000 mg schedule was adopted for subsequent phase II and III investigations (Cartron et al. 2016).

Obinutuzumab Safety

The safety profile of obinutuzumab is similar to the other anti-CD20 antibodies rituximab and ofatumumab.

Y⁹⁰ ibritumomab and I¹³¹ tositumomab tiuxetan are murine RICs. Yttrium-90 is a beta-emitter, while iodine-131 is a beta-emitter and an emitter of gamma radiation. Their activity is mainly achieved through their radioisotopes rather than their intrinsic antibody activity (Fig. 23.6). The radiation can penetrate through several cell layers, which is termed “cross fire” or “bystander” effect (Fig. 23.7). Adjacent cells not marked by the antibody are also affected. Although active, these antibodies never gained a broad acceptance in the medical community. As of February 2014, the production of tositumomab and I¹³¹ tositumomab has been discontinued by the manufacturer and is no longer available (NCI, 2014). Y⁹⁰ ibritumomab tiuxetan is still approved in the EU but is rarely used.

Fig. 23.6

Comparison of conventional radiotherapy (left) with an external radiation source. Radiation from inside (right) by RICs

Fig. 23.7

“Cross fire” or “bystander” effect of radiation, penetrating more cell layers than marked by the antibody

Due to the historical development, up to this point, anti-CD20 mAbs (unconjugated as well as conjugated RICs) were presented together. For the remainder of the chapter, however, ADCs will be discussed separately, even if the therapeutic target of an ADC is the same as that of an unconjugated mAb (e.g., anti-HER2). The mode of action of the payload will be addressed in a separate section but will be cross-referenced in the corresponding target antigen sections.

Other Anti-CD Antibodies: Unconjugated

Alemtuzumab was another anti-lymphocytic antibody, directed against the CD52 antigen. The CD52 antigen is mainly found on mature B and T lymphocytes but can also be found on monocytes/macrophages, NK cells, eosinophils, and epithelial cells of the male reproductive tract. Alemtuzumab was used for the treatment of double refractory B-CLL (refractory to an alkylator plus fludarabine). This hematological approval was retracted in 2012 to favor the development as an immunosuppressant against multiple sclerosis, for which it is currently approved.

Anti-myeloma Antibodies

Elotuzumab

Elotuzumab is an immunostimulatory monoclonal antibody that recognizes signaling lymphocytic activation molecule family member 7 (SLAMF7; CD319), a glycoprotein highly expressed on myeloma and natural killer cells but not on normal tissues (Hsi et al. 2008). Elotuzumab causes myeloma cell death via a dual mechanism of action (Collins et al. 2013) (Fig. 23.8):

• 1. Direct activation: Binding to SLAMF7 directly activates natural killer cells (Collins et al. 2013) but not myeloma cells (Guo et al. 2015).

• 2. Tagging for recognition: Elotuzumab activates natural killer cells via CD16, enabling selective killing of myeloma cells via antibody-dependent cellular cytotoxicity (ADCC) with minimal effects on normal tissue (Collins et al. 2013).

Fig. 23.8

Dual mechanism of action of elotuzumab. (a) Direct activation. (b) Tagging for recognition (Collins et al. 2013)

Elotuzumab is not active as a monotherapy (Radhakrishnan et al. 2017), so it is combined with an immune-modulating drug, lenalidomide or pomalidomide, and dexamethasone. This combination lowers the killing threshold for myeloma cells. Further combinations, for example, with proteasome inhibitors, are currently being tested.

Elotuzumab Safety

Elotuzumab needs an intensive premedication. Despite correct pretreatment with antipyretics, antihistamines, and multiple doses of oral and i.v. corticosteroids, IRRs occur during elotuzumab administration, which makes a close patient monitoring mandatory. A specified infusion rate may be increased in a stepwise fashion as described in the FPI/SmPC. Several interruptions of the administration might be necessary before completion. According to the SmPC, patients experiencing an IRR shall be supervised in regard to their vital parameters every 30 min for 2 h after completing the elotuzumab infusion (EuropeanMedicinesAgency 2022b).

Anti-CD38 Antibodies

Daratumumab

Daratumumab is directed against the CD38 antigen, found on T cells (precursors, activated), B cells (precursors, activated), myeloid cells (monocytes, macrophages, dendritic cells), NK cells, erythrocytes, and platelets. However, current results show that CD38 is not lymphocyte specific. It is ubiquitously expressed in virtually all tissues. It is present not only on cell surfaces but also in various intracellular organelles, including the nucleus (Lee 2006).

CD38 has multiple functions. It acts as a receptor in close contact with the B-cell receptor complex and CXCR4. In engagement with CD31 or hyaluronic acid, it activates NF-kappaB, ZAP-70, and ERK1/2 pathways. It also works as an ectoenzyme. CD38 interacts in a pH-dependent manner (Fig. 23.9) with NAD⁺ and NADP⁺, which are converted to cADPR, ADPR, and NAADP, all intracellular Ca²⁺-mobilizing agents (Malavasi et al. 2011). The ectoenzyme activity of CD38 is independent of its receptor functions (Deaglio et al. 2007).

Fig. 23.9

Multiple pH-dependent enzymatic reactions catalyzed by CD38. cADPR and NAADP are two different Ca²⁺ messengers (Lee 2006) in the sense of “new” second messengers. Abbreviations used: cADPR cyclic ADP-ribose, ADPR ADP-ribose, NAADP nicotinic acid adenine dinucleotide phosphate, ADPRP ADP-ribose-2′-phosphate

Because of marked quantitative differences in expression levels of CD38 between normal cells and leukemic cells, combined with its role in cell signaling, suggests that CD38 is an attractive target for immunotherapy treatment of multiple myeloma (MM) (Malavasi et al. 2008). It is highly and uniformly expressed on myeloma cells (Lin et al. 2004; Santonocito et al. 2004), whereas only a relatively low expression was detected on normal lymphoid and myeloid cells and in some tissues of non-hematopoietic origin (Deaglio et al. 2001). In summary, daratumumab binding to CD38 elicits a signaling cascade and immune effector function engagement, leading to (de Weers et al. 2011):

• Complement-dependent cytotoxicity (CDC).

• Antibody-dependent cell-mediated cytotoxicity (ADCC).

• Antibody-dependent cell-mediated phagocytosis (ADCP), a rather new/unknown mode of action (Overdijk et al. 2015).

• Induction of apoptosis.

• Modulation of cellular enzymatic activities associated with calcium mobilization and signaling.

• Combination of these activities leads to elimination of plasma cells from the bone marrow in MM patients.

Daratumumab also kills myeloid-derived suppressor cells Tregs or negatively regulating T cells.

Isatuximab

Isatuximab is the second approved anti-CD38 antibody with qualitatively the same mode of action (MOA). There are, however, several differences: First, both mAbs bind to CD38 but on different epitopes. Isatuximab and daratumumab recognize 23 and 27 amino acids of human CD38, respectively (Fig. 23.10) (Martin et al. 2019).

Fig. 23.10

Different CD38 epitope binding sites for daratumumab and isatuximab (Martin et al. 2019)

A graphical overview of the mode of action of CD38 mAbs is presented in Fig. 23.11.

Fig. 23.11

Modes of action of anti-CD38 antibodies (de Weers et al. 2011; Overdijk et al. 2015). Antibody-mediated phagocytosis is a rather newly recognized pharmacodynamic process. For details, see text

In vitro measurable differences, without any evidence regarding their clinical relevance, are listed in Table 23.2 (Martin et al. 2019).

Table 23.2 In vivo measurable distinctions between daratumumab and isatuximab. This is explicitly not a comparison in the sense of “is better than” (Martin et al. 2019)

Another difference between these two mAbs is that daratumumab can induce programmed cell death through cross-linking (bridging) several CD38 molecules (compare type I and type II anti-CD20 antibodies in “Rituximab” section), whereas isatuximab exerts a strong pro-apoptotic activity independent of cross-linking. Isatuximab and its F(ab’)2 fragments also induce apoptosis of myeloma cells highly expressing CD38 through the activation of caspases 3 and 7, lysosome-dependent cell death, and by upregulation of reactive oxygen species (Aits and Jaattela 2013). It has also been demonstrated that isatuximab-mediated killing of myeloma cells is enhanced by compounds like pomalidomide (Morandi et al. 2018) (see also “Anti-CD20 Antibodies” section).

Daratumumab and Isatuximab Safety

Both antibodies need a premedication. Despite correct pretreatment with antipyretics, antihistamines, and corticosteroids, IRRs occur during daratumumab and isatuximab administration. Especially for daratumumab, IRRs can occur hours after completing the infusion. The range of IRR onset was up to over 70 h. Administration of an intermediate- or long-acting corticosteroid orally for 2 days starting the day after the administration of daratumumab is recommended for the monotherapy setting. In combination regimens, an intermediate- or long-acting corticosteroid beginning the day after the administration of a daratumumab infusion shall be given. If a background regimen-specific corticosteroid (e.g., dexamethasone, prednisone) is administered the day after the daratumumab infusion, additional corticosteroids may not be needed (JanssenBiotech 2022a, b). While some cancer centers administer montelukast prophylactically in off-label use, others change to s.c. administration (see below).

For patients with a history of chronic obstructive pulmonary disease, prescribing short- and long-acting bronchodilators and inhaled corticosteroids should be considered (JanssenBiotech 2022a, b). A suddenly stuffy nose, cough, throat irritation, allergic rhinitis, and hoarseness during the infusion can be regarded as early signs of an upcoming IRR.

During clinical trials, the median duration of first intravenous (i.v.) infusion of daratumumab with 16 mg/kg was 7.0 h. As longer infusion times to minimize the risk of IRR were impractical, the idea came up to split the very first dose of daratumumab from 16 mg/kg day 1 to 8 mg/kg days 1 and 2 for week 1 in cycle 1. This was justifiable by the PK, investigated through computer simulations for various recommended dosing regimens and in a multi-cohort, phase 1b study. A transient difference in concentration on day 1 in cycle 1 is not expected to have any impact on overall clinical outcomes (Xu et al. 2018). The safety of this modification was evaluated in a retrospective observational study at community-based oncology clinics. These findings provided real-world evidence on the infusion time and safety of the first infusion of daratumumab (Rifkin et al. 2018). This split dosing was approved in the EU (EuropeanMedicinesAgency 2022a).

Due to necessary i.v. infusion interruptions caused by these IRR, a s.c. formulation as described for rituximab was developed to resolve the problem. Dissolved in recombinant hyaluronidase, the fixed dose of 1800 mg daratumumab (15 mL) achieves the same trough level on day 1 of cycle 3, compared to the i.v. dosage form. The IRR rate could be reduced from approximately >30% to 11%. Delayed systemic administration-related reactions could be reduced to 1% (JanssenBiotech 2022a, b). Of note, body weight seems to have an influence on the trough levels and mAb exposition.

Although delayed IRRs were not observed during the ICARIA study with isatuximab, the magnitude of IRRs was similar, with grade 3/4 reactions below 3%. Because CD38 is also expressed on erythrocytes, the indirect Coombs test used prior to blood transfusions is false positive, up to 6 months after the last infusion of anti-CD38 mAbs. Blood typing at baseline is therefore highly recommended before the first daratumumab/isatuximab infusion. Meanwhile, daratumumab and isatuximab in blood samples can be destroyed by incubation with dithiothreitol (DTT) prior to the blood compatibility testing. DTT cracks the structure stabilizing disulfide bridges of the antibody.

