Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

During the past half century, therapeutic research has identified several chemotherapeutic agents that are effective in the treatment of osteosarcoma. These agents were incorporated into a number of therapeutic regimens. With their application in innovative multimodal strategies, cure rates were escalated from <20% prior to the 1907s to current levels of 65–75%. Accompanying this escalation has been the ability to offer limb salvage to approximately 80% of patients.

The principal agents currently in use comprise high-dose methotrexate with leucovorin rescue, doxorubicin (Adriamycin), cisplatin, cyclophosphamide, and ifosfamide. Earlier investigations with nitrogen mustard, mitomycin C, and vincristine had yielded minimal response, and these agents were abandoned.1

Conpadri/Compadri Series

In the early 1960s, Sutow and coworkers2 demonstrated anti-tumor activity in osteosarcoma with l-phenylalanine mustard. Temporary regression in 10–43% of patients was achieved.2 This result prompted an investigation of l-phenylalanine mustard as adjuvant chemotherapy for patients with nonmetastatic osteosarcoma after ablation of the primary tumor, and a disease-free survival rate of 14% was achieved.3 In 1969, the combination of vincristine, dactinomycin (actinomycin D), and cyclophosphamide (VAC) was demonstrated to be effective in treating rhabdomyosarcoma.4 The success achieved in rhabdomyosarcoma prompted Sutow et al5 to investigate the efficacy of the regimen as adjuvant treatment for osteosarcoma after ablation of the primary tumor. During this same period, osteosarcoma was shown to be responsive to cyclophosphamide (as discussed below).6 To potentiate the efficacy of the VAC regimen, Sutow administered cyclophosphamide in an intensive intermittent pulse schedule based on studies reported by Finklestein et al.7 This regimen designated “pulse VAC,” was administered to 12 patients and resulted in a 33% disease-free survival rate in all of them.8

With the demonstration that doxorubicin was effective in the treatment of osteosarcoma, 9 Sutow elected to substitute doxorubicin for dactinomycin in the pulse VAC regimen and to augment its efficacy with the addition of l-phenylalanine mustard. This regimen [pulsed cyclophosphamide, vincristine (Oncovin), l-phenylalanine mustard, and doxorubicin (Adriamycin)] was designated “Conpadri” or Conpadri I.10 It yielded a 55% disease-free survival rate. Sequential changes in the composition and acronym of the Conpadri regimen followed. Methotrexate was incorporated and it was designated “Compadri,” commencing with Compadri II. Each successive number indicated an evolution in the regimen.11 Sutow also observed that pulmonary metastases were appearing later in patients treated with the Compadri regimen.12 This change in the pattern of development of pulmonary metastases inaugurated new concepts in the treatment of patients with metastatic osteosarcoma. The conceptual development and the evolution of programs designed for this purpose were outlined in two publications.13,14 Prior to 1970, the survival rate for patients with metastases was considerably less than 2%. After 1970, according to Sutow, it improved to approximately 40%.13 The best postmetastatic survival rates occurred in patients whose metastatic lesions developed at least 13 months after initial treatment; the worst rates occurred when metastases were present at diagnosis.11 Sutow’s observation was confirmed by Jaffe et al,15 who noted an alteration in the pattern of relapse in several patients treated with adjuvant chemotherapy: metastases in these patients appeared later than in untreated or inadequately treated patients. They also tended to be single or isolated.15 This development permitted successful multidisciplinary intervention in an increasing number of patients.

The Compadri II and III regimens yielded disappointing results. It was surmised that their lack of efficacy was due to reduced doses of doxorubicin, and the approach was adjusted in Compadri IV and Compadri V: high-dose methotrexate and doxorubicin were intensified, and aggressive “front loading” was adopted. Unfortunately, Wataru Sutow’s untimely death precluded evaluation of the last two Compadri studies. However, before he died, an updated review of the Compadri I, II and III regimens was published: 81 of 200 patients (41%) were alive without evidence of disease, 18 months and longer after diagnosis.16

The Compadri regimens represented the first rational attempt to promote the use of combination chemotherapy as adjuvant therapy in osteosarcoma. They comprised different agents with different modes of action and minimal overlapping toxicity. Compadri was later superseded by other chemotherapeutic regimens.

