Keywords

Introduction

Up to 85% of patients with pancreatic cancer have unresectable disease at the time of diagnosis, either due to local tumor invasion or metastatic spread. As is the case for many other solid tumors, cytotoxic chemotherapy remains the mainstay of treatment for this group of patients. Goals of systemic therapy include prolongation of survival and palliation of symptoms. Occasionally, patients with locally advanced disease may become eligible for surgical resection after initial systemic therapy. Improvements in chemotherapy regimens over the past decade have led to important, albeit modest, gains in the survival of patients with advanced pancreatic cancer. Below we will review treatment approaches from the medical oncology perspective for patients with locally advanced and metastatic disease.

Locally Advanced Disease

Approximately one-third of patients present with unresectable, locally advanced disease at the time of diagnosis with pancreatic cancer. While several definitions of the stages of resectability have been developed over the past decade and a half, each centers on the relationship between the tumor and surrounding vasculature [1,2,3]. The MD Anderson Criteria for resectability for pancreatic cancer define locally advanced disease as tumors that either encase the superior mesenteric artery, the celiac axis, or hepatic artery without a reconstructive option or occlude the superior mesenteric or portal veins without a reconstructive option (Fig. 9.1) [2].

Fig. 9.1
A contrast-enhanced C T scan depicts the pancreas. The encircled area marks the tumor arising while the arrows indicate the celiac artery which appears wrapped by the tumor that makes it irresectable.

Locally advanced pancreatic cancer. Tumor arising from head and body of pancreas is circled. Arrow indicates celiac artery which is being encased by tumor, rendering it unresectable

Full size image

Given the contraindication to surgery and the high probability of occult metastatic disease in this population, systemic chemotherapy is the recommended first-line treatment in patients with locally advanced pancreatic cancer (LAPC). The GERCOR group retrospectively analyzed patients with LAPC treated in their trials with first-line chemotherapy, which was either 5-fluorouracil (5-FU) + leucovorin + gemcitabine or gemcitabine + oxaliplatin [4]. They found that 29% of patients developed disease progression during the first 3 months of chemotherapy, a subset that had a median overall survival of 4.5 months. This data highlights the heterogeneity in aggressiveness of LAPC. Upfront systemic therapy identifies patients with particularly aggressive biology as those who would not benefit from local therapy such as radiation.

Guidelines on the optimal initial chemotherapy regimen for patients with LAPC rely primarily on data from patients with metastatic disease. Combination regimens such as FOLFIRINOX and gemcitabine + nab-paclitaxel are frequently utilized in this setting, owing to their proven efficacy in phase 3 studies of patients with metastatic pancreatic cancer [5, 6]. A 2016 systematic review evaluated the role of FOLFIRINOX in LAPC in 315 patients across 11 studies [7]. The authors reported a median overall survival of 24.2 months (95% CI 21.7–26.8) and median progression-free survival (PFS) of 15.0 months (95% CI 13.8–16.2). The pooled proportion of patients with LAPC who underwent resection was 25.9%, with 78.4% having an R0 (microscopically negative) resection.

Gemcitabine-based combination regimens also have limited data in the locally advanced setting. A phase 2 study in Austria evaluated gemcitabine + oxaliplatin as “neoadjuvant” therapy in patients with LAPC [8]. Of the 33 patients in their study, 13 underwent resection with a 69% R0 resection rate. The patients who underwent resection had a median OS of 22 months, compared to 12 months for those who did not undergo resection. An important caveat to these results is that 15 of the 33 patients were considered to have borderline resectable disease at the time of diagnosis when evaluated by centralized imaging review. The results of the multicenter phase 2 LAPACT trial which evaluated the combination of induction gemcitabine + nab-paclitaxel in patients with LAPC were reported in 2018 [9]. One hundred seven patients were included in the study, and the median PFS was 10.2 months. Neutropenia was the most common (42%) grade 3 or 4 adverse event, but the authors found the regimen to be largely tolerable. Sixteen patients (15%) in this study underwent surgical resection with 7 having an R0 resection.

For patients who are not eligible for combination chemotherapy regimens due to comorbidities or poor performance status, single agent gemcitabine is a reasonable alternative. Evidence for the use of gemcitabine monotherapy in this setting is largely extrapolated from older clinical trials that included patients with LAPC and metastatic disease [10,11,12]. Results published from the LAP07 trial, which studied the benefit chemoradiation compared with continuation of chemotherapy in patients with LAPC, demonstrated that induction chemotherapy with single agent gemcitabine resulted in a median overall survival of 13.6 months [13]. Radiotherapy with or without concurrent chemotherapy is another first-line option for patients who cannot tolerate combination chemotherapy, an approach that will be discussed later.

