117: Systemic Chemotherapy for Mesothelioma

Mesothelioma is a rare malignancy arising from the mesothelial surfaces of the pleura, peritoneum, pericardium, or tunica vaginalis. There are roughly 2000 to 3000 new cases of mesothelioma annually in the United States.¹ Although the incidence of this disease peaked in the United States about a decade ago, its incidence in many other parts of the world continues to rise, particularly in regions without sufficient asbestos regulation.²

Even with aggressive multimodality therapy for seemingly localized disease, disease recurrence or progression occurs in the vast majority of cases. Effective systemic treatment options for mesothelioma are sorely needed. To date, there is only one agent—pemetrexed—that has received specific approval from the U.S. Food and Drug Administration (FDA) for the treatment of mesothelioma. The development of effective systemic therapies for patients with mesothelioma has faced several significant challenges.

The relatively low incidence of this cancer makes accrual to clinical trials—particularly large, randomized trials—challenging. In addition, the relative rarity of this cancer can dampen enthusiasm and limit funding from pharmaceutical companies and funding agencies. As a result, much of the clinical data in this disease consists of smaller, single-arm, investigator-initiated studies. Furthermore, comparisons of outcomes between these smaller studies are complicated by significant variability in staging systems among the trials, as well as by differences in the histologic composition of each trial, which consists of varying percentages of the three main histologic types of mesothelioma: epithelioid, sarcomatoid, and biphasic disease. Epithelioid histology consistently has been associated with a better prognosis. Therefore, differences in the histologic composition of each trial can confound useful comparisons between studies.

Another obstacle to determining treatment efficacy in mesothelioma is the assessment of tumor response: the applicability of standard Response Evaluation Criteria in Solid Tumors (RECIST) has been limited in mesothelioma, particularly with respect to serial assessments of the pleural rind. To address this problem, RECIST have been modified specifically for the assessment of mesothelioma (Table 117-1).³

Faced with these challenges and dissuaded by low response rates in early clinical trials of cytotoxic chemotherapy in mesothelioma, drug development in this disease had slowed. More recently, though, the clinical success of agents like pemetrexed and the preclinical activity of novel targeted agents have helped to spur a number of promising clinical trials and have offered a renewed sense of hope for patients suffering from this disease.

For patients with unresectable mesothelioma, combination with cisplatin plus pemetrexed is the standard first-line therapy and remains the only FDA-labeled treatment option in this disease. Though both cisplatin and antifolate agents were long known to have activity in this disease, the efficacy of this combination in mesothelioma was not demonstrated until 2003.

Cisplatin, an alkylating agent first described in 1845,⁴ is an integral component of combination chemotherapy regimens in many cancer types. A meta-analysis of clinical trials reported between 1965 and 2001 highlights the activity of cisplatin in mesothelioma.⁵ When used as a single agent, though, activity is modest. Experiences in other cancers had shown that cisplatin is best used in combination with other agents.

Antifolates were also long known to have some efficacy against mesothelioma. This class of agents was among the first chemotherapy agents to show activity in mesothelioma. Administration of high-dose methotrexate (3 g/m² every 10–21 days) with leucovorin rescue was associated with response in 22 of 60 patients evaluable for response (37%), with a median survival of 11 months.

The development of the multitargeted antifolate pemetrexed in the late 1990s and early 2000s would ultimately establish a new standard for systemic therapy in mesothelioma. This agent inhibits several key enzymes in purine and pyrimidine synthesis: thymidylate synthase, dihydrofolate reductase, and glycinamide ribonucleotide formyltransferase. In a large phase III trial, patients with previously untreated mesothelioma were randomized to receive cisplatin plus pemetrexed or cisplatin alone.⁶ The addition of pemetrexed was associated with improvements in response (41.3% vs. 16.7%, p < 0.0001), median time to progression (TTP) (5.7 vs. 3.9 months, p = 0.001), and median overall survival (OS) (12.1 vs. 9.3 months, p = 0.02). The combination of cisplatin plus pemetrexed was generally well tolerated. Although there were three patient deaths in the combination therapy arm that were considered to be treatment-related, all occurred in patients who had not received supplementation with folic acid and vitamin B₁₂ . There were no deaths in the vitamin-supplemented group. The most common grade 3 or 4 toxicities in the fully supplemented group included neutropenia (23.2%), nausea (11.9%), vomiting (10.7%), and fatigue (10.1%).