High-Dose Methotrexate with Leucovorin Rescue

The use of methotrexate against osteosarcoma was initiated in the 1970s (see Chap.11). Methotrexate binds stoichiometrically and irreversibly to dihydrofolate reductase, thereby inhibiting the formation of tetrahydrofolate from dihydrofolate. This inhibition interferes with the de novo biosynthesis of purine and pyrimidine. Ultimately, thymidylate biosynthesis is inhibited; this is the key event leading to cell death. The antidote to methotrexate is leucovorin (5-formyl tetrahydrofolate). Within the cell, leucovorin is converted to 5,10 methylene tetrahydrofolate and 5-methyl tetrahydrofolate, thereby replenishing the product surceased by methotrexate.

The methotrexate-leucovorin “rescue” regimen (MTX-l) took wing following publications by Jaffe et al.17–20 Methotrexate is usually administered intravenously in doses of 10–12.5 G/m2 over 4–6 h, with leucovorin “rescue” commencing 24 h after the initiation of the methotrexate infusion. When deployed as single-agent therapy for “intensification” or “consolidation,” MTX-l should optimally comprise 4–8 doses administered at 10–14-day intervals. When combined with other agents, the interval between MTX-l and the other agents (generally doxorubicin, which may be administered 8–10 days after MTX-l) is usually extended to 21–28 days before initiating the next MTX-l dose.

MTX-l administered postoperatively as the sole agent to patients with osteosarcoma after ablation of the primary tumor, yielded a 40% disease-free survival.18 When MTX-l was combined with other agents as pre- and post-operative therapy for osteosarcoma, a disease-free survival rate of 65–75% was achieved.17–27

The Children’s Cancer Study Group considered high-dose MTX-l and intermediate-dose MTX-l in combination with doxorubicin as adjuvant therapy for nonmetastatic osteosarcoma; no benefit was observed for patients who received MTX-l. The overall outcome in these patients was not superior to that in patients who received doxorubicin alone.28 Two Osteosarcoma studies, one in Europe and the other in the United Kingdom, found an inferior disease-free survival rate in patients who received a three-drug regimen: doxorubicin, cisplatin and MTX-l as compared with the rate in patients who received the regimen without MTX-l.29,30 Several factors could possibly account for the inferior results achieved with high- or intermediate-dose MTX-l in these studies. The factors include inadequate tumoricidal concentrations due to substandard doses (vide the Children’s Cancer Study Group investigation), and dilution of methotrexate because of the excessive hydration that was designed to eliminate the drug. Other factors, including age and pharmacokinetics, also influence serum methotrexate levels and tumor response.21,31–33

It has been suggested that methotrexate levels of 700–1,000μmol/L at 4–6 h (generally 1,000μmol/L) after initiation of the infusion are required for optimum results.31,32 However, for unexplained reasons, inter- and intra-patient variability in methotrexate concentration is often encountered, despite administration of a standard dose, and optimum levels are not constantly obtained. In the author’s experience, a methotrexate concentration of 1,500μmol/L or higher at 4–6 h is desirable. This concentration is more likely to be obtained by limiting pre- and intra-therapeutic intravenous (alkaline) hydration to 3 L/m2/24 h.

Efforts to increase the therapeutic effect of methotrexate by enhancing the local concentration at the tumor site using the standard dose with intra-arterial, rather than intravenous, administration were unsuccessful.34 Apparently, response requires a critical dose, and any dose escalation above this level will not enhance therapeutic efficacy. Similarly, in pharmacokinetic and clinical studies of a 24-h infusion of high-dose methotrexate with leucovorin administered after completion of the infusion, the efficacy of the drug did not improve over that of a shorter 4–6-h infusion.35 These observations indicate that the optimum therapeutic tumoricidal concentration is achieved with the intravenous dosages described above.