Despite improvements in response rates with combination chemotherapy, most patients who are initially considered to have LAPC never become candidates for curative surgical resection. However, patients who demonstrate response to induction systemic therapy without the development of overt metastatic disease should be re-evaluated for the possibility of resection by an experienced multi-disciplinary team. This is an approach supported by both the National Comprehensive Cancer Network (NCCN) and the American Society of Clinical Oncology (ASCO) [14, 15]. A study of 415 patients with LAPC treated at Johns Hopkins from 2013 to 2017 found that 84 (20%) underwent surgical resection after a median of 5 months of pre-operative therapy. Median OS was higher in the resected cohort at 35.3 months, compared to 16.3 months in the non-resected patients. They also found that the use of FOLFIRNOX and stereotactic body radiation therapy (SBRT) was associated with an increased probability of surgical resection.

The German phase 2 NEOLAP study randomized 130 LAPC patients at 33 different institutions who had not progressed on 2 initial cycles of gemcitabine + nab-paclitaxel to either continue the doublet for 2 more cycles or switch to 4 cycles of FOLFIRINOX [16]. The study’s results, presented in 2019, included a primary endpoint of conversion rate to resectability. The conversion rate in the gemcitabine + nab-paclitaxel group was 30.6%, compared to 45.0% in the FOLFIRINOX group (p = 0.135). Median OS was not different between the two chemotherapy arms, but conversion to resectability was associated with an improved median OS (27.4 vs 14.2 months, p = 0.0035).

Determining a patient’s potential for downstaging and future resectability at the time of diagnosis with LAPC is very challenging. An attempt has been made by clinicians at the Medical College of Wisconsin to categorize LAPC patients into “type A” (potentially resectable) or “type B” (very likely not resectable) based on vascular involvement and the potential for surgical resection after neoadjuvant therapy [17]. An analysis of 108 consecutive patients with LAPC who were categorized into type A or B under this schema found that 62% of type A and 24% of type B patients underwent surgical resection [18]. The authors concluded that such a classification may help establish appropriate expectations and goals of care with patients.

Patients who continue to have stable but unresectable disease after induction chemotherapy of 4 to 6 months should be considered for either continuation of systemic therapy, a treatment break, or consolidative chemoradiotherapy. This decision should be made taking into account the patient’s goals of care, performance status, and symptoms. ASCO and NCCN guidelines recommend observation for this population in the absence of a clinical trial. For patients who are treated with chemoradiotherapy, there may be some benefit to maintenance systemic chemotherapy if their disease continues to be localized [19].

The role of radiation therapy in locally advanced disease will be discussed in greater depth later in this chapter. For LAPC patients who are treated with radiation, the addition of concurrent chemotherapy as a sensitizing agent has become commonplace. A 2009 systematic review analyzed the benefits of chemoradiotherapy in 21 published studies and concluded that chemoradiotherapy increases OS compared to radiotherapy alone, at the cost of higher toxicity [20]. The SCALOP trial was a randomized phase 2 study comparing gemcitabine-based chemoradiotherapy to capecitabine-based chemoradiotherapy for patients who had not developed disease progression during 12 weeks of induction chemotherapy with the combination of gemcitabine and capecitabine [21]. Seventy-four LAPC patients were ultimately randomized to one of the chemoradiotherapy groups in the study, and median OS was improved with capecitabine compared to gemcitabine (15.2 vs 13.4 months, hazard ratio [HR]: 0.39, p = 0.012). The adverse effect profile also favored capecitabine-based chemoradiotherapy, though quality of life scores were not different between the arms.

Metastatic Disease

The medical management of a patient with metastatic pancreatic cancer is frequently complex and often requires attention to pain control, thromboembolic disease, biliary or gastrointestinal obstruction, infection, pancreatic exocrine insufficiency, and anorexia/weight loss, in addition to side effects from chemotherapy (Fig. 9.2). However, administration of chemotherapy has been shown to have the dual benefit of prolonging survival and improving quality of life [22, 23]. The two most commonly utilized first-line chemotherapy regimens are FOLFRINOX and gemcitabine + nab-paclitaxel (Table 9.1).