From these results, the combination of cisplatin plus pemetrexed has become the standard first-line therapy in mesothelioma. The agents are delivered intravenously once every 21 days: cisplatin 75 mg/m² and pemetrexed 500 mg/m². Patients should receive supplementation with vitamin B₁₂ (1000 mcg intramuscularly starting 1 week before cycle 1 and then given every 9 weeks thereafter) and folic acid (400 mcg by mouth daily). In addition, patients also typically receive dexamethasone premedication (4 mg PO BID on the day before, day of, and day after chemotherapy) in order to prevent pemetrexed-associated rash. They also receive aggressive antiemetic therapy in light of the emetogenic potential of cisplatin. Restaging scans are recommended after every two to three cycles of treatment to monitor for progression. In the Vogelzang study, patients who tolerated therapy well and who achieved ongoing response or stable disease were continued on treatment. The median number of cycles of combination therapy received in that study was six (range 1–12), with only 7% of fully supplemented patients continuing on eight or more cycles of combination therapy.⁶ In practice, the optimum length of therapy has not been established, though extrapolation from other tumor types such as non–small-cell lung cancer has led many investigators to stop combination treatment after four to six cycles.

For those patients whose age, overall condition, and/or medical comorbidities preclude the use of combination therapy with cisplatin plus pemetrexed, treatment with pemetrexed plus carboplatin or with pemetrexed monotherapy could be considered. Although there has been no trial that directly compares pemetrexed-based combinations with cisplatin versus carboplatin, useful data are available from phase II trials as well as from the International Extended Access Program. In a phase II trial of carboplatin plus pemetrexed, 19 out of 102 patients achieved a response (18.6%, 95% CI 11.6%–27.5%), with 48 patients (47%) achieving stable disease.⁷ Median TTP was 6.5 months and median OS was 12.7 months, both comparable to the results seen with cisplatin plus pemetrexed in the Vogelzang study. In the Expanded Access Program, patients received carboplatin plus pemetrexed had a response rate of 21.7%, TTP of 6.9 months, and a 1-year survival of 64%.⁸

For those patients for whom combination chemotherapy with a platinum-based agent is deemed intolerable, pemetrexed monotherapy can be considered. A phase II study of pemetrexed monotherapy as well as data from the pemetrexed Expanded Access Program have provided insights into the activity of single-agent activity in mesothelioma. As a single agent, pemetrexed achieves response rates of about 10% to 12%, with median survivals of 10.7 to 14.1 months.^9,10

Pemetrexed Maintenance or Reinitiation

Just as the optimal duration of first-line therapy is not well established, the role of pemetrexed maintenance therapy is also an area of ongoing investigation. A small phase II study followed 27 patients with nonprogressing disease after six cycles of first-line therapy with a pemetrexed-containing combination. In this study, 13 patients received pemetrexed maintenance versus 14 who did not. Both TTP (8.5 vs. 3.4 months) and OS (17.9 vs. 6.0 months) favored those patients receiving pemetrexed maintenance. Furthermore, treatment was generally well tolerated, with fatigue (15%) as the only non-hematological grade 3 toxicity encountered. A randomized study of the role of maintenance therapy is currently ongoing.

For patients who had received a first-line regimen that did not contain pemetrexed, consideration should be given to pemetrexed as a second-line agent at the time of progression.^11,12 For patients whose disease recurs well after completing first-line therapy with a pemetrexed-containing combination and who had never progressed while on that therapy, physicians can consider reinitiating pemetrexed-based therapy. Though there is no specific trial exploring this indication, it is not unreasonable to return to pemetrexed, given the proven activity of the agent, the existing data for its use in the maintenance setting (see previous), and the general paucity of other commercially available systemic options in mesothelioma.

Other Cytotoxic Chemotherapeutics with Activity in Mesothelioma

Although there are no agents besides pemetrexed that have been specifically FDA-approved for the treatment of mesothelioma, there are existing, commercially available therapies with known activity in this tumor type. The two most commonly considered agents in patients who have progressed on pemetrexed-based therapy are vinorelbine and gemcitabine.

Vinorelbine

Vinorelbine, a semisynthetic vinca alkaloid, also has demonstrated activity against mesothelioma, alone or in combination. In a randomized trial, vinorelbine monotherapy had a response rate of 16% and demonstrated a trend toward improved median OS (9.5 months) compared with active symptom control alone (7.6 months) in previously untreated mesothelioma patients (HR 0.80, 95% CI 0.63%–1.02%; p = 0.08).¹³ Had the trial not suffered from poor accrual, it might have ultimately attained sufficient power to show a statistically significant benefit for vinorelbine over active symptom control alone. Other phase II trials of vinorelbine monotherapy in mesothelioma have shown similar response rates (16%–24%) and median OSs (9.6–10.6months).^14,15