With the optimum dosage administered over 4–6 h and the appropriate hydration, the following peak methotrexate levels may be anticipated at specific time intervals after initiation of the infusion (Fig. 1):

  • 24 h 30–300μmol/L

  • 48 h 3–30μmol/L

  • 72 h <0.3μmol/L

Fig. 1
figure 1_12

Methotrexate decay curve following a 6-h infusion at 7.5 G/m2 dissolved in 600 cc of 5% dextrose water administered over 6 h. Reproduced with permission from Advances in Chemotherapy. Jaffe N. Antifolate rescue use of high-dose methotrexate and citrovorum factor. In: Rossowsky A, ed. Advances in Chemotherapy. New York and Basel: Marcel Decker Inc.; 1979:111-141

Full size image

Values in excess of these concentrations, particularly at the 48-h level, portend potential toxicity.35–38

Leucovorin is administered according to specially designed algorithms that are available in most institutions. The algorithm generally advocates an intravenous dose of 10 mg after completion of the methotrexate infusion and a similar dose every 6 h until the methotrexate level is ≤0.1μmol/L. This usually occurs at 72 h and requires 12 doses. However, it may be necessary to prolong the leucovorin administration if the ≤0.1μmol/L methotrexate level is not attained at 72 h. In some institutions, a methotrexate level of ≤0.3μmol/L may be an acceptable endpoint.

Prerequisites for MTX-l therapy include normal renal and hepatic function, a normal hemogram, and absence of infection. The prerequisites are usually determined by obtaining a corrected creatinine clearance rate, a serum electrolyte study, urinalysis, liver function studies, and a complete blood count prior to each course of therapy. Collections of fluid (pleural, pericardial and peritoneal effusions) may cause a delay in methotrexate excretion by sequestering methotrexate into the fluid collection and are contraindicated in MTX-l treatment.

Toxic reactions are infrequent. They are generally induced by incomplete (delayed) renal clearance and are usually associated with methotrexate precipitation in the renal tubules. This reaction manifests with gastrointestinal mucosal ulceration, myelosuppression, and hepatorenal failure. Measures for aborting or treating toxic reactions may comprise any or all (usually the latter) of the following:

  1. 1.

    Increasing fluid intake to 4 L/m2/24 h.

  2. 2.

    Increasing leucovorin dose to 50–100 mg (or higher) every 6 h, as stipulated by the institution’s algorithm.

  3. 3.

    Administering carboxypeptidase G-2 if the serum 24- or 48-h methotrexate level is extremely high and/or anuria or oliguria appears to be developing.

  4. 4.

    Considering high-flux renal dialysis at any time in the above circumstances.

In addition to its efficacy as a pre- and post-operative agent, methotrexate potentiates the tumoricidal effects of radiation therapy.39–42 The effects are limited to the portals of radiation and include dermatologic reactions. Radiation effects are more likely to occur if the methotrexate administration coincides with radiation or is juxtaposed with the immediate postradiation period. With longer intervals between radiation and methotrexate, response and skin reactions are less observed. The combination of radiation therapy and methotrexate may be used for treatment of resistant pulmonary metastases and inoperable primary tumors. This combination is also extremely useful in alleviating cord compression and relieving bone pain.

Doxorubicin

Doxorubicin came into use for treatment of osteosarcoma in the early 1970s. The agent intercalates into DNA and induces topoisomearase II-mediated single- and double–strand breaks in the DNA. Initial studies in which doxorubicin was administered intravenously, alone or in combination with dacarbazine [dimethyldiethyl triazeno imidazole carboxamide, (DTIC)] produced responses in 35–40% of patients with pulmonary metastases.43–45 Responses included complete disappearance of lung lesions and a 40% reduction in tumor volume. The onset of the responses occurred within 1–2 months with doses of 30–35 mg/m2 administered daily for 3 days, at 3–4-week intervals. When administered as the sole agent after ablation of the primary tumor, doxorubicin also improved survival rates in patients with osteosarcoma.8,43–47

Doxorubicin can also potentiate the therapeutic effects of radiation therapy.48 In one study, doxorubicin was administered intra-arterially over 24 h in combination with radiation (3.5 Gy) to treat the primary tumor. More than 75% tumor destruction was reported in 24 of 36 patients.49 However, the procedure was complicated by erythema and ulceration of the skin and subcutaneous tissue in several patients. Selective entry of the drug into a small vessel was implicated, and it was suggested that the complication could possibly be prevented by positioning the catheter in a large-caliber vessel proximal to the tumor. Ulceration precludes limb-salvage procedures, and consequently intra-arterial doxorubicin is generally not advocated as local treatment for potential limb-salvage candidates.50