Fig. 9.2
Two contrast-enhanced C T scans. The C T scan on the left depicts three encircled areas that mark the hypodense prominence of the pancreatic head on its left, while the right image encircles the tumor that rises from the tail of the pancreas.

Metastatic pancreatic cancer. Left: CT image of hypodense liver lesions consistent with metastatic spread (circled). Right: CT image of tumor arising from tail of pancreas (circled). Patients with tumors of the pancreatic tail often present with metastatic disease

Full size image
Table 9.1 First-line chemotherapy regimens for metastatic pancreatic cancer
Full size table

FOLFIRINOX was established as a standard-of-care first-line regimen based on data from the large phase 3 study conducted by Conroy et al. comparing the combination with gemcitabine monotherapy, the previous standard-of-care [5]. A total of 342 patients with metastatic pancreatic cancer were randomized to either FOLFIRINOX or gemcitabine. Patients were required to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 and adequate liver and renal function. Compared to gemcitabine, FOLFIRINOX improved median OS (11.1 vs 6.8 months, HR, 0.57, p < 0.001). The FOLFIRINOX arm also demonstrated an improved objective response rate (31.6% vs 9.4%) and median PFS (6.4 vs 3.3 months, HR, 0.47, p < 0.001). Importantly, rates of grade 3 or 4 adverse events were higher in the FOLFIRINOX group, including neutropenia (45.7% vs 21.0%) and diarrhea (12.7% vs 1.8%).

Two years after the data on FOLFIRINOX was published, the results of the phase 3 MPACT trial were published, comparing gemcitabine plus albumin-bound paclitaxel (nab-paclitaxel) to gemcitabine monotherapy in 861 patients. Similar to FOLFIRINOX, this doublet improved median OS compared to gemcitabine (8.5 vs 6.7 months, HR, 0.72, p < 0.001). Median PFS was 5.5 months with gemcitabine + nab-paclitaxel compared to 3.7 months for the single agent (HR: 0.69, p < 0.001). As expected, grade 3 or 4 adverse effects were higher in the doublet arm including neutropenia (38% vs 27%), fatigue (17% vs 7%), and neuropathy (17% vs 1%).

The selection of which first-line chemotherapy, if any, to utilize depends on patient comorbidities and performance status. For patients with an excellent performance status, either regimen is appropriate. Most clinicians consider gemcitabine + nab-paclitaxel to be a more tolerable therapy and may recommend the doublet for a patient with a borderline performances status. Single agent gemcitabine represents another option for patients whose performance status or comorbidities preclude a combination regimen [10, 24]. Comorbid conditions such as pre-existing peripheral neuropathy may necessitate treatment that omits platinum or taxanes, such as FOLFIRI or gemcitabine alone. Additionally clinicians should be cautious about prescribing irinotecan and potentially gemcitabine in patients with hepatic impairment. Despite a cutoff of 75 years in the pivotal 2011 study leading to the approval of FOLFIRINOX, age, alone, should not necessarily dictate which first-line chemotherapy is selected, as performance status is often a better predictor of tolerability [25].

Patients who maintain an ECOG performance status of ≤2 may be considered for second-line therapy. NAPOLI-1 was a phase 3 study of second-line therapy in 417 patients with metastatic pancreatic cancer whose disease had progressed on gemcitabine-based chemotherapy [26]. Patients were randomized to one of the three arms: (1) 5-FU + leucovorin, (2) nanoliposomal irinotecan, or (3) combination of 5-FU + leucovorin + nanoliposomal irinotecan. The primary endpoint, median OS, in the 5-FU + leucovorin + nanoliposomal irinotecan group was 6.1 months compared to 4.2 months in the 5-FU + leucovorin group (HR: 0.67, p = 0.012). These results led to the approval of 5-FU + leucovorin + nanoliposomal irinotecan as a second-line regimen for gemcitabine-refractory patients in 2015. For patients who were treated with first-line 5-FU-based therapy, including FOLFIRINOX, gemcitabine + nab-paclitaxel represents a reasonable option for eligible patients, despite limited prospective data [27, 28]. Many patients who progress on first-line therapy are not candidates for further systemic therapy and should be managed with best supportive and/or hospice care.