All three antibodies elotuzumab, daratumumab, and isatuximab also interfere with serum protein electrophoresis or immune fixation assays (IFA), leading to false-positive results in patients with IgGκ myeloma protein, making the assessment of the initial response difficult. Meanwhile, specific assays were developed to circumvent these problems. The DIRA assay contains murine anti-daratumumab antibodies, which bind daratumumab to keep it out of the M-protein region of the immune fixation assay (van de Donk et al. 2016). For isatuximab, a so-called Hydrashift assay was developed (Finn et al. 2020).

Belantamab Mafodotin

Belantamab mafodotin is the first anti-BCMA-directed monoclonal antibody, a humanized, afucosylated IgG1 anti-B-cell maturation antigen mAb. BCMA is a transmembrane glycoprotein in the tumor necrosis factor receptor (TNFR) superfamily 17 (TNFRSF17). It is expressed at significantly higher levels in all multiple myeloma cells but not on other normal tissues, except normal plasma cells. Its level in patient serum is further correlated with disease status and prognosis. B-cell activating factor (BAFF) and a proliferation-inducing ligand (APRIL; CD256), together with other growth factors, bind to BCMA. BAFF and APRIL are important for the development and function of B cells.

Altogether, there are four mechanisms of action for belantamab mafodotin:

• 1. ADC toxicity (see “Vedotins and Mafodotins in the Auristatins section” in the ADC section).

• 2. ADCC, via FcγRIIIa receptors.

• 3. Immunogenic cell death (ICD)—by cytokines.

• 4. Inhibition of the BCMA-receptor pathway, in the sense of a receptor antagonism.

During ICD, calreticulin is expressed pre-apoptotic on the cell surface, thereby generating an “eat me” signal for dendritic cells (DC). Autophagy facilitates the release of ATP from dying cells, constituting a “find me” signal for DCs and their precursors as well as a proinflammatory stimulus. Additionally, ICD is associated with the post-apoptotic release of the non-histone chromatin-binding protein high-mobility group box 1 (HMGB1). HMGB1 binds to the toll-like receptor 4 on DCs and thus stimulates their antigen-presenting functions (Martins et al. 2014). Belantamab mafodotin induces the ICD (Bauzon et al. 2019) and possesses the abovementioned effector mechanisms (Fig. 23.12).

Fig. 23.12

Mechanism of action of belantamab mafodotin in a clockwise direction. 12 o’clock: belantamab mafodotin binds to cell surface BCMA and is rapidly internalized. Inside the myeloma cell, the payload is released, disrupting the microtubule network, leading to cell cycle arrest and apoptosis. 3 o’clock: the known effector mechanisms ADCC and ADCP. 6 o’clock: ICD; for details, see text. 9 o’clock: classical antagonizing ligand binding to the BCMA receptor, resulting in signal transduction inhibition. Abbreviations used: ADC antibody–drug conjugate, ADCC/ADCP antibody-dependent cell-mediated cytotoxicity/phagocytosis, APRIL a proliferation-inducing ligand, BAFF B-cell activating factor, BCMA B-cell maturation antigen, CRT calreticulin, an “eat me” signal for phagocytes, CTLs cytotoxic T cells, DC dendritic cells, HMGB1 high-mobility group box 1 protein, ICD immunogenic cell death

Belantamab Mafodotin Safety

Likewise other mAbs, IRRs are possible under belantamab mafodotin. However, a premedication is not mandatory. If grade 2 or higher IRRs occur during administration, the infusion rate has to be reduced or the infusion has to be stopped. Supportive care is recommended according to the local institutional rules. The infusion can be resumed with a reduced infusion rate by at least 50% once the symptoms have resolved. If grade 2 or higher IRRs occur, premedication, not otherwise specified, is recommended for subsequent infusions. Not surprisingly, like all other anti-myeloma therapeutics, belantamab mafodotin causes a certain hematotoxicity. New in this context is an eye disorder, designated as keratopathy or corneal adverse reaction, with or without changes in visual acuity. The underlying pathology—an unspecific, reversible off-target mechanism—is described by Farooq et al. (Farooq et al. 2020) and is recommended for the interested reader. Ophthalmic exams are required at baseline, prior to each dose, and promptly for worsening symptoms (GlaxoSmothKline 2020).

Bispecific Antibodies

Blinatumomab

Blinatumomab is a first-in-class bispecific antibody, directed against CD3 on T cells and CD19 on B cells. It is also designated as a BiTE, a bispecific T-cell engager. This antibody construct is made of two single-chain variable fragments (scFv) of anti-CD3/CD19, enabling a patient’s T cells to recognize malignant B cells (Fig. 23.13). CD3 is part of the T-cell receptor. CD19, originally a B-cell marker, is expressed on the majority of B-cell malignancies in normal to high levels. By linking these two cell types, T cells are activated to exert cytotoxic activity on the target cell. The usual clinical use is the Philadelphia chromosome negative B (precursor) ALL.

Fig. 23.13

Two fused scFv of anti-CD3/anti-CD19 antibodies result in the blinatumomab construct. One arm of blinatumomab binds to CD3; the other binds to CD19. This engages unstimulated T cells, which destroys the CD19⁺ cells

Blinatumomab has a molecular weight of 54.1 kDa, approximately one-third of the size of a traditional monoclonal antibody (Wu et al. 2015). It is eliminated rather quickly by the kidneys with a half-life of 1.2 h. Therefore, blinatumomab has to be continuously infused over 1 month.

Blinatumomab Safety

Blinatumomab can cause cytokine release as well as fulminant neurological toxicities even with fatal outcome. Some of them were caused by prescription, calculation, or handling errors. When changing the infusion lines, for instance, the used lines must not be flushed. This would result in an erroneous bolus application with severe toxicities or death. To avoid all this and to make blinatumomab therapy safer, the manufacturer developed risk minimization information brochures for physicians, other healthcare professionals, and patients. The preparation protocol is rather complicated, and a high-precision infusion pump is necessary.

Mosunetuzumab

Mosunetuzumab is a full-length, humanized T-cell-recruiting bispecific antibody targeting CD20-expressing B cells (anti-CD20 x anti-CD3). The antibody contains a N²⁹⁷G amino acid substitution in the Fc region. This substitution results in a non-glycosylated heavy chain that has minimal binding to Fcγ receptors, which reduces Fc effector functions (minimization of cytokine release probability; see below). Similar to blinatumomab, the mechanism of action of mosunetuzumab involves engaging T cells via CD3 with CD20⁺-expressing cells, leading to T-cell activation and T-cell-mediated cytolysis of the CD20⁺-expressing cells. The initial approval for mosunetuzumab is the double relapse/refractory follicular lymphoma after at least two previous systemic therapies.

Mosunetuzumab Safety

A cytokine release syndrome (CRS), like that associated with CAR-T cells, might occur with bispecific mAbs as a consequence of T-cell activation. This reaction can be life threatening. Therefore, patients have to be pretreated with corticosteroids, antihistamines, and antipyretics at a minimum up to cycle 2. An intervention for the treatment of an emerging CRS is, besides corticosteroids, tocilizumab (approved indication for the treatment of CAR-T-cell-induced CRS).

Amivantamab

Amivantamab is a fully human anti-EGFR x anti-MET bispecific mAb. It was designed to engage these two mentioned distinct driver pathways in non-small-cell lung cancer (NSCLC) (Moores et al. 2016; Vijayaraghavan et al. 2020; Yun et al. 2020). By binding to each receptor’s extracellular domain, amivantamab can inhibit ligand binding, promote receptor–antibody complex endocytosis and degradation, and induce antibody-dependent cellular cytotoxicity by natural killer cells and trogocytosis by macrophages. Trogocytosis (derived from the ancient Greek trogo, meaning “gnaw”) is a process whereby lymphocytes conjugated to antigen-presenting cells extract surface molecules from these cells and express them on their own surface. In contrast to phagocytosis, both cells survive this contact.

NSCLC with epidermal growth factor receptor (EGFR) exon 20 insertion mutations exhibits inherent resistance to approved tyrosine kinase inhibitors. Amivantamab with its immune cell-directing activity binds to each receptor’s extracellular domain, bypassing resistance at the tyrosine kinase inhibitor binding site.

Dosing of amivantamab appears a little bit exceptional. Patients with a body weight below 80 kg at baseline receive 1050 mg, and those with equal or greater are dosed with 1400 mg. This two-tiered weight-based dosing was established using population pharmacokinetic analysis. This dosing reduces pharmacokinetic variability and exposure differences. Of note, this dosage is based on baseline body weight and only changed in case of adverse events, as specified by the FPI/SmPC. Amivantamab exhibits linear pharmacokinetics at 350-1,750 mg and nonlinear pharmacokinetics below 350 mg. A dose reduction below 350 mg therefore makes no sense (compare dosing week 1, day 1, Table 23.3). The dose of 1050 mg, however, provides saturation of circulating serum EGFR and MET targets (Park et al. 2021). To reduce the IRR risk, the initial infusion must be split to be administered in two consecutive days (Table 23.3).

Table 23.3 Initial and subsequent dosing of amivantamab depending on baseline body weight

Amivantamab Safety

An IRR prophylaxis has to be administered. Antihistamine and antipyretic prophylaxis is required at all doses. Glucocorticoid is also required at the initial dose (week 1, days 1 and 2), then optional for subsequent doses in the absence of IRR.

The antibody can cause a drug-induced interstitial lung disease (ILD), sometimes called nonbacterial, nonviral pneumonitis. Recognized too late, an ILD can be fatal within a very short time period. Patients have to be monitored for new or worsening symptoms indicative of ILD/pneumonitis.

Ocular toxicity, including keratitis, dry eye symptoms, conjunctival redness, blurred vision, visual impairment, ocular itching, and uveitis, can be associated with amivantamab (keratitis in 0.7%, uveitis in 0.3% of the safety population). All reported events were grades 1–2. Patients presenting with eye symptoms shall be promptly referred to an ophthalmologist.

Other toxicities are consistent with on-target inhibition of the EGFR and MET pathways. EGFR inhibition causes dermatologic adverse reactions as described in the anti-EGFR antibodies section, including the long-eyelash syndrome. MET inhibition causes hypoalbuminemia, typically during the first 8 weeks of treatment, and peripheral edema.

Anti-Growth Factor Receptor Antibodies: Anti-EGFR

Anti-EGFR Strategies

The cell membranous receptors of the epidermal growth factor family consist of four related, functionally different members with tyrosine kinase activity, the homologues EGFR1 to EGFR4 (also called HER1 to HER4 or ErbB1 to ErbB4). The inactive monomers homo- (HER1 with HER1) or hetero-dimerize (HER1 with HER2 or HER3 or HER4) after external ligand binding. Conformational change leads to auto-phosphorylation and subsequent signal transduction for diverse cellular functions including proliferation, cell survival—comprising inhibition of apoptosis, adhesion, and DNA damage repair. In tumors, this pro-survival signal transduction is activated permanently, without binding of a physiological, external ligand such as transforming growth factor-alpha (TGFA), heparin-binding EGF-like growth factor, betacellulin, amphiregulin, epiregulin, epigen, and neuregulins (Rubin and Yarden 2001; Schneider and Wolf 2009; Singh et al. 2016). Diseases with malignancy-associated overexpression of EGFR1 (HER1) are, for example, colorectal carcinomas (CRC), NSCLC, pancreas carcinomas, and squamous cell cancers of the head and neck. Her2 is overexpressed by certain breast and gastric cancers. Over-expression is also known to occur in ovarian (Tuefferd et al. 2007; English et al. 2013; Teplinsky and Muggia 2014), stomach, and aggressive forms of uterine cancer, such as uterine serous endometrial carcinoma (Santin et al. 2008; Buza et al. 2014). HER2 is over-expressed in 30% of salivary duct carcinomas (Chiosea et al. 2015), however, without any therapeutic consequences up to now. Overexpression must be demonstrated by immunohistochemistry (IHC) or fluorescent in situ hybridization (FISH) as a prerequisite for the use of anti-EGFR mAbs.