Doxorubicin may induce cardiac failure. To prevent this complication, the cumulative dose is generally limited to 300 mg/m2 in children under 6 years of age and to 450–500 mg/m2 in adolescents. However, based on experiences with adult patients with breast cancer, the cumulative dose may possibly be extended to 600 mg/m2 (or more) with liposomal formulations of the drug (e.g., Doxil51). Dexrazoxane has also been administered in combination with doxorubicin to prevent cardiac failure.52 The potential salubrious effect of the agent in preventing cardiac complications in long-term survival remains to be determined.

Doxorubicin has been claimed to be the most effective agent for the treatment of osteosarcoma.53 It is incorporated in most combination chemotherapy regimens used for this disease. A cardiac assessment comprising an electrocardiogram and echocardiogram should optimally be obtained prior to the administration of each course. Cardiac assessments should also be obtained at regular intervals in long term survivors.

Cisplatin

Cisplatin (cis-diamminedichloroplatinum II) was first used for treatment of osteosarcoma in the 1970s. It exerts its cytotoxic action by platination of DNA. It may be administered intravenously or intra-arterially. In a series of studies in which cisplatin was administered intravenously, it produced responses of 30–50%.54–56 The responses were obtained in patients with unresectable or metastatic disease who received the agent alone or in combination with doxorubicin. In contrast, in studies in which cisplatin was administered intra-arterially as the sole agent for treatment of the primary tumor, response rates were 60–90%.57,58 The intra-arterial route achieves higher local cytotoxic concentrations which improves penetration across the cell membrane.57 This strategy was investigated principally at The University of Texas M.D. Anderson Cancer Center.58–62 The procedure, which involved general anesthesia or conscious sedation of the patient, required placement of an arterial catheter via the Seldinger technique through the brachial or femoral artery (Fig. 2) under fluoroscopic guidance. Concurrently, a systemic intravenous infusion was initiated to provide hydration at 3 L/m2/24 h and Manitol. The tip of the arterial catheter was positioned into a vessel that supplied the neoplasm. A pulsatile infusion pump was used to induce turbulence of the cisplatin with saline59; this turbulence prevented laminar flow and reduced the possibility of a “platinum burn” because of selective entry of high platinum concentrations into small vessels. Occasionally, tumor embolization may be performed to improve the direction and concentration of chemotherapy to the tumor, if excessive neovascularity is present.

Fig. 2
figure 2_12

Arterial catheter containing cisplatin (CDP) attached to pulsatile infusion pump (PUMP) inserted into left femoral artery and directed to the tumor in the contralateral limb via the bifurcation of the aorta. Systemic hydration and Mannitol are provided through a venous catheter (I.V.Inf) in the right antecubital fossa. The diagram also depicts sites of venous catheters inserted to determine cisplatin concentrations in the local tumor draining vein and systemic circulation. I.V. Inf = Intravenous in fusion; CDP = Cisplatin; PUMP = Pusatile Infusion Pump

Full size image

In the initial studies, the dosage was 150 mg/m2 administered with Mannitol over 2 h at 3-week intervals.57,58 Four preoperative courses were administered. In more recent studies, 120 mg/m2 cisplatin is administered intra-arterially over 4 h, and concurrently 95 mg/m2 doxorubicin is administered over 24 h. In a schedule similar to the sole treatment with intra-arterial cisplatin, four preoperative courses are administered at 4-week intervals.