Incorporation of targeted therapies into the armamentarium of pancreatic cancer treatment has largely been ineffective. A mutation in the KRAS gene is an almost universal finding in pancreatic cancers, and its subsequent activation is a well-described driver of pancreatic tumor development [29,30,31,32]. Despite decades of effort by researchers from the bench to clinic, the discovery of an effective therapy to target this crucial oncogene has been elusive thus far [33, 34]. Downstream to KRAS are potential therapeutic targets including mTOR, RAF, MEK, though inhibition of these proteins in pancreatic cancer patients has not demonstrated clinical efficacy in multiple clinical trials [35,36,37,38,39]. The addition of monoclonal antibodies targeting VEGF and EGFR to gemcitabine chemotherapy has similarly shown no benefit in large phase 3 clinical trials [40, 41]. One notable exception is erlotinib, an oral tyrosine kinase inhibitor with activity against EGFR, which demonstrated a statistically significant improvement in OS when added to gemcitabine compared to gemcitabine alone in a randomized phase 3 study [42]. However, erlotinib is seldom utilized in clinical practice as its addition to gemcitabine improved median OS by fewer than 2 weeks, and it is associated with non-trivial toxicities.

One targeted strategy with recent therapeutic success aims to exploit aberrancies in DNA damage repair pathways. While germline mutations in BRCA1 and BRCA2 are found in only up to 7%, broadening the umbrella of DNA damage repair mutations to include genes such as PALB2, ATM, RAD51, and CHEK1/2 may increase this population to almost a quarter of patients with pancreatic cancer [30, 43,44,45,46,47,48]. Patients with aberrant DNA damage repair are more susceptible to DNA-damaging therapies, including platinum chemotherapy as well as radiation. More recently, the development of inhibitors of poly(adenosine diphosphate-ribose) polymerase (PARP), a crucial component of the homologous recombination pathway for single-strand DNA breaks, has offered a novel therapeutic agent. Phase 2 studies of the PARP inhibitors olaparib and rucaparib as single agents have shown promising activity in pre-treated pancreatic cancer patients with germline or somatic BRCA1 and BRCA2 mutations [49, 50]. The phase 3 POLO trial randomized 154 metastatic pancreatic cancer patients with germline BRCA1 or BRCA2 mutations whose disease had not progressed on at least 16 weeks of first-line platinum-based chemotherapy to either maintenance placebo or olaparib [51]. The median PFS in the olaparib group was significantly longer (7.4 vs 3.8 months, HR: 0.53, p = 0.004). At the interim analysis, there was no difference in overall survival between the groups. In December 2019, olaparib was granted FDA approval for this indication, marking the first biomarker-based targeted therapy approved for pancreatic cancer [52]. Another rare but clinically significant molecular aberration is a fusion in the NTRK gene, which occurs in up to 1% of patients with pancreatic cancer and has been demonstrated in case reports to be a clinically significant target with the inhibitor, entrectinib [53]. In 2019, the FDA approved entrectinib as a tumor-agnostic treatment for patients with advanced solid tumors harboring NTRK fusions [54]. Tumor molecular profiling that includes evaluation for NTRK gene fusions should be considered for patients with advanced pancreatic cancer.

Over the past several years, immunotherapy has completely changed the treatment landscape in a number of malignancies including melanoma, non-small cell lung cancer, and genitourinary cancer. Unfortunately, most patients with gastrointestinal cancers have not reaped these benefits. Pancreatic cancer, in particular, has repeatedly failed to demonstrate response to single agent immune checkpoint inhibition outside of the approximately 1% of patients whose tumors harbor mutations in mismatch repair proteins or have high microsatellite instability (MSI-H) [55,56,57,58,59]. It has been hypothesized that immunotherapy in pancreatic cancer fails based on the presence of an immunologically “cold,” immunosuppressive, tumor microenvironment with abundant tumor-associated macrophages, regulatory T cells, cancer-associated fibroblasts, and myeloid-derived suppressor cells [60,61,62]. Clinical studies aiming to introduce cytotoxic lymphocytes or target these immunosuppressive components of the pancreatic tumor microenvironment in order to enhance the activity of both chemotherapy and immune checkpoint inhibitors are ongoing (e.g. NCT03336216, NCT02588443).