Cetuximab

Cetuximab is a chimeric IgG1 antibody composed of the Fv region of a murine anti-EGFR antibody with human IgG heavy-chain and kappa light-chain constant regions. It binds to the extracellular domain of EGFR1 with an affinity five- to tenfold greater than endogenous ligands, resulting in inhibition of EGFR signaling. Moreover, it also exerts the cytotoxic immune effector mechanisms described in Chap. 8. It is indicated in the treatment of metastatic colorectal cancer as monotherapy and combination therapy. Cetuximab also binds to a number of EGFR antigen-negative tissues; therefore a saturating loading dose of 400 mg/m² has to be given as a first dose and again, if the weekly interval is exceeded. The weekly follow-up dose is 250 mg/m². Many commonly used chemotherapy regimens for CRC, including irinotecan monotherapy or in combination with infusional fluorouracil, like FolFOx or FolFIri, are administered every 2 weeks (FolFIri treatment regimen (= folinic acid, fluorouracil, irinotecan): irinotecan 180 mg/m² day 1; calcium folinate 400 mg/m² day 1; fluorouracil (5-FU) 400 mg/m² day 1 undiluted bolus injection over 5 min; fluorouracil (5-FU) 2400 mg/m² days 1–2 as 48 h as prolonged infusion; FolFOx treatment regimen (= folinic acid, fluorouracil, oxaliplatin): oxaliplatin 100 mg/m² day 1; calcium folinate 400 mg/m² day 1; fluorouracil (5-FU) 400 mg/m² day 1 undiluted bolus injection over 5 min; fluorouracil (5-FU) 2400 mg/m² day 1–2 as 48 h as prolonged infusion).

The ability to synchronize the administration of cetuximab and concomitant chemotherapy is more convenient for the patients. Dosing of 500 mg/m² every 2 weeks exhibited predictable pharmacokinetics, which were similar to those of the approved weekly dosing regimen. No differences between the weekly and two-weekly regimen on pharmacodynamics were observed. Efficacy and safety were also similar (Tabernero et al. 2008). Cetuximab is also used in the treatment of certain squamous cell cancers of the head and neck in combination with radiation or in combination with a platin-based chemotherapy.

Panitumumab

Panitumumab is the fully human version of an anti-EGFR1 mAb and shares the same target as cetuximab with slightly different binding affinity and specificity. In contrast to cetuximab, no loading dose is necessary—it is continuously dosed with 6 mg/kg q2w.

Both antibodies are only effective in pan-RAS mutation-negative tumors (wild type). K- and N-RAS proteins are G proteins (GTPases) downstream of EGFR and components of EGFR signaling, propagating EGFR signaling events. Activating mutations result in constitutive activation without necessity of ligand binding to the EGFR (Fig. 23.14). In other words, an anti-EGFR antibody therapy for patients with activating RAS mutations has no therapeutic benefit but exposes the patient to the risk of anti-EGFR mAb-related toxicity.

Fig. 23.14

Mutated RAS carcinomas are resistant to anti-EGFR antibodies due to constitutively activated downstream signaling and thus growth promotion

Pan-RAS tissue testing is mandatory according to clinical guidelines (Van Cutsem et al. 2014) before initiating a cetuximab or panitumumab therapy. The benefit of adding cetuximab to RAS wild type was shown by the pooled analysis of the OPUS and CRYSTAL studies (Bokemeyer et al. 2012).

Approximately up to 10% of CRC tumors also carry a BRAF mutation. RAS mutations and BRAF mutations are usually mutually exclusive (De Roock et al. 2010), so double testing is not necessary. A BRAF mutation is a strong negative prognostic biomarker, and patients with a BRAF mutant CRC have a very poor prognosis (Van Cutsem et al. 2011). However, patients with a KRAS wild type but a BRAF mutation benefit from a cetuximab-containing chemotherapy (Van Cutsem et al. 2011).

Side Matters: Location of the CRC. It has been observed for years that patients with right-sided colorectal cancer had worse outcomes than those with left-sided disease (Brule et al. 2015). Meanwhile, right- and left-sided tumors are regarded as molecular distinct (Tejpar et al. 2016). This translates into therapeutic consequences. Whereas treatment of patients with right-sided tumors should be started with an anti-angiogenic component (bevacizumab), patients with left-sided tumors benefit from an anti-EGFR treatment (Venook et al. 2016) (Fig. 23.15).

Fig. 23.15

Median survival dependent on the tumor location and first-line regimen in colorectal cancer (CRC (based on Venook et al. (2016)). KRAS-WT = KRAS wild type, not mutated. HR hazard ratio

Necitumumab

Necitumumab is an anti-EGFR recombinant human monoclonal antibody of the IgG1κ isotype, currently used against NSCLC in combination with cisplatin (75–80 mg/m² on day 1) and gemcitabine 1200–1250 mg/m² on days 1 and 8). It was approved with a statistically significant difference in median overall survival (OS) of 1.6 months, compared to cisplatin and gemcitabine without antibody (11.5 vs 9.9 months). There is a fixed dosing for necitumumab of 800 mg q3w.

Anti-EGFR mAb Safety

Common anti-EGFR therapy associated adverse events include acneiform rash, diarrhea, hypomagnesemia, hypocalcemia, and infusion reactions. Most common are papulopustular rash of the upper trunk and face skin (60–90%), dry and itchy skin (12–16%), and resulting microbial infections (38–70%), conditioned by open pustules as portal of entry for germs. Less frequently, pruritus, hair modifications, and paronychial inflammation occur (Holcmann and Sibilia 2015). By now, the mechanisms underlying skin disorders induced by EGFR inhibitors are well understood (Holcmann and Sibilia 2015). Alterations in chemokine and cytokine production in keratinocytes may result in attraction of inflammatory cells. Disturbed keratinocyte differentiation impairs proper formation of tight junctions and physiological barrier function. The barrier defect and reduced expression of antimicrobial peptides result in bacterial infections. Prophylactic—rather than reactive—management of skin reactions for all patients receiving EGFR inhibitors is recommended. Appropriate prophylaxis can effectively reduce the severity of skin reactions and also a stigmatization as a cancer patient, discernible by their skin eruptions. Skin care has the potential to directly benefit patients and their quality of life. Several recommendations and guidelines have been published (assorted samples: Gutzmer et al. (2011), Lacouture et al. (2011), Potthoff et al. (2011), Hofheinz et al. (2016)). Skin changes during therapy proceed in three phases. Skin care has to be adapted to these phases:

• Phase I: acneiform skin changes.

• Phase II: desiccation phase.

• Phase III: dry, sensitive skin.

During phase III, the skin is very sensitive to sunlight. An unnecessary sun exposure should be avoided, long-sleeved outer clothing should be worn, and the use of sun blockers should be considered. Furthermore, micro-traumatization should be avoided. That is:

• No mechanical manipulation of alleged spots.

• No hot hair drying.

• No use of curlers, especially no tight wrapped curlers.

• No long and hot showering or bathing.

• No vigorous rubbing with towels (hard to comply in case of additional, tantalizing itching).

• No occlusion, for instance, with rubber gloves.

• No too tight footwear.

• Shaving, electric or blade shaving, always causes (unavoidable) micro-traumatization.

Skin cracking that is hardly healing and possibly painful can be sealed with instant glue (like cyanoacrylate glue). This can keep these lesions from worsening or becoming infected. Superglue is also supposed to have some local anesthetic properties (Lacouture et al. 2011).

Electrolyte disturbances (Ca²⁺, Mg²⁺) were initially “simply reported” in the SmPC/FPI. However, especially Mg²⁺ loss occurs frequently. The EGFR is also found in the distal part of the collecting duct and other parts of the kidneys but with a rather high expression in the ascending part of Henle’s loop and in the distal convoluted tubule, where active reabsorption of magnesium ions takes place (~70%). The EGFR regulates the protein TRPM6 (= transient receptor potential cation channel, under family M member 6) (Schlingmann et al. 2007). By blocking this pump with anti-EGFR antibodies, magnesium losses develop (Izzedine et al. 2010). The elimination half-life of cetuximab is approximately 3–7 days. Assuming that the half-life is 3 days and 5 half-lives are necessary to quantitatively eliminate a drug to have no residual pharmacodynamic activity, then one can assume that the EGFR-dependent magnesium pump is more or less permanently inhibited during the weekly dosing schedules. The same applies for panitumumab with its two-weekly schedule and a half-life of 7.5 days. As a consequence, cumulative hypomagnesaemia of grade 3/4 can develop. This is termed magnesium wasting syndrome (MWS). The (randomly) observed frequency of MWS ranges between 27% (Fakih et al. 2006) and 36% (Schrag et al. 2005), although in a prospective cohort study, 97% of the patients were affected (Tejpar et al. 2007). Besides the known symptoms of magnesium loss such as excitability and cardiac effects, a severe fatigue syndrome can be observed. To detect a MWS early, magnesium levels have to be monitored from baseline (as comparative value), because a loss can occur up to 8 weeks after completing therapy. A hypomagnesemia grade 1/2 can be treated by oral substitution. A grade 3/4 hypomagnesemia must be treated by parenteral substitution. Fakih et al. recommend 6–10 g (!) magnesium sulfate daily or twice weekly (Fakih 2008). This magnitude of necessary substitution was confirmed for cetuximab- (Schrag et al. 2005) as well as for panitumumab-induced MWS (Cheng et al. 2009). Such an amount of magnesium sulfate has to be given as a protracted infusion, because of a quick compensatory elimination from short infusions by the kidneys. Ambulatory pumps such as those for infusional 5-FU can be used. In contrast to anti-EGFR mAbs, no magnesium deficiency—not to mention MWS—has been reported with the use of small molecular inhibitors EGFR tyrosine kinase inhibitors such as erlotinib or gefitinib, except in context with diarrhea. During a phase I study with erlotinib combined with the multikinase inhibitor sorafenib, phosphate deficiencies were observed (Izzedine et al. 2010). MWS seems to be a class effect of anti-EGFR mAbs. Detected relatively late after approval of cetuximab and panitumumab, a frequency for hypomagnesemia of 83% is stated in the FPI/SmPC of necitumumab. Meanwhile, electrolyte monitoring (potassium, calcium, magnesium) is recommended to prevent cardiopulmonary arrest.

Hypersensitivity Risks by Tick Bite: An Unexpected Connection with Cetuximab’s Structure. As with other mAbs, hypersensitivity prophylaxis by premedication with an antihistamine is recommended. European product specifications also recommend the use of a corticosteroid premedication. However, severe anaphylaxis during the first exposure to cetuximab has been observed. These reactions to cetuximab developed rapidly, and symptoms often peaked during or within 20 min following the first infusion of the antibody and occasionally proved fatal. A lot of patients also reported to be intolerant to mammalian (red) meat such as beef and pork. These first events were associated with a specific geographical region: a group of southern US states. It was not until 2006 that the true severity of the reactions became obvious. Tick bites can induce an immunological reaction against the oligosaccharide galactose-alpha-1,3-galactose (alpha-gal). The enzyme beta-galactosyl alpha 1,3 galactosyltransferase, needed for the formation of alpha-gal, is inactivated in humans and higher mammals due to an evolutionary process. As a result, immunocompetent individuals may form IgG isotype antibodies to alpha-gal, which makes alpha-gal an immunogenic carbohydrate. The distribution of these antibodies first became clear from the states in which reactions to cetuximab were occurring, i.e., Virginia, North Carolina, Tennessee, Arkansas, Oklahoma, and Missouri (Chung et al. 2008). Subsequently, it has become clear that the syndrome of delayed anaphylaxis to red meat is most common in these same states (Commins et al. 2009). In fact, it was the similarity between the region for reactions to cetuximab and the maximum incidence of Rocky Mountain spotted fever that suggested that tick bites might be relevant to these reactions. The alpha-gal component is also a partial structure Fab fragment of cetuximab’s heavy chain. Each cetuximab molecule contains two alpha-gal epitopes that can cross-link the high-affinity receptor for IgE on mast cells, leading to mast cell activation and release of hypersensitivity mediators (Saleh et al. 2012) (Fig. 23.16). IgE binding to alpha-gal was later linked to allergic reactions to red meat in America and Europe.