Pharmacokinetic studies were conducted as part of the initial study of 150 mg/m2 cisplatin administered intra-arterially. Evaluation of the local venous effluent and concurrent systemic venous concentrations demonstrated consistently higher cisplatin concentrations in the local vein than in the peripheral vein57 (Fig. 3, left and right respectively). The highest single concentrations in the local vein were 10 and 9μg/mL at 60 and 90 min, respectively, as opposed to the highest concentrations in the peripheral vein, which were 1.7 and 3.9μg/mL at 30 and 120 min, respectively. From 90 min, the local venous concentrations plotted on a log scale were linear by curve fitting. The concentration in the systemic circulation was sufficiently tumoricidal to destroy pulmonary metastases. Figure 4 demonstrates complete disappearance of pulmonary metastases after two courses of 150 mg/m2 intra-arterial cisplatin administered for a primary tumor in the distal femur. These pharmacokinetic and clinical studies contradict the claim that the systemic concentration following intra-arterial administration is insufficient to destroy pulmonary metastases.63 Systemic concentrations in these studies were also sufficient to cause adverse side effects, including auditory and renal dysfunction.64,65

Fig. 3
figure 3_12

Left: Local venous cisplatin concentrations (tumor draining vein). Right: Systemic venous cisplatin concentrations. CDP cisplatin; Pt total platinum (Mcg/ml)

Full size image
Fig. 4
figure 4_12

Computer scan of lungs demonstrating pulmonary metastases (left) and disappearance of pulmonary metastases (right) following two courses of intra-arterial cisplatin 150 mg/m2

Full size image

The efficacy of intra-arterial therapy may be demonstrated by angiographic study with the disappearance of tumor neovascularity and staining (Fig. 5). It is also capable of inducing a complete response in patients with pathological fractures (Fig. 6).

Fig. 5
figure 5_12

Arteriogram demonstrating recurrent tumor in the distal end of the femur manifesting with neovascularity and stain (left). Complete disappearance of tumor neovascularity was obtained with after four courses of intra-arterial cisplatin, 150 mg/m2/course (right). Pathological examination of resected tissues demonstrated absent tumor or complete tumor necrosis in sites where minimal residual tumor was suspected to be present

Full size image
Fig. 6
figure 6_12

Pathological fracture of the humerus at diagnosis (left) and complete healing after four courses of intra-arterial cisplatin, 150 mg/m2/course (right)

Full size image

Tissue determinations of cisplatin levels were obtained in the tumor and the surrounding tissues in some patients (Fig. 7). These revealed that concentrations of 17–40μg/g were associated with tumor destruction of 60–100%

Fig. 7
figure 7_12

Tissue cisplatin determinations in the proximal tibia following four course of intra-arterial cisplatin, 150 mg/m2/course

Full size image

Figure 8 demonstrates 100% tumor destruction after four courses of intra-arterial cisplatin. This result contrasted sharply with those for concentrations of 12μg/g or less, which were associated with tumor destruction of less than 60%. The difference in the mean cisplatin tumor concentrations between the groups with greater than 60% tumor destruction and those with less than 60% destruction was 16.7μg/g. Using the one-tailed t-test, this difference was significant at a level of <0.025

Fig. 8
figure 8_12

Photomicrograph of tumor at diagnosis (left) and specimen obtained after treatment with four courses of intra-arterial cisplatin, 150 mg/m2/course (right). Complete necrosis was induced. There is a complete absence of tumor cells in the specimen comprising residual bone trabeculae

Full size image

Cisplatin uptake also varied with tumor subtype in this study. Smaller concentrations were detected in patients with telangiectactic osteosarcoma and malignant fibrocystic histiocytoma, as opposed to chondroblastic osteosarcoma. In patients with malignant fibrocystic histiocytoma, 60% tumor destruction was noted with a cisplatin concentration of 2.4μg/g.57 Cisplatin tumor concentration and tumor destruction were also related to the number of infusions: the percentage of tumor destruction was greater with three or more infusions than with two infusions.58

Patients whose tumors initially respond and later relapse may experience a response again with reinstatement of cisplatin at the same dose (Fig. 5 shows an example). Intra-arterial therapy is extremely useful when an immediate response is desired, particularly in the treatment of pathological fractures72 (Fig. 6) or with the threat of tumor invasion and potential imminent compromise to the neurovascular bundle.