The treatment of patients with advanced pancreatic care invariably requires clinicians to incorporate supportive care. Common symptoms include pain, nausea, diarrhea, depression, anxiety, anorexia, and weight loss. Early recruitment of supportive or palliative care specialists to the treatment team can reduce the complexity of management by oncologists and has been demonstrated to decrease the aggressiveness of care near the time of death [63]. Pain is a near universal symptom, and its etiology is often multifactorial owing to local invasion into the celiac plexus or effects of distant metastases. Opioid-based therapy is the recommended treatment of cancer-related pain, but interventions such as celiac plexus neurolysis are also commonly utilized [64]. Biliary obstruction is another common effect of pancreatic cancer that can lead to infection and often precludes adequate delivery of chemotherapy. When required, biliary stenting with metal rather than plastic stents tends to improve stent patency and lowers infection risk [65]. Antidepressants, appetite stimulants, antiemetics, and pancreatic enzyme replacements are all part of the armamentarium of clinicians and should be considered for appropriate patients.

Approach to LAPC at MD Anderson

While an individualized approach is necessary for patients with locally advanced, unresectable disease, the general approach at MD Anderson Cancer Center is to begin with systemic therapy. FOLFIRINOX is typically given to patients with good performance status and no contraindications to treatment with oxaliplatin, such as pre-existing peripheral neuropathy. For those patients with frailty or comorbidities, gemcitabine and nab-paclitaxel given once every 2 weeks (rather than weekly × 3, every 28 days) are more commonly utilized. Restaging evaluations consisting of routine laboratory studies and serum tumor markers in addition to contrast-enhanced computed tomography of the chest, abdomen, and pelvis are performed every 2 months. If a patient develops metastatic disease or progression of the primary tumor with first-line chemotherapy, second line chemotherapy is usually recommended with efforts to enroll such patients on active clinical trials. For the subset of patients who have not progressed after 4–6 months of systemic therapy, referral to a radiation oncologist to consider consolidating radiation therapy is advised. Whenever possible, delivery of consolidating radiation is conducted in the context of a clinical trial. Importantly, the smaller subset of patients who have a clinical and radiographic response to chemotherapy with or without radiation at the primary tumor site without interval metastatic disease should be referred to an experienced surgical oncologist to consider surgical resection with curative intent. This is particularly true of those patients who have normal or near-normal serum tumor markers after systemic therapy ± radiation.

MD Anderson Approach to Metastatic Pancreatic Cancer

Combination chemotherapy is the most common recommendation for patients who present with metastatic disease. The choice of FOLFIRINOX or gemcitabine and nab-paclitaxel is based on the individual patient’s personal wishes, performance status, comorbidities, and frailty. FOLFIRINOX is usually preferred as first-line therapy for patients with ECOG performance status 0–1. Gemcitabine and nab-paclitaxel are more commonly offered to those patients with ECOG performance status 1–2. In some cases, only gemcitabine monotherapy should be considered, particularly for patients with frailty or those likely to experience increased toxicity with combination therapy (Table 9.2). When feasible, enrollment in a front-line clinical trial is considered, especially for patients with well-preserved performance status.

Table 9.2 Association between performance status and survival in pancreatic cancer
Full size table

Whenever possible, germline genetic testing and molecular profiling of biopsy material are recommended to identify patients who may benefit from subsequent targeted therapy, most commonly olaparib for patients with germline BRCA mutations delivered as maintenance therapy. This is usually appropriate after 4–6 months of cytotoxic therapy with a platinum agent such as oxaliplatin. Clinical trial enrollment is considered for patients with progressive disease after front-line therapy who maintain good performance status. Importantly, attention to the results of germline testing and/or molecular profiling of biopsy material for enrollment in biomarker-specific clinical trials is strongly encouraged.

Conclusion

Patients with locally advanced or metastatic disease comprise the overwhelming majority of those with pancreatic cancer. Cytotoxic chemotherapy continues to be the only treatment that offers a clear survival benefit for this group of patients. The development of combination chemotherapy regimens over the past decade, particularly gemcitabine + nab-paclitaxel and FOLFIRINOX, has extended the life expectancy of most patients with advanced disease. While a small percentage of initially locally advanced pancreatic cancers may ultimately become surgically resectable, the goal of therapy in the vast majority of patients is palliative in nature. Improvements in biomarker selection and novel targeted agents have already begun to expand therapeutic options for a minority of patients such as those with germline BRCA1/2 mutations, and there is optimism for similar future successes in other pancreatic cancer patients. While immune checkpoint inhibitors largely have no role in pancreatic cancer, manipulation of the immunosuppressive tumor microenvironment appears necessary to unlock the therapeutic potential of immunotherapy in this disease. Finally, supportive care is a crucial, if underappreciated, component of care for patients with advanced pancreatic cancer.