Fig. 23.16

Infusion reactions with cetuximab are linked to the presence of IgE antibodies directed against the alpha-gal component of the Fab fragment of the cetuximab heavy chain

Despite the development of an ELISA test for anti-cetuximab IgE for the identification of patients with high risk (Mariotte et al. 2011), no recommendations for pre-therapeutic testing exist up to now. The phenomenon, however, is mentioned in the SmPC. For further reading, consult Chung et al. (2008), Commins et al. (2009), Commins et al. (2011), Saleh et al. (2012), Berg et al. (2014), and Steinke et al. (2015).

Anti-Growth Factor Receptor Antibodies: Anti-EGFR2 (= Anti HER2/neu)

Trastuzumab

Trastuzumab is mainly indicated for HER2/neu overexpressing breast cancers in different clinical situations, as monotherapy as well as in combination therapy. It is also approved for HER2/neu overexpressing gastric cancers in combination with chemotherapy. Overexpression has to be verified by IHC or FISH testing. Dosing depends on the interval. For the weekly regimen, a 4 mg/kg loading dose is necessary, followed by 2 mg/kg weekly. The loading dose for the 3-weekly interval is 8 mg/kg followed by 6 mg/kg. Like rituximab, a fixed dose preparation of s.c. trastuzumab together with recombinant hyaluronidase, 2000 U/mL, is available (trastuzumab and hyaluronidase-oysk in the United States). 600 mg is given q3w over 2–5 min in the thigh. No loading dose is necessary. However, this time-saving procedure compared to i.v. infusions is counteracted by the EMA’s demand to supervise the patients for 6 h (!) after the first and for 2 h after subsequent injections.

Pertuzumab

Pertuzumab is another anti-HER2 antibody. Trastuzumab binds to the HER2 subdomain IV and disrupts ligand-independent HER2 signaling (Fig. 23.17). Pertuzumab binds to subdomain II and blocks ligand-dependent HER2 heterodimerization with HER1, HER3, and HER4. Therefore, pertuzumab has been classified as the first HDI = HER2 dimerization inhibitor. The combination of pertuzumab and trastuzumab significantly augmented antitumor activity in HER2-overexpressing xenograft models, which was the rationale for the development of this double antibody-based anti-Her2 strategy. Improved anticancer activity was proven in patients treated with pertuzumab in combination with trastuzumab compared to either drug alone (Baselga et al. 2012; Cortes et al. 2012).

Fig. 23.17

(a) HER2/3 heterodimerization with subsequently the strongest mitogenic signaling and activation of two key pathways that regulate cell survival and growth. (b) Pertuzumab binds to subdomain II and blocks ligand-dependent HER2 heterodimerization with HER1, HER3, and HER4. (c) Trastuzumab binds to subdomain IV and disrupts ligand-independent HER2 signaling. Trastuzumab in combination with pertuzumab provides a more comprehensive blockade of HER2-driven signaling pathways

Pertuzumab is given as a combination therapy with trastuzumab on a 3-weekly basis (with docetaxel). The initial loading dose is 840 mg followed by 240 mg q3w (trastuzumab: 8 mg/kg loading dose followed by 6 mg/kg q3W). Since 2020, a fixed combination of pertuzumab, trastuzumab, and hyaluronidase for s.c. injection with 1,200 mg pertuzumab, 600 mg trastuzumab, and 30,000 units hyaluronidase for initial dosing and 600 mg pertuzumab, 600 mg trastuzumab, and 20,000 units hyaluronidase for maintenance dosing is available.

Ado-trastuzumab/Trastuzumab Emtansine (T-DM1)

Ado-trastuzumab/trastuzumab emtansine (T-DM1) is the third anti-HER2 antibody, an ADC coupled with a toxin (Fig. 23.24; see “ADC” section of this chapter). Dosing is based on body weight with 3.6 mg/kg q3w.

Trastuzumab Deruxtecan

Trastuzumab deruxtecan (fam-trastuzumab deruxtecan-nxki in the United States) is a HER2-directed antibody with a topoisomerase I inhibitor as payload (see “ADC” section of this chapter).

Margetuximab

The latest anti-Her2 family member is margetuximab. Margetuximab is a chimeric, Fc-engineered, anti-Her2 IgG1 mAb that shares epitope specificity and Fc-independent antiproliferative effects with trastuzumab. Fc engineering of margetuximab alters 5 amino acids from wild-type IgG1 to increase affinity for activating Fcγ receptor FcγRIIIa and to decrease affinity for inhibitory FcγRIIb. Margetuximab was approved on the basis of an international, randomized, open-label, phase 3 study (SOPHIA). This was a direct comparison with trastuzumab, each combined with single-agent chemotherapy, in pretreated patients with Her2⁺ advanced breast cancer. Patients must have had progressive disease after two or more lines of prior Her2-targeted therapy, including pertuzumab, and 1–3 lines of nonhormonal metastatic breast cancer therapy. Allowed combination chemotherapy was capecitabine, eribulin, gemcitabine, or vinorelbine, based on the investigator’s choice. The primary endpoint was PFS. The centrally assessed PFS of margetuximab plus chemotherapy was prolonged for 0.9 months in median, compared with trastuzumab plus CTX (5.8 months vs. 4.9 months with p = 0.03) (Rugo et al. 2021).

Anti-HER2 mAb Safety

The Her2 receptor has essential roles in embryonal and fetal development and tissue protection, i.e., via anti-apoptosis. Particularly neuronal and non-neuronal tissues, including cardiac myocytes, require Her2 for normal development (Zhao et al. 1998; Negro et al. 2004). By blocking these protective functions, the corresponding toxicity, that is, cardiotoxicity, is observed. This applies to all three anti-HER2 mAbs. All of them carry a warning message in their FPI/SmPC for cardiomyopathy and/or the development of a reduced left ventricular ejection function. That means all anti-HER2 antibodies have the potential to cause ventricular dysfunction and congestive heart failure. There is a higher risk for patients who receive anthracyclines, taxanes, or cyclophosphamide in combination and/or during previous therapy. An evaluation of left ventricular function in all patients prior to and during treatment with these mAbs is mandatory. Post approval surveys revealed a potential for lung toxicity, including interstitial lung disease. Some (rare) cases developed respiratory distress syndrome, including some with fatal outcome. Severe IRRs, including hypersensitivity and anaphylaxis, have also been reported with trastuzumab and hyaluronidase-oysk (s.c.). In the HannaH and SafeHER trials, 9% and 4.2% of patients experienced grade 1–4 hypersensitivity and anaphylaxis, respectively, whereas grade 3–4 hypersensitivity and anaphylactic reactions occurred in 1% and <1% of the patients (GenentechInc 2019). Exacerbation of chemotherapy-induced neutropenia can occur with trastuzumab and can also occur when combining pertuzumab and trastuzumab. For T-DM1, hepatotoxicity has been reported, predominantly as asymptomatic elevations of transaminases, and neurotoxicity. This is most probably attributable to MMAE. The most common adverse reactions are infusion reactions (usually mild to moderate), which rarely require discontinuation of therapy. A routine prophylaxis with antihistamines, antipyretics, and/or corticosteroids is not recommended. Skin toxicity like rash, pruritus, and dry skin is described. However, an intense prophylaxis therapy as required for the anti-EGFR mAbs is apparently not necessary. For trastuzumab deruxtecan, there is an explicit warning for the development of an ILD (DaiichiSankyo 2022). Of all ≥ grade 3 toxicities, an ILD was observed in 3% of the patients. Of these, 2.6% had a fatal outcome. Patients should be informed about a potential lung toxicity, monitored for signs and symptoms of ILD/pneumonitis, and should be advised to immediately report cough, dyspnea, fever, and/or any new or worsening respiratory symptoms.

Anti-Angiogenic Antibodies

A small (undetectable) tumor is nourished by passive diffusion. When it has reached dimensions of 1–2 mm across, diffusion is not sufficient any longer. This leads to evolutionary pressure for the tumor and, in the sense of gain of function, it can secrete autocrine pro-angiogenic growth factors such as vascular endothelial growth factor (VEGF). The angiogenic switch has been turned on. The subsequent formation of new blood vessels from preexisting vessels (angiogenesis) is crucial for tumor development, growth, and metastasis (Ohta et al. 1996). VEGF, with its different subtypes (VEGF-A to VEGF-E), is a secretory, pro-angiogenic cytokine and one of the major regulators of the process of neovascularization (Ohta et al. 1996). An important regulating process in angiogenesis is the interaction of the VEGF subtypes with their receptors. Figure 23.18 shows the members of the VEGF family with their receptors and the resulting effects.

Fig. 23.18

VEGF family and their corresponding receptors. PlGF was discovered in human placenta and shares a 50% homology to VEGF-A. VEGF vascular endothelial growth factor, PlGF placenta growth factor, VEGF-R vascular endothelial growth factor receptor, NRP neuropilin

Anti-angiogenic therapy leads to

• Regression of existing microvasculature, known as antivascularization.

• Normalization of the surviving vasculature offering optimal chemotherapy delivery within the tumor, thus enhancing antitumor properties.

• Anti-angiogenic effect leading to the inhibition of new vessel growth, offering improved response rates and tumor death with the potential of improving patient relevant outcomes (progression free survival, PFS; time to progression, TTP; overall survival, OS).

Up to now, there are two principles of anti-angiogenic therapy:

• Intercepting the soluble growth factors in the peripheral blood before they can interact with their receptors (bevacizumab, aflibercept).

• Blocking the receptor (a) by a mAb (ramucirumab) from the extracellular space and (b) by a small-molecule kinase inhibitor from the intracellular space (e.g., axitinib, cabozantinib, nintedanib, pazopanib, regorafenib, sorafenib, sunitinib—not further discussed here).

Bevacizumab

Bevacizumab is directed against VEGF-A and its isoforms. VEGF is intercepted in the peripheral blood and tumor microenvironment and thus neutralized before it can exert pro-angiogenic effects. Bevacizumab is used against several solid tumors like CRC (initial approval), NSCLC, and breast and renal cancer. Dosing depends on the schedule interval and the underlying disease. 5–10 mg/kg q2w and 7.5–15 mg/kg q3w are given.

Ziv-aflibercept/Aflibercept

Ziv-aflibercept/aflibercept (in EU) is a recombinant fusion protein consisting of the VEGF-binding extracellular domain of human VEGF receptors 1 and 2, fused to the Fc part of human IgG1. It binds to all VEGF-A isoforms with a 100-fold higher affinity than bevacizumab. It also binds to VEGF-C and VEGF-D and to PlGF. It is designated as VEGF trap. The clinical use in oncology is restricted to CRC in combination with FolFIri at a dose of 4 mg/kg q2w. A non-oncologic indication of aflibercept (of note the INN) is the treatment of patients with neovascular (wet) age-related macular degeneration by ophthalmic intravitreal injection.