A review of the results of treatment with intra arterial cisplatin in several publications revealed an average sensitivity of 95% and a specificity of 87%.66–71

The tumoricidal effects achieved with the first or second course of intra-arterial cisplatin may also be achieved with conventional intravenous therapy administered over a more prolonged period (several courses). In contrast, intra-arterial cisplatin is capable of producing a rapid definitive attack on the primary tumor. The rapidity and immediacy of response with intravenous cisplatin are not as impressive as that achieved with the first or second course of intra-arterial cisplatin. In addition, the efficacy of intra-arterial treatment may also be assessed reasonably early on the arteriogram by observing the reduction of tumor neovascularity and staining after the first or second course.

Oxazaphosphorines

The oxazaphosphorines, cyclophosphamide and ifosfamide, are alkylating agents that require hepatic microsomes for activation. They possess moderate to high efficacy in the treatment of osteosarcoma.5,6,73–75 Cyclophosphamide was probably the first agent discovered to have activity in osteosarcoma5 (Fig. 9). In 1962, Pinkel6 stated that he knew of no reports concerning responses of “osteogenic sarcoma to other alkylating agents” at that time.

Fig. 9
figure 9_12

Chest radiograph demonstrating pulmonary metastases from osteosarcoma (left) and partial response to oral cyclophosphamide (right). The latter manifested with disappearance and reduction of tumor masses. This material is reproduced with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc. (Pinkel6: Figs. 3a,b)

Full size image

Cyclophosphamide may be administered in combination with etoposide (VP-16), which is thought to augment the efficacy of the alkylating agent through a synergistic interaction. An early study of this combination yielded a 58% response rate in various malignant diseases, including osteosarcoma.76 A follow-up of the study concluded that the combination of cyclophosphamide and etoposide was effective therapy for both primary and metastatic osteosarcoma.77 Eighty-eight percent of the patients experienced complete or partial responses. Because of its putative synergistic effect, etoposide is also frequently combined with ifosfamide.

Cyclophosphamide and ifosfamide produce a metabolite, acroline, that causes hemorrhagic cystitis. This complication can be prevented with the administration of Mesna. The latter absorbs the acroline, providing uroprotection. This strategy permits the administration of high doses of cyclophosphamide and ifosfamide. The intake of liberal amounts of fluid is another means of preventing hemorrhagic cystitis.

The activity of the alkylating agents, particularly ifosfamide, can be augmented by fractionating the dose. The efficacy can also be enhanced by dose escalation. Investigators have reported initial responses of 10–40% with doses of 6–9 g/m2 78–80; escalating the dose to 12 or 14 g/m2 yielded enhanced responses of 60%.81–84 These results were noted in patients who had had relapses or in whom conventional therapy had failed. This experience, however, was not observed by Harris et al,85 who noted a complete and partial response rate of only 30% with a dose of 12 g/m2. The alkylating agents also are not cross-resistant: in patients who have relapses after treatment with a specific alkylating agent, responses may again be achieved by substituting an alternative alkylating agent (e.g., ifosfamide for cyclophosphamide or vice versa).

Goorin et al86 treated patients with newly diagnosed osteosarcoma in a “therapeutic window” at 3–4-week intervals and achieved a 59% response rate with a combination of ifosfamide (3.5 g/m2/day for 5 days, for a total of 17.5 g/m2) and etoposide (100 mg/m2/day for 5 days). This experience was duplicated by investigators at M. D. Anderson Cancer Center. However, in contrast to the patients treated by Goorin et al, patients at M.D. Anderson had been heavily pretreated with high-dose methotrexate, doxorubicin, cisplatin, and ifosfamide (9 g/m2). Etoposide was usually omitted. The response of one such patient with pulmonary metastases who was treated with ifosfamide only is illustrated in Fig. 10. The total dose of ifosfamide, 17.5 g/m2, was associated with moderate myelosuppression and mild renal dysfunction. In addition, two other patients developed moderate renal failure following the fifth course of 17.5 g/m2 ifosfamide. Thus, the use of high-dose ifosfamide (17.5 g/m2) should probably be limited to four courses. If there is evidence of renal dysfunction, cyclophosphamide, which is unlikely to affect the kidneys, may be substituted for ifosfamide at an equivalent dose. This is determined by dividing the ifosfamide dose (17.5 g/m2) by 3.5. The resulting dose of cyclophosphamide (3–4 g/m2) may be administered over two consecutive days (i.e., 1.5–2 g/m2/day) at 3–4-week intervals with Mesna.