Gemcitabine

Gemcitabine, an IV pyrimidine analog, has activity in mesothelioma, both as a single agent and as a part of combination therapy. An early trial of gemcitabine monotherapy reported a 31% response rate, with an additional 40% of patients experiencing significant symptomatic improvement in their disease; however, these results must be viewed with caution owing to the small trial size (n = 23) and a high percentage of patients with early-stage disease and epithelial histology.¹⁶ Indeed, two other trials of gemcitabine monotherapy in mesothelioma delivered more sobering results: a multicenter Cancer and Leukemia Group B (CALGB) trial (n = 17) had no clinical responses and a median survival of only 4.7 months,¹⁷ whereas a European trial (n = 27) showed a response rate of 7% and a median survival of 8 months.¹⁸

The sentiment for gemcitabine as an agent with clear activity in mesothelioma is perhaps better supported by outcomes exploring its use in combination with platinum. Response rates across six clinical trials that used combination therapy with gemcitabine plus cisplatin have varied from 12% to 48%, with median times to progression that have ranged from about 6 to 8 months.^19–24 In fact, a retrospective review of cases within British Columbia found no difference in median survival between patients treated with cisplatin plus pemetrexed and those treated with cisplatin plus gemcitabine.²⁵

Anthracyclines

Along with methotrexate, doxorubicin was among the earliest agents to demonstrate activity in mesothelioma. Clinical trials of single-agent treatment with doxorubicin or epirubicin have demonstrated response rates ranging from 0% to 15%.^26–28 Although anthracycline-based combination regimens also have shown reasonable activity in this disease, cardiotoxicity concerns and the development of other less toxic agents (e.g., pemetrexed) have blunted enthusiasm for further investigation of the anthracyclines in mesothelioma.

Investigators across disciplines continue to explore the role of multimodality approaches in patients with potentially resectable mesothelioma. However, the specific target population, the exact nature and order of the interventions, and the extent of benefit of these aggressive approaches remain undefined. To this point, the only specific intervention that has ever been shown to prolong survival in mesothelioma in a randomized trial is chemotherapy with pemetrexed and cisplatin for patients with unresectable disease. This should not dampen our enthusiasm for exploring newer and better ways of treating localized disease in mesothelioma. Rather, it should challenge us to devise and conduct focused clinical trials to refine our interventions and define our target populations more closely.

Chemotherapy is the only modality ever shown to prolong survival in any randomized trial in mesothelioma. As a consequence, it is considered an important part of multimodality approaches. Based on the experiences in unresectable disease, pemetrexed plus cisplatin is the most common regimen used for localized therapy. Although there are no clinical trials to address whether this is best given before or after surgery, many prospective clinical trials that incorporate chemotherapy have used it neoadjuvantly. In non–small-cell lung cancer, patients were more likely to complete chemotherapy and chemotherapy was more easily tolerated when given neoadjuvantly rather than postoperatively.²⁹ When one considers the relatively greater potential morbidity of extrapleural pneumonectomy (EPP) for mesothelioma, concerns about tolerability become even more important.

In addition, analysis from a prospective multicenter phase II trial of multimodality therapy suggests that initial response to neoadjuvant chemotherapy may be the most important predictor of survival in patients undergoing trimodality therapy. Seventy-seven patients with T1-3 N0-2 mesothelioma were prospectively enrolled in a trial of four planned cycles of cisplatin plus pemetrexed.³⁰ Patients with nonprogressing disease then underwent EPP, followed by hemithoracic radiation. Eighty-three percent of patients completed all four cycles of chemotherapy, with an overall response rate of 32.5% (95% CI, 22.2–44.1). Fifty-seven patients proceeded to the operating room, with 54 patients undergoing EPP. Forty-four patients received hemithoracic postoperative radiation, with 40 completing the full planned course. In the context of this planned aggressive approach, median survival in the overall study population was only 16.8 months (95% CI, 13.6 – 23.2 months; censorship, 33.8%). However, it appeared that there was clearly a subset of patients who were more likely to benefit: the 2-year survival on the trial was 37%, and 20% of patients were estimated to live at least 3 years. Univariate analysis of multiple patient and disease factors showed that the only factor associated with prolonged survival was radiologic response to initial chemotherapy. In patients who achieved a complete or partial response to chemotherapy, the median survival was 26 months, whereas those patients who had only stable disease or progressive disease on chemotherapy had a median survival of 13.9 months (p = 0.05). Furthermore, patients who were able to complete all three modalities (n = 40) had a median survival of 29.1 months and a 2-year survival rate of 61.2%.

With this in mind, if (1) neoadjuvant chemotherapy might be more tolerable than adjuvant treatment and might, therefore, increase the chance of completing all three modalities, and (2) upfront chemotherapy has a potentially prognostic value in determining who might best benefit from aggressive trimodality therapy, it is reasonable to consider chemotherapy upfront for patients with potentially resectable mesothelioma who are being considered for multimodality treatment.

Angiogenesis Inhibition

The potential role of antiangiogenic agents in mesothelioma has provoked significant interest. Preclinical work has detected vascular endothelial growth factor (VEGF) and its receptors VEGFR1 