Ramucirumab

Ramucirumab is a recombinant human IgG1 monoclonal antibody that specifically binds to vascular endothelial growth factor receptor 2. Its clinical use comprises NSCLC and gastric and colorectal cancer as single agent (gastric or gastroesophageal junction adenocarcinomas) or in combination with selected chemotherapy regimens. The dosage for gastric cancer/CRC is 8 mg/kg qw2, whereas for NSCLC, it is 10 mg/kg q3w.

Anti-angiogenic mAb Safety

Initially, it was assumed that VEGF-targeted therapies would be toxicity free. However, clinical trials and experience revealed several adverse events associated with anti-angiogenic agents, mAbs, as well as small kinase inhibitors, which can be summarized as a class effect (Hutson et al. 2008; Chen and Cleck 2009). These treatment-emergent adverse events comprise:

• Hypertension.

• Impaired wound healing.

• Hemorrhage, including severe courses.

• Proteinuria.

• Nephrotic syndrome or thrombotic microangiopathy.

• Gastrointestinal perforation.

• Fistula formation.

• Arterial thromboembolic events.

• Grade 4 venous thromboembolic events (including pulmonary embolism).

• Posterior reversible encephalopathy syndrome (PRES), also known as reversible posterior leukoencephalopathy syndrome (RPLS), colloquially referred to as “cortical blindness.”

Pre-therapeutic hypertension has to be adjusted and monitored through therapy. Sometimes, “aggressive” interventions for blood pressure control have to be undertaken, or therapy has to be discontinued. Blood pressure is mainly driven by the VEGF-R2, e.g., by NO and prostacyclin I₂ release from endothelial cells. By blocking VEGF-R2 and intercepting its ligands, the synthesis of vasodilators is suppressed, which in turn increases the peripheral resistance and in consequence the blood pressure (Verheul and Pinedo 2007). Elevated peripheral vascular resistance can also contribute to a reduced left ventricular ejection function and thus be the cause for (congestive) heart insufficiency.

Wound healing depends on neoangiogenesis. Therefore, it is easy to understand that wound healing is impaired by anti-angiogenic therapy. The mentioned mAbs may be used at the earliest 4 weeks after surgery and after complete wound healing. They also must be interrupted for 4 weeks before a planned surgery. Gastrointestinal perforations, especially along surgical sutures, have occurred.

PRES or RPLS is a rare (<1%) but severe side effect of anti-angiogenic or blood pressure-elevating drugs (examples of the latter: proteasome inhibitors). RPLS is a brain–capillary leak syndrome related to hypertension, fluid retention, and the cytotoxic effects of immunosuppressive agents on the vascular endothelium. It seems that severe hypertensive encephalopathy leads to RPLS and vasogenic edema of the posterior cerebral white matter, induced by endothelial dysfunction and a disrupted blood–brain barrier (Verheul and Pinedo 2007; Widakowich et al. 2007). This rare syndrome regained attention after the approval of bevacizumab, but for most of the anti-angiogenic kinase inhibitors, this side effect has been described, as well as for proteasome inhibitors such as carfilzomib, which also elevates blood pressure. Patients who experience a RPLS must never be exposed again to the causative drug.

Toxin-Conjugated Antibodies: Antibody–Drug Conjugates (ADCs)

Antibody–drug conjugates (ADCs) are chemically linked combinations of mAbs and small-molecule drugs with antitumor activity. As mentioned above, the unconjugated drug alone is too toxic for the patient. Therefore, common features of toxin-coupled antibodies are:

• The ADC has to be stable in circulation in vivo.

• The antigen has to be predominantly tumor specific.

• After binding to the tumor antigen, the ADC has to be internalized.

• When internalized, the ADC has to be degradable inside the cell to release the lethal cargo.

The linker between the mAb and toxin can be distinguished into enzymatically cleavable linkers, such as the dipeptide linkers used in brentuximab vedotin, or enzymatically uncleavable linkers, such as the thioether linker in ado-trastuzumab emtansine (trastuzumab emtansine in the EU, syn. T-DM1). Linker technologies are depicted in short in Table 23.4.

Table 23.4 Linker technology and designated release mechanisms, according to Pettinato (2021)

The linker itself contributes to the properties of the ADC including PK and ADME (overview (Han and Zhao 2014)).

Not applying to the ADC characterization are the fusion toxins moxetumomab pasudotox and tagraxofusp. The former is the murine immunoglobulin variable domain of CD22, genetically fused to a truncated form of pseudomonas exotoxin, PE38. The latter is composed of recombinant human interleukin-3 and truncated diphtheria toxin fusion protein, not even a mAb. These fusion proteins are not further discussed in this chapter. Table 23.5 lists the currently used ADCs.

Table 23.5 Currently approved ADCs, in alphabetical order of the INN

Naming of the Conjugate

The name of an ADC consists of a prefix and a suffix. The prefix denominates the mAb component. The suffix consists of a linker (see above) plus the pure payload, i.e., the toxin itself. Sometimes the suffix terms the payload plus a linker plus a necessary spacer and/or attachment group for steric reasons, as exemplarily shown in Fig. 23.19.

Fig. 23.19

The suffix in the naming of an ADC is comprised of the payload itself, a linker, and sometimes a spacer and/or attachment group. Exemplifying the ADC of brentuximab vedotin with MMAE and the attached mAb-linker conjugate via a spacer (para-aminobenzyl carbamate), MMAE is attached to the cathepsin cleavable amino acid VC linker, followed by an attachment group, consisting of maleimide and caproic acid

There are no general guidelines for the designation of ADC suffixes. At the time of writing, for example, in all approved vedotin ADCs (brentuximab vedotin, enfortumab vedotin, polatuzumab vedotin, tisotumab vedotin), the vedotin part consists of maleimidocaproyl-valine-citrulline-p-aminobenzyloxycarbonyl (VC) linker plus MMAE.

Characterization of the Toxins: Mechanisms of Action

Ozogamicins and Tesirines

The payload of the ozogamicins is a cytotoxic enediyne-antibiotic N-acetyl derivative from Micromonospora echinospora ssp. Calichenensis, termed calicheamicin γ₁^I_. The calicheamicin cleaves double-stranded DNA via a radical mechanism. Like bleomycin, it belongs to the so-called free radical-based DNA-cleaving natural products. Calicheamicin binds with sequence specificity to TCCT-, TCTC-, and TTTT <-- in the minor grove of the DNA (minor groove-binding alkylating agent).

Gemtuzumab Ozogamicin

Gemtuzumab ozogamicin is an anti-CD33-directed mAb, coupled with calicheamicin γ₁^I (gamma-one-iodine, Fig. 23.20).

Fig. 23.20

The complex chemical structure of N-acetyl-calicheamicin γ₁^I coupled to an anti-CD33 mAb (symbolized in the upper left; © J. Barth)

Gemtuzumab ozogamicin originally received accelerated approval in May 2000 as a stand-alone treatment for older patients with CD33-positive AML who had experienced a relapse. It was voluntarily withdrawn from the market after subsequent confirmatory trials failed to verify clinical benefit and demonstrated safety concerns, including a high number of early deaths. Gemtuzumab ozogamicin was reapproved in September 2017 with a modified dosing regimen. This approval includes a lower recommended dose, a different schedule in combination with chemotherapy or on its own, and a new patient population (FDA 2017).

Gemtuzumab Ozogamicin Safety. Worrisome adverse events are end organ damages such as veno-occlusive disease or the sinusoidal obstruction syndrome, which can lead to fatalities. Other, in part severe, treatment-emergent adverse events are hemorrhage, mucositis, nausea, vomiting, constipation, headache, increased liver enzymes (AST and ALT), rash, fever, and infection.

Inotuzumab Ozogamicin

Inotuzumab ozogamicin carries the same toxin as gemtuzumab in gemtuzumab ozogamicin, but the mAb component of the ADC is directed against the CD22 antigen on (precursor) B-ALL cells. Safety and treatment emergent adverse events are similar to those of gemtuzumab ozogamicin.

Another group of minor groove-binding alkylating agents are the payloads of the tesirines—a pyrrolobenzodiazepine (PDB) dimer. PDBs are a class of natural products produced by various actinomycetes. The first pyrrolobenzodiazepine antitumor antibiotic, anthramycin, synthesized by Streptomyces refuineus, was discovered in 1965 (Fig. 23.21). PDBs are cardiotoxic, which has precluded their continued clinical use as unmodified molecules.

Fig. 23.21

Above: the PDB dimer active “warhead” used in the ADC loncastuximab tesirine. Below: the naturally occurring precursor anthramycin and the PDB scaffold (formulas © by J. Barth)

PDBs act as sequence-selective DNA alkylating compounds. PBD dimers bind in the minor groove of DNA, where they form covalent amine cross-links between the N2 of guanine and the C11 position of the PBD. The resulting PBD–DNA adducts cause replication forks to stall cell proliferation and tumor cells to arrest at the G2-M boundary, ultimately resulting in apoptosis at low nanomolar to picomolar concentrations (EditorialReviewTeam 2022). A dramatic increase in cytotoxicity and sequence selectivity has been achieved by linking two PBD units to form PBD dimers as cross-linking agents on opposite DNA strands (e.g., interstrand cross-links). Pyrrolobenzodiazepines typically demonstrate IC₅₀ values in the low to mid picomolar range in a variety of cell types in vitro.

Loncastuximab Tesirine

Loncastuximab tesirine is a CD19-directed ADC indicated from third line onward, i.e., after two or more lines of systemic therapy, of adult patients with relapsed or refractory large B-cell lymphoma. This includes diffuse large B-cell lymphoma (DLBCL), not otherwise specified (NOS) DLBCL arising from low-grade lymphoma, and high-grade B-cell lymphoma. NOS means without Myc and BCL2 and/or BCL6 rearrangement, thus no “double-” or “triple-hit” lymphoma. Loncastuximab tesirine is composed of a humanized IgG1κ mAb, conjugated to SG3199 PDB dimer. SG3199 is the most cytotoxic payload employed in a marketed ADC to date. This PBD dimer acts as a minor groove-binding alkylating agent as described above, linked through a protease-cleavable valine–alanine linker to the anti-CD19 mAb. Baseline dosing is 0.15 mg/kg q3w for two cycles. From cycle 3 onward, the dose is bisected to 0.075 mg/kg q3w. For patients with a body mass index (BMI) ≥35 kg/m², an adjusted body weight (ABW) has to be used. The ABW calculates as follows: ABW [kg] = 35 kg/m²× [height in meters]. The monoclonal antibody portion of loncastuximab tesirine is expected to be metabolized into small peptides by catabolic protein pathways. The small-molecule cytotoxin SG3199 is metabolized by CYP3A4/5 in vitro.

Loncastuximab Tesirine Safety

A premedication with dexamethasone 4 mg orally or intravenously twice daily for 3 days beginning the day before loncastuximab tesirine infusion should be given. If dexamethasone administration does not begin the day before the planned infusion, dexamethasone should begin at least 2 h prior to administration of loncastuximab tesirine. Typical for alkylating agents, the PDB dimer is myelotoxic. Patients should be monitored for the development of pleural and pericardial effusions, ascites, and peripheral and general edema. Serious cutaneous reactions occurred in patients treated with loncastuximab tesirine. Grade 3 cutaneous reactions occurred in 4% and included photosensitivity reaction, rash (including exfoliative and maculo-papular), and erythema. Monitoring and dermatologic consultation should be considered.

Vedotins and Mafodotins: Auristatins

Auristatins are dolastatin 10-based analogues. Clinical trials of dolastatin 10 did not progress because of its nonspecific toxicity. The cytotoxic compounds of vedotins and mafodotins are synthetic derivatives of dolastatin 10, which are pentapeptides found in the wedge sea hare, a shell-less mollusk, Dolabella auricularia. They are designated as monomethyl auristatin E (MMAE) and monomethyl auristatin F (MMAF)—mnemonic: vEdotin = MMAE and maFodotin = MMAF (Fig. 23.22). The difference between these two molecules is a phenylalanine present at the C-terminus of MMAF. The charged carboxylic acid terminus limits its passive diffusion into surrounding cells. That means that MMAF has hardly any bystander effect cell killing, whereas MMAE can enter (and leave) cells via passive diffusion (Doronina et al. 2006). Monomethyl auristatins are 100–1,000 times more toxic than doxorubicin (Nilsson et al. 2010). These synthetic agents interact with the vinca alkaloid binding site on α-tubulin and block its polymerization and prevent the formation of the mitotic apparatus (Bouchard et al. 2014), thus inducing cell cycle arrest in G2/M phase, causing cells to undergo apoptosis (Chen et al. 2017).

Fig. 23.22

Above: the synthetic dolastatin derivatives MMAE and MMAF, the latter with an ionizable, C-terminal phenylalanine moiety (for comparison, notice the pink circles). This prevents passive diffusion out of the target cell in contrast to its uncharged counterpart, MMAE. Below: the naturally occurring parent substance dolastatin 10 (formulas © by J. Barth)

MMAE is primarily metabolized by CYP3A4 in vitro, whereas MMAF is mainly hydrolyzed and dehydrated to a cyclized isomeric form. MMAE and MMAF are hemato-, neuro-, and hepatotoxic in general and show some skin and ocular toxicity (see below). Toxicity of the gastrointestinal tract can also occur (Mecklenburg 2018).

Brentuximab Vedotin

Brentuximab vedotin is directed against the CD30 antigen, expressed on classical Hodgkin’s lymphoma Reed–Sternberg and anaplastic large cell lymphoma cells and, in part, mycosis fungoides. It is expressed in embryonal carcinomas but not in seminomas and is thus a useful marker in distinguishing between these germ cell tumors. The dosing is 1.8 mg/kg with a cutoff weight of max. 100 kg. In combination with AVD (= doxorubicin, vindesine, dacarbazine), the dosing is 1.2 mg/kg.

Brentuximab Vedotin Safety

The drug has a myelotoxic potential with possible grade 3/4 neutropenia, thrombocytopenia, and anemia. Neutropenia can result in severe infections and/or opportunistic infections. In the presence of severe renal or hepatic impairment, a higher frequency of ≥ grade 3 toxicities and deaths was observed. Besides severe but rare dermatologic reactions including Stevens–Johnson syndrome and toxic epidermal necrolysis, a cumulative, peripheral sensory neurotoxicity has been observed. This may require a delay, change in dose, or discontinuation of treatment.

Brentuximab and ado-trastuzumab (see below) each carry a toxin that, in case of extravasation, could cause skin necrosis like the vinca alkaloids. As mentioned, the antibodies are conjugated with a cleavable or an uncleavable linker to the toxin. Whether cleavable linkers will be lysed in the extravasation area is unclear at the current time. ADCs with an uncleavable liker will be degraded like proteins, and after that, the toxin is released. If and how this may happen to a clinically relevant degree in an extravasation area is also unclear at the current time. Limited clinical experience so far suggests that extravasated ADCs may only cause slight skin reactions.

Enfortumab Vedotin

Enfortumab vedotin is a fully humanized, IgG1 anti-nectin-4 mAb, conjugated to the microtubule inhibitor MMAE through a protease-cleavable linker. Nectin-4 is a type I transmembrane glycoprotein belonging to the Ig superfamily of proteins. Nectins are primarily involved in the formation and maintenance of adherence and tight junctions [(Takai et al. 2008). Physiologically, nectin-4 is mainly expressed by embryo and fetal tissues, while its expression is low in normal adult tissues (Nishiwada et al. 2015). Pathologically, it has been found to be overexpressed by several cancers and demonstrated to promote tumor growth and proliferation (Takano et al. 2009, Derycke et al. 2010, Athanassiadou et al. 2011). Enfortumab vedotin is indicated for the treatment of adult patients with locally advanced or metastatic urothelial cancer who have previously received a checkpoint inhibitor (anti-PD-1 or anti-PD-L1) and a platinum-containing chemotherapy in the neoadjuvant/adjuvant, locally advanced or metastatic setting. The dose is 1.25 mg/kg up to a maximum dose of 125 mg, accordingly to 100 kg body weight.

Enfortumab Vedotin Safety

Hyperglycemia occurred in patients treated with enfortumab vedotin, including death, and diabetic ketoacidosis. Diabetic ketoacidosis may occur in patients with and without preexisting diabetes mellitus. Blood glucose level monitoring is recommended, especially in patients with, or at risk for, diabetes mellitus or hyperglycemia. Patients should also be monitored for peripheral neuropathy. This neuropathy is predominantly sensory in nature. Prophylactic artificial tears should be considered to minimize ocular disorders like dry eyes. However, vision changes may occur. Ophthalmic topical steroids after an ophthalmic exam might be necessary. Skin reactions can occur during enfortumab vedotin treatment (maculopapular rash, pruritus). Grade 3–4 skin reactions occurred in 10% of patients and included symmetrical drug-related intertriginous and flexural exanthema, bullous dermatitis, exfoliative dermatitis, and hand–foot syndrome. For severe reactions, the treatment has to be interrupted until improvement or resolution.

Polatuzumab Vedotin

Polatuzumab vedotin is an ADC with an anti-CD79b mAb designed for the treatment of DLBCL. The current approval is first-line therapy in combination with CHP. CHP is the cyclophosphamide, doxorubicin, vincristine, and prednisone/prednisolone (CHOP) regimen, where the vinca alkaloid vincristine (O = Oncovin) is replaced by polatuzumab vedotin. For the relapsed/refractory situation, polatuzumab vedotin is combined with bendamustine–rituximab. This means that polatuzumab vedotin is approved in combination with another oncologic mAb, the anti-CD20 antibody rituximab. Anti-CD20 monoclonal antibodies are part of almost any regimen for CD20⁺ B-cell lymphomas. The molecular rationale for combining these two mAbs in DLBCL was elucidated by Kawasaki et al. (2022) in vitro.

In short, both AKT signaling and ERK signaling have been reported to regulate CD20 expression. Immunoblotting analysis identified that AKT and ERK phosphorylation was increased after polatuzumab treatment. However, it was noticed that the anti-CD79b antibody increased the phosphorylation of AKT but inhibited the phosphorylation of ERK. In contrast, MMAE potentiated phosphorylation of ERK but slightly attenuated the phosphorylation of AKT. Taken together, these results suggested that polatuzumab vedotin activates both AKT signaling and ERK signaling and shows a significant effect on the upregulation of CD20 expression levels with an increased sensitivity to rituximab, in terms of complement mediated cytotoxicity and antibody-dependent cellular cytotoxicity. These effects occur as a mechanism of action independent of the payload. Conversely, a decreased activity of polatuzumab vedotin can be deduced if the DLBCL cells display a low density of the CD79b antigen and potentially in a monotherapy setting of polatuzumab vedotin. A sequence-specific administration (polatuzumab vedotin always before rituximab) is especially in the steady-state situation due to the long half-life of polatuzumab vedotin not necessary (adults with non-Hodgkin’s lymphoma 22 days [range 6.1–52]). Dosing is based on body weight with 1.8 mg/kg. Due to limited clinical experience in patients treated with 1.8 mg/kg polatuzumab vedotin at a total dose >240 mg, it is recommended not to exceed the dose 240 mg/cycle according to EuropeanMedicinesAgency (2022d). This recommendation cannot be found in the US FPI. 240 mg corresponds to a body weight of 133.3 kg.

Polatuzumab Vedotin Safety

The payload MMAE is a mitotic spindle poison. Therefore, peripheral neuropathies are typical to the spectrum of side effects. Serious cases of hepatotoxicity, including elevations of transaminases and/or bilirubin, have occurred in patients treated with polatuzumab vedotin. Grade 3 and 4 transaminase elevations develop rarely (1.9% and 1.9%, respectively).

Tisotumab Vedotin

One might be surprised that a fully humanized anti-tissue factor mAb such as tisotumab vedotin is conjugated with the MMAE payload. However, tissue factor (TF) contributes to tumor growth, angiogenesis, metastasis, and thrombosis in patients with cancer. TF expression by tumor cells may increase the growth of tumors by increasing cell survival and/or increasing angiogenesis. TF expression by tumor cells in the blood enhances metastasis by activating coagulation and platelets. It is well known that coagulation in cancer patients is shifted to the procoagulatory side. It is ascertained that release of TF-positive microparticles by tumor cells and host cells into the blood may trigger venous thromboembolism (Fig. 23.23) (Kasthuri et al. 2009).

Fig. 23.23

Interplay of TF tumor growth, angiogenesis, metastasis, and thrombosis in cancer patients. TF expression by tumor cells may increase the growth of tumors by increasing cell survival and/or increasing angiogenesis as well as enhancing metastasis by activating coagulation and platelets. Microparticles released by TF-positive tumor cells and host cells into the blood may trigger venous thromboembolism (Kasthuri et al. 2009)

Preclinical evidence about tisotumab vedotin showed that in contrast to other TF-targeting mAbs, it does not alter coagulation parameters while maintaining high cytotoxic power (Breij et al. 2014). The approved indication is recurrent or metastatic cervical cancer with disease progression on or after chemotherapy. Dosing is 2 mg/kg with a maximum dose of 200 mg, given as an intravenous infusion every 3 weeks.

Tisotumab Vedotin Safety

Because of the broad distribution of TF in the human body, one might expect a broad spectrum of toxicity. Tisotumab vedotin shares the same potential of peripheral neuropathy as all the other MMAE/MMAF conjugates. By targeting a component of the coagulation cascade, there is an elevated risk for bleeding. Hemorrhage occurred in 62% of patients with cervical cancer treated with tisotumab vedotin across clinical trials. Grade 3 hemorrhage occurred in 5% of patients. Severe, life-threatening, or fatal pneumonitis/ILD can occur in patients treated with antibody–drug conjugates containing vedotin including tisotumab vedotin. Patients have to be monitored concerning neuropathies and pulmonary symptoms. Several ocular adverse reactions occurred in 60% of patients with cervical cancer treated with tisotumab vedotin, making a widespread eye care necessary. Besides an ophthalmic exam including visual acuity and slit lamp exam at baseline, prior to each dose, and as clinically indicated, the patient needs to use corticosteroid eye drops for 72 h after each infusion, vasoconstrictor eye drops immediately prior to each infusion, cooling eye pads during the infusion of tisotumab vedotin, and lubricating eye drops for the duration of therapy and for 30 days after the last dose of tisotumab vedotin. The patient also needs to avoid wearing contact lenses unless advised by their eye care provider for the entire duration of therapy (SeagenInc. 2021).

Maytansinoids (DMs)

Maytansinoids (DMs) are thiol derivatives of maytansine (Fig. 23.24). Maytansines bind to tubulin at the rhizoxin/vinblastine binding site and thus inhibit microtubule formation. They are believed to have a high affinity for tubulin located at the ends of microtubules. Their suppression of microtubule dynamics causes cells to arrest in the G2/M phase of the cell cycle, ultimately resulting in apoptosis (Bouchard et al. 2014).

Fig. 23.24

The maytansine parent compound and the thiol derivatives DM1 and DM4. DM1 is known from trastuzumab emtansine (syn. T-DM1). The DM4 compound is under investigation as ravtansine

Trastuzumab Emtansine

Trastuzumab emtansine is an anti-Her2 mAb with DM1 as payload (Fig. 23.25). T-DM1 was the first-in-class ADC approved for the treatment of solid tumors. The ADC treatment of solid tumors fell short due to numerous biological barriers in the tumor microenvironment (e.g., poor vascularization, diffusion through dense stroma, overcoming tumor interstitial fluid pressure). These circumstances limited drug penetration. Unlike hematologic malignancies, the concept of lineage-specific antigen expression (e.g., anti-CD20, anti-CD33) is not applicable to solid tumors. The solid tumor antigens are expressed mainly “tumor associated” rather than “tumor specific” (Criscitiello et al. 2021). This means that the antigens are expressed on tumor cells but also on normal cells, weakly or limited to a given tissue type. This implies that ADCs in these indications have on-target/off-tumor toxicity dependent on the expression of the specific target by normal cells.

Fig. 23.25

The antibody–toxin construct in T-DM1

Camptothecins

Deruxtecan (DXd)

The toxin component of trastuzumab deruxtecan is an exatecan derivative, DX-8951 (syn. DXd). It is a structural analogue of camptothecin, a topoisomerase I (TOP I) inhibitor (see Fig. 23.27), with strong antineoplastic activity. DXd is 10 times more effective than the active metabolite of irinotecan SN-38 (EuropeanMedicinesAgency 2022c). The TOP I inhibitory concentration (IC₅₀) is 2.78 μM for SN38 compared to 0.31 μM for DXd, i.e., ~9-fold lower.

Trastuzumab Deruxtecan

The recommended dosage of trastuzumab deruxtecan (T-DXd) is 5.4 mg/kg given as an intravenous infusion once every 3 weeks (21-day cycle) for unresectable Her2⁺ breast cancer after two or more prior regimens, until disease progression or unacceptable toxicity. Observed with high interest is the apparent potential of T-DXd against brain metastases. This is more or less unexpected because mAbs are believed not to be able to cross the blood–brain barrier to a significant degree. However, a subanalysis of the DESTINY-Breast03 study comparing T-DXd vs. T-DM1 demonstrated superior progression-free survival with T-DXd vs. T-DM1 (Hurvitz et al. 2023) (Fig. 23.26).

Fig. 23.26

PFS curves of a direct comparison of T-DXd vs. T-DM1 in patients with HER2⁺ metastatic breast cancer, with and without brain metastasis at baseline; for details, see Hurvitz et al. (2023)

Govitecan

Govitecan is a hydrolyzable antibody linker, bearing the active metabolite of irinotecan, that is, SN38, a TOP I inhibitor. Figure 23.27 compares the camptothecin payloads and the unconjugated used TOP I inhibitors.

Fig. 23.27

Comparison of the camptothecin parent compound camptothecin (in blue, right below) with the unconjugated TOP I inhibitors topotecan and irinotecan, its active metabolite SN-38, the payload of sacituzumab govitecan and DXd, and the payload of trastuzumab deruxtecan (formula © by J. Barth)

Sacituzumab Govitecan

Sacituzumab govitecan consists of a fully humanized IgG1 anti-Trop-2 antibody conjugated to the payload SN-38, a topoisomerase I inhibitor. Trop-2 (trophoblast antigen 2) is a transmembrane glycoprotein coded by the gene TACSTD2, which primarily acts as intracellular calcium signal transducer (Shvartsur and Bonavida 2015). Trop2 is expressed in several normal tissues, including the skin, uterus, bladder, esophagus, oral mucosa, nasopharynx, and lungs. It is overexpressed in many epithelial tumors, like breast, urothelial, lung, and gynecological and gastrointestinal carcinomas (Stepan et al. 2011). Its overexpression directly promotes tumor growth by inducing several different oncogenic pathways and has been associated to poor prognosis. The ADC gained FDA approval as a treatment for triple-negative breast cancer (TNBC) after at least two prior therapies for metastatic disease with the recommended dose of 10 mg/kg. Other entities than TNBC are under investigation (NSCLC and urothelial carcinoma).

Sacituzumab Govitecan Safety

Just like irinotecan, sacituzumab govitecan can cause severe diarrhea, which can occur early (cholinergic early symptom) or even delayed (cAMP mediated). For the cholinergic early onset, an intervention with atropine is recommended (plus fluids and electrolytes as needed). Antiemetic prevention is necessary and the use of G-CSF as secondary prophylaxis should be considered. Especially patients with a reduced UGT1A1 activity (homozygous for the uridine diphosphate-glucuronosyltransferase 1A1 (UGT1A1*28 allele) are at increased risk for (febrile) neutropenia following initiation of sacituzumab govitecan.

Immune Oncology

Immune Agonistic Antibodies

The group of immune agonistic antibodies are also designated as immune checkpoint inhibitors (ICIs), checkpoint inhibitors (CI), or immune oncologics (IO), expressions that can all be considered synonymous. In principle, the human immune system can prevent tumorigenesis. Tumor cells, however, have the capability to escape the immune system (Vesely et al. 2011). (Re-)activating the immune system to eliminate cancer cells to produce clinically relevant responses has been a long-standing “dream of mankind.” T cells have an important role to play in fighting cancers. To avoid collateral damage of host tissues and organs, T-cell action is balanced by inhibitory signals and molecules—the so-called immune checkpoints. They are critical for maintaining self-tolerance, on the one hand, and modulating the duration and amplitude of immune responses, on the other.

Normally, after T-cell activation, cytotoxic t-lymphocyte antigen 4 (CTLA4; CD152) is upregulated on the plasma membrane. Its functions is to downregulate T-cell activity through a variety of mechanisms, including preventing co-stimulation by outcompeting CD28 for its ligand, B7, and also by inducing T-cell cycle arrest. Through these mechanisms and others, CTLA4 has an essential role in maintaining normal immunologic homeostasis. CTLA4 downmodulates the amplitude of T-cell activation. Blockade of CTLA4 by mAbs such as ipilimumab keeps T cells stimulated (Fig. 23.28).

Fig. 23.28

Simplified mechanism of action of anti-CTLA4 antibodies. Left: T-cell activation by antigen presenting to the T-cell receptor with the co-stimulatory signal B7 to CD28. Middle: Under physiological conditions, the activated T cells are downregulated after 48–72 h by the displacement of costimulatory CD28 with CTLA4. The B7–CTLA4 axis slows down the T cells up to complete inhibition and maintains self-tolerance in the periphery. Right: By blocking CTLA4 with mAbs, the T cell remains activated, attacking tumor tissues. APC antigen-presenting cell, CTLA4 cytotoxic T-lymphocyte antigen 4, MHC major histocompatibility complex, TCR T-cell receptor

While CTLA4 primarily regulates the amplitude of the early stages of T-cell activation, the major role of PD-1 is to limit the T-cell activity in peripheral tissues. Despite its name, programmed cell death protein-1- (PD-1) does not induce cell death directly. When engaged with one of its ligands, PD-L1 or PD-L2, PD-1 inhibits kinases that are involved in T-cell activation. In summary, this pathway is a “stop signal” for T cells. Tumors use these inhibitory pathways to evade an immune attack through overexpression of PD-L1. As a result, this blocks the generation of an immune response to the tumor (cf. Figs. 23.28 and 23.29).

Long-term exposure to antigens in the presence of inflammatory cytokines induces a distinct phenotype in T cells, characterized by loss of effector functions, sustained expression of inhibitory receptors, poor proliferative capacity, and decreased cytotoxic functions. This progressive loss of T-cell effector functions is commonly seen during chronic viral infections and in cancer. It is termed “T-cell exhaustion.” T cells detect the tumor, accumulate in the tumor microenvironment, however, and are silenced by inhibitory molecules expressed on the tumor surface such as PD-L1/2. PD-1/PD-L1 inhibitors pharmacologically prevent the PD-1/PD-L1 interaction, thus facilitating a positive immune response to kill the tumor. For further immune oncology information, the reader is referred to Pardoll (2012), Ferris (2013), Intlekofer and Thompson (2013), Li et al. (2016), Suzuki et al. (2016), Alsaab et al. (2017), and Rotte et al. (2018).

Other known negative or inhibitory checkpoint molecules include:

• Lymphocyte activation gene-3 (AG-3) suppresses an immune response by action on Tregs as well as direct effects on CD8+ T cells.

• T-cell Immunoglobulin domain and mucin domain-3 (TIM-3) is expressed on activated human CD4+ T cells and regulates Th1 and Th17 cytokines. TIM-3 acts as a negative regulator of Th1/Tc1 function by triggering cell death upon interaction with its ligand, galectin-9.

• Indoleamine 2,3-dioxygenase (IDO) is a tryptophan catabolic enzyme with immune-inhibitory properties. IDO is known to suppress T and NK cells, generate and activate Tregs and myeloid-derived suppressor cells, and promote tumor angiogenesis.

Immune oncology mAbs do not show a clear dose–response relationship. They are either dosed by body weight or with a fixed dose. For the future, it may be possible that antibodies currently dosed by body weight may be switched to fixed dosing.

Due to their mechanism of action, infiltrating the tumor tissue followed by tumor cell killing, an initial increase of tumor lesions was observed. From the formal point of view, these patients were RECIST progressive (RECIST = Response Evaluation Criteria In Solid Tumors). However, it would have been a mistake to stop treatment too early, as biopsies confirmed inflammatory cell infiltrates with an apparent enlargement of the tumor. This phenomenon is termed pseudoprogression or “tumor flare” (short overview (Chiou and Burotto 2015)). Thus, to ascertain a clinical benefit, it can take 12 weeks or more after the first infusion and may include the emergence of (temporary) new lesions. To avoid misclassification according WHO in immune oncology, immune-related response criteria (irRC) were developed in 2009 (Wolchok et al. 2009).

Anti-CTLA4 Antibodies

Ipilimumab is the first anti-CTLA4, recombinant human monoclonal antibody. It is used as monotherapy for the treatment of melanoma in different clinical situations. 3 mg/kg q3w for 4 doses is used for unresectable or metastatic melanoma. In a certain adjuvant setting (cutaneous melanoma with pathologic involvement of regional lymph nodes of more than 1 mm that have undergone complete resection, including total lymphadenectomy), 10 mg/kg q3w for 4 doses followed by 10 mg/kg q12w up to 3 years is recommended. Figure 23.28 illustrates the simplified mechanism of action. Tremelimumab is the second approved anti-CTLA4 antibody (only in combination with durvalumab).

Anti-PD-1 and Anti-PD-L1 Antibodies

While development of new anti-PD-1/anti-PD-L1 antibodies is rapidly progressing and will likely provide new agents, the following mAbs were on the market at the time of the creation of this manuscript (Table 23.6).

Table 23.6 Approved anti-PD-1 and anti-PD-L1 antibodies, in alphabetical order per column

The anti-PD-1/anti-PD-L1 antibodies are used against different solid tumors. Depending on the underlying disease and the clinical situation (first-line treatment or relapsed tumor), PD-L1 testing is sometimes mandatory. The relevance and usefulness of PD-L1 as a predicting biomarker are still under debate. Figure 23.29 illustrates the pathological mechanisms and Fig. 23.30 the pharmacodynamic intervention with mAbs.

Fig. 23.29

Naive T cells are activated by the presentation of tumor antigens. During this priming phase, PD-1 is upregulated as a physiological reaction to protect unaffected host tissue. The corresponding ligand, PD-L1, expressed on tumor cells, downregulates T cells, pretending to be “friendly,” healthy tissue. The magnifier shows the details of the T-cell receptor (TCR) interaction with tumor antigen-presenting major histocompatibility complex (MHC). Inactivating programmed cell death receptor 1 (PD-1) is stimulated by the corresponding ligands (PD-L1/2), resulting in T-cell exhaustion. MHC major histocompatibility complex, PD-1 programmed cell death receptor1, PD-L1/2 PD-1/2-ligand, TCR T-cell receptor

Fig. 23.30

Anti-PD-1 and/or anti-PD-L antibodies prevent the inactivation of immune competent T cells

LAG-3 Antibodies

LAG-3 (CD223) stands for “lymphocyte-activation gene 3,″ a transmembrane protein of the immunoglobulin family. It is a co-inhibitory receptor to suppress T-cell activation and cytokine secretion. This ensures a state of immune homeostasis (Andrews et al. 2017). LAG-3 exerts differential inhibitory effects on various types of lymphocytes. The precise molecular mechanisms of LAG-3 inhibitory signaling and interaction with other immune checkpoints are mostly unclear. However, LAG-3 shows a striking synergy with PD-1 (Long et al. 2018). In particular, combination therapy of the anti-LAG-3 mAb relatlimab plus anti-PD-1 mAb nivolumab has shown impressive clinical efficacy in melanoma patients and was approved as a fixed combination consisting of 240 mg nivolumab and 80 mg relatlimab in a single-dose vial. Patients 12 years of age or older receive a total dose of 480 mg nivolumab and 160 mg relatlimab.

Immune Oncology mAb Safety: Playing with Fire?

Due to their mechanism of action, immune checkpoint inhibitors in the form of mAbs against CTLA4, PD-1, PD-L1, and LAG-3 have a unique spectrum of toxicity that differs from the typical adverse events seen with chemotherapeutic agents. By inhibiting checkpoint molecules, the immune system is activated or even over-activated or, from the viewpoint of the normal tissue, deregulated (non-recognizing “self”). These new kinds of autoimmune-like symptoms are based on the loss of self-tolerance and termed immune-related adverse events (irAEs). For CTLA4-blocking antibodies, toxicities seem to be dose related, because the rate of grade 3–4 drug-related serious irAEs increased from 5% to 18% when the dose was increased from 3 to 10 mg/kg, whereas it was 0% at a dose of 0.3 mg/kg. In contrast, the toxicities related to PD-1 blockade are similar at doses ranging from 0.3 to 10 mg/kg, exemplary observed with nivolumab. The observed toxicities depend on:

• The patient population/the underlying disease

• The dose (especially for CTLA4 antibodies)

• The schedule

That means that the same dose or schedule in different diseases might result in different toxicity profiles. Combinational checkpoint blockade can result in an increase of grade 3/4 treatment emergent adverse events (TEAE) up to 53% (Wolchok et al. 2013). The toxicity pattern is based on induced autoimmunity and comprises:

• Skin irAEs, including Stevens–Johnson syndrome and toxic epidermal necrolysis (both being rare).

• Gastrointestinal irAEs—autoimmune colitis.

• Autoimmune hepatitis.

• Autoimmune pancreatitis.

• Autoimmune thyroiditis.

• Autoimmune hypophysitis.

• Neurological irAEs.

• Autoimmune pneumonitis.

• Ocular irAEs (rare), like autoimmune uveitis and autoimmune episcleritis.

In December 2016, the Federal Institute for Drugs and Medical Devices, Germany, informed the medical professional circles about suspected cases of pancytopenia/agranulocytosis. Onset was between 12 and 274 days after start of therapy. Three instances were fatal. Another rare irAE was myocarditis (nine cases), with two of them having an additional myositis and rhabdomyolysis. Four cases had a fatal outcome. Immune-mediated myocarditis and hematologic toxicity were underrated or even unexpected at that time. This opinion is not understandable from the present point of view. A large retrospective analysis of the World Health Organization pharmacovigilance database (VigiLyze) comprising more than 16,000,000 adverse drug reactions, reported from 2009 through January 2018 in VigiLyze, was evaluated. 613 fatalities due to irAEs in patients treated with immune checkpoint inhibitors were detected. Patients were exposed to anti-CTLA4 (ipilimumab or tremelimumab), anti-PD-1 (nivolumab, pembrolizumab), or anti-PD-L1 (atezolizumab, avelumab, durvalumab). The spectrum differed widely between regimens. In a total of 193 anti-CTLA4 deaths, most were usually from colitis (135 [70%]). Anti-PD-1/PD-L1–related fatalities were often from pneumonitis (333 [35%]), hepatitis (115 [22%]), and neurotoxic effects (50 [15%]). Combination PD-1/CTLA4 deaths were frequently from colitis (32 [37%]) and myocarditis (22 [25%]). Myocarditis had the highest fatality rate (52 [39.7%] of 131 reported cases), whereas endocrine events and colitis had only 2–5% reported fatalities; 10% to 17% of other organ–system toxic effects reported had fatal outcomes (Wang et al. 2018). Hematological immune-related adverse events are meanwhile termed as hem-irAEs. Case reports or case series of the following hem-irAEs are published:

• Aplastic anemia/bone marrow failure.

• Hemophilia A.

• Acute thrombosis.

• Large granular lymphocytosis.

• Hemophagocytic lymphohistiocytosis/macrophage activation syndrome.

• Eosinophilia.

• Hematological cytopenias affecting one or more hematological cell lines. Literature reports include cases of ir-neutropenia.

• Autoimmune hemolytic anemia.

• Ir-thrombocytopenia (ir-TCP).

• Pancytopenia.

Awareness about potential myocarditis should be kept in mind. Recommendations for diagnostics can be found in Spallarossa et al. (2019) and Guo et al. (2020). However, in principle, every tissue and organ can be affected (Fig. 23.31).

Fig. 23.31

Affected tissues and organs and symptoms by autoimmune-related adverse reactions. Listing is not intended to be exhaustive

Toxicity/irAE Management

Physicians and pharmacists should know about and be able to recognize irAE as such and the recommended interventions (Weber et al. 2012; Haanen et al. 2015; Weber et al. 2015; Davies et al. 2017).

In severe diarrhea with associated signs of colitis, the usual countermeasures (fluid + electrolytes ⇨ loperamide ⇨ budesonide) are insufficient. Intravenous or oral steroid therapy has to be initiated. For patients in whom intravenous steroids followed by high-dose oral steroid therapy do not lead to initial resolution of symptoms within 48–72 h, treatment with infliximab at 5 mg/kg can be used as an “emergency brake.” Once relief of symptoms is achieved with infliximab, it should be discontinued. A prolonged steroid taper over 45–60 days should be instituted. There may be a waxing and waning of the GI adverse effects. As steroids are tapered, there can be a recrudescence of symptoms, mandating a retapering of steroids starting at a higher dose, a more prolonged taper, and the (re-)use of infliximab. Prophylactic use of budesonide cannot be recommended, based on a phase II study (Wolchok et al. 2010).

Autoimmune hepatitis is likewise treated with (high-dose) corticosteroids. If serum transaminase levels do not decrease within 48 h after initiation of systemic steroids, oral mycophenolate mofetil 500 mg q12h should be considered. Infliximab has to be avoided, because of its potential own hepatotoxicity. As described above, resurgence of symptoms and steroid tapering is necessary.

Difficult in diagnosis can be the auto-hypophysitis with symptoms of headache, nausea, vertigo, behavior change, visual disturbances such as diplopia, and weakness. In this context, new occurrence of brain metastases has to be excluded. Baseline measurement of pituitary, thyroid, adrenal, and gonadal status, i.e., serum morning cortisol, adrenocorticotropic hormone [ACTH], free triiodothyronine [T3], free thyroxine [T4], thyroid-stimulating hormone [TSH], testosterone in males and follicle-stimulating hormone, luteinizing hormone, and prolactin in females, is recommended.

Immune-related pancreatitis generally manifests as an asymptomatic increase of amylase and lipase. Some patients experience more or less unspecific symptoms like fevers, malaise, nausea and vomiting, and/or abdominal pain. As described for other irAEs, a steroid taper is indicated, but often, this has minimal immediate effects. The symptoms of an autoimmune pancreatitis resolve slowly.

The irAEs caused by checkpoint blockade exhibit a characteristic pattern in the timing of their occurrence, shown in Fig. 23.32 (Weber et al. 2012).

Fig. 23.32

irAEs exhibit a characteristic pattern in the timing of their occurrence. The frequency is not considered. (From Weber et al. (2012))

After 2–3 weeks of initiation of a checkpoint blocker therapy, skin-related irAEs can be expected. Gastrointestinal side effects emerge after approximately 5–6 weeks. Hepatic irAEs occur after 6–7 weeks and endocrinological irAEs (autoimmune hypothyroiditis or hypophysitis) after an average of 9 weeks. Long-term effects of immune oncology therapy are currently still largely unknown.

Self-Assessment Questions

Questions

• 1. Against what kind of molecular targets can mAbs be directed? Give examples.

• 2. By which factors can the pharmacokinetics of mAbs be influenced?

• 3. What does ADC stand for?

• 4. Why is rituximab in CLL dosed with 375 mg/m² initially but escalated to 500 mg/m²in subsequent cycles?

• 5. Describe the differences in clinical effects between type I and type II anti-lymphoma antibodies.

• 6. How many subtypes of the epidermal growth factor receptors exist, and how are they designated?

• 7. Which diagnostic actions are mandatory before using anti-EGFR/anti-Her2 antibodies?

• 8. Does it make therapeutically a difference if a colorectal carcinoma is right sided or left sided?

• 9. Does it make sense to combine different mAbs?

• 10. What is the main difference in the mechanism of action between “classical” antibody treatment of cancers and immune agonistic antibodies?

• 11. Describe the toxicity profile of checkpoint inhibitors.

Answers

• 1. mAbs can be directed against:

  • Surface antigens like CD antigens (CD20, CD30)

  • Receptors (EGFR)

  • Growth factors (VEGF)

  • Activating immune checkpoints on T cells (PD-1)

  • Suppressing immune checkpoints on tumor cells (PD-L1)

• 2. Tumor burden that corresponds to the amount of target antigen:

  • Neutralizing anti-mAb-antibodies

  • Gender

  • Age

• 3. ADC is the abbreviation for antibody–drug conjugate, where an antibody is chemically linked to a toxin with antitumor activity. The toxins are too toxic to be given in unconjugated form.

• 4. In CLL, the CD20 antigen density is lower than in other CD20-positive lymphomas. The reduced dose for the first infusion is necessary because of a usually high tumor load in need for therapy in CLL blast crisis. Responding to rituximab therapy bears a high risk of tumor lysis and cytokine release syndrome as well as other infusion-related reactions.

• 5. Type I antibodies cause predominant CDC, while type II antibodies ADCC.

• 6. Four subtypes, EGFR1–4 or ErbB1–4 or Her1–4

• 7. Verification of EGFR or Her2 overexpression, in case of Her2 with IHC or FISH. Pan-RAS wild-type testing for cetuximab and panitumumab.

• 8. Depending on the localization, these tumors are regarded as biologically different. Patients with a right-sided tumor benefit from an anti-angiogenic drug containing first-line therapy, whereas left-sided tumors benefit from an anti-EGFR strategy if they are not RAS mutated.

• 9. Yes, for instance, trastuzumab plus pertuzumab, polatuzumab vedotin plus rituximab, and nivolumab plus relatlimab.

• 10. The so-called checkpoint-inhibiting antibodies do not attack the tumor itself; they restore their immunogenicity and/or (re-)activate immune-competent cells.

• 11. The toxicity pattern consists of loss of self-tolerance, resulting in autoimmune reactions (immune-related adverse events) such as:

  • Skin irAEs, including Stevens–Johnson syndrome and toxic epidermal necrolysis (both being rare)

  • Gastrointestinal irAEs—autoimmune colitis

  • Autoimmune hepatitis

  • Autoimmune pancreatitis

  • Autoimmune thyroiditis

  • Autoimmune hypophysitis

  • Neurological irAEs

  • Autoimmune pneumonitis

  • Ocular irAEs (rare), such as autoimmune uveitis and autoimmune episcleritis