Fig. 10
figure 10_12

Chest radiograph of patient who responded to four courses of Ifosafamide 17.5 G/m2/course administered over 5 days (3.5 G/m2/dx5). The patient had previously been treated with, and responded to, Ifosfamide 14 G/m2 and later relapsed

Full size image

Less Frequently Employed Chemotherapeutic Agents

Carboplatin, has been used in several combination regimens.87–90 In a study of patients with metastatic lung lesions, a regimen containing 560 mg/m2/day carboplatin was not as effective as 100 mg/m2/day cisplatin.88 Petrilli et al90 investigated intra-arterial carboplatin (Study III) in a series of patients. They were also treated with epirubicin, ifosfamide and MTX-l. In contrast, in Study IV, intravenous carboplatin in conjunction with cisplatin, doxorubicin and Ifosfamide was employed. The overall survival rate for Study III and Study IV in patients who had nonmetastatic osteosarcoma at the time of their original diagnosis, was 60.5%, and the event-free survival rate, 45.5% at 5 years. Since other agents in addition to carboplatin were employed, the contribution of the latter to the final result cannot be assessed.

Novel antifolate agents, including trimetrexate, have been investigated in patients who have had relapses. Although these agents produced isolated responses, they have not been evaluated in formal clinical trials. In addition, gemcitabine, which has been reported to produce responses in osteosarcoma, awaits further investigation.91,92

High-dose radioactive samarium153 [153Sm-EDTMP (Quadramet)] has been used to treat bone metastases and has afforded patients appreciable relief with regard to symptoms.93 However, this treatment is usually complicated by severe myelosuppression and may require peripheral blood or stem-cell support.

A 2×2 factorial design study using two chemotherapy regimens, one standard and one experimental, was employed in conjunction with liposomal muramyl tripeptide phosphatidyl ethanolamine (l-MTPPE).94 The latter induces infiltration of inflammatory macrophages into lung metastases. The study was designed to evaluate the activity of l-MTPPE in osteosarcoma. As such, each standard and experimental arm included or did not include l-MTPPE. All patients received identical cumulative doses of cisplatin, doxorubicin, and MTX-l. The results published in the initial report found no statistically significant advantage for l-MTPPE in disease-free survival, the primary endpoint, although the trend favored the combination of ifosfamide and l-MTPPE. Overall survival, which was not prespecified in the protocol, showed a 76% six year survival rate for patients who received l-MTPPE with ifosfamide, compared with a rate of 66% for patients who did not receive the combination (p=0.183).

The U.S.Food and Drug Administration95 concluded that the l-MTPPE single study did not provide substantial evidence of effectiveness: the results for the primary endpoint did not reach statistical significance, and the overall survival analysis was not part of the study plan. The report stated: “Follow-up data have not been rigorously collected and are incomplete with insufficient follow-up for a significant proportion of patients.”95 l-MTPPE was not sanctioned for general clinical distribution.

In a follow-up report of the above study the authors confirmed a trend for improved event free survival (p=0.08) and improved overall survival (p=0.03) for the MTPPE arm.96 These results were discussed in several letters to the Editor of the Journal of Clinical Oncology.97–99 It was suggested that additional investigation be performed to substantiate the utility of MTPPE and define its exact role in the treatment of osteosarcoma. The agent is available through a Compassionate Investigational New Drug (CIND) application and in certain investigational trials.

Inhalation therapy with granulocyte-macrophage colony-stimulating factor (GM-CSF) is currently under investigation by the Pediatric Oncology Group for treating pulmonary metastases.

Chemotherapy Regimens

Spawned by the efficacy of chemotherapy, combination regimens were devised for treating patients with osteosarcoma. Currently most regimens, following the principles and example provided by Sutow, comprise agents with different mechanisms of action and minimal or non overlapping toxicity. An important additional principle in the construction of each regimen is the attempt to deliver agents at maximum dose intensity. Chemotherapy is integrated into a multidisciplinary approach to assist surgical extirpation of the primary tumor and resection of pulmonary metastases. Chemotherapy may also potentiate the action of radiation therapy in resistant and recurrent tumors.

The results obtained with most regimens have been similar. Except for an occasional publication, 71 there does not appear to be a regimen which can claim superiority over those published in most communications. With current strategies, utilizing pre- and postoperative chemotherapy, cure rates of 60–75% in newly diagnosed nonmetastatic patients have been reported.17–20,22,27,30,71 Limb salvage and, occasionally, rotationplasty have been reported in as many as 80% of the patients in these studies and in several communications devoted specifically to this topic.100–102 In this context, an attempt was made to perform limb salvage exclusively with chemotherapy and abrogating surgery.103 It was successful in three patients only: Figs. 11 and 12 demonstrate the result in one of these patients.

Fig. 11
figure 11_12

Initial radiograph of patient with osteosarcoma of the proximal tibia (left). There is a mixed osteoblasic and osteolytic lesion. Follow-up after seven courses of high dose methorexate (7.5 G/m2/course) demonstrates healing by calcification of the medulary lytic lesions and solid periosteal bone formation (right). This was accompanied clinically by a complete absence of symptomatology. This material is reproduced with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc. (Jaffe N et al: Cancer 1985;56:461-466)

Full size image
Fig. 12
figure 12_12

Photograph of patient with lesion depicted in Fig. 11 10 years after completion of treatment exclusively with chemotherapy. The patient has remained alive and well without recurrent tumor 28 years after diagnosis and therapy

Full size image

Inadequacy of Chemotherapy

While chemotherapy has produced remarkable successes in osteosarcoma, it has also been marred by failures. The survival rate for patients with pulmonary metastases following aggressive multimodal treatment is of the order of 25–30%.104–113 It has not improved over the past quarter century. Bielack et al reported that overall and event-free survival rates, respectively, were 16% and 9% for second, 14% and 0% for third, 13% and 6% for fourth, and 18% and 0% for fifth recurrences: “The exact role of retreatment with chemotherapy, particularly in the adjuvant situation, remains to be defined.”114

Patients with partially treated tumors in whom survival has been prolonged, have also developed extrapulmonary metastases in uncommon sites, notably the kidneys, brain, heart, mediastinum, and epidural space.115–117 Such metastases may cause severe complications, and considerable pain and discomfort, and require extensive palliative maneuvers. Attempts to prevent and eradicate such extra pulmonary metastases constitute a major challenge. New therapy for patients with multifocal (sclerosing) osteosarcoma must also be developed.

Conspectus: Impact on the Conduct of Future Investigation and Treatment

Despite major discoveries in the chemotherapeutic pantheon for osteosarcoma over the past 35 years, the survival rate has plateaued. This is due principally to the fact that the arsenal of available effective chemotherapeutic agents over this period has not changed substantially . New agents and alternative strategies for the conquest of this malignancy are urgently required. Employing new paradigm shifts should receive serious consideration, while acquisition of new agents might include biotherapy, gene therapy, anti-angiogenic agents, targeted therapy, and attempts to harness the power of the immune system.

The reviews and reports cited above make no mention of identification markers to detect silent pulmonary micrometastases. Biomarkers or mechanisms with reliable specificity and sensitivity to identify such metastases would constitute a significant saltation in planning new strategies of treatment. Molecular profiles with rigorous characterization, gene expression patterns and phenotypes of osteosarcoma could afford an opportunity for planning risk-adjusted or personalized chemotherapy with reduced toxicity. Identification of patients free of micrometastases would permit treatment with chemotherapy limited to downstaging the primary tumor for surgical extirpation.

Despite the absence of major advances in chemotherapy and of any significant improvement in survival during the past 35 years, the ability to offer limb salvage to approximately 80% of newly diagnosed patients is noteworthy. Further, although the attempt to treat patients exclusively with chemotherapy, avoiding surgical ablation of the primary tumor was not entirely unsuccessful102, three of the 31 patients were cured with this approach. With new and more effective agents, the ability to cure osteosarcoma without surgical ablation of the primary tumor may yet be realized. The ability to rescue most patients with recurrent pulmonary metastases may also be attained. Considering the phenomenal strides made in the treatment of this disease over the past 35 years, the current lack of new agents notwithstanding, these possibilities may become a reality in the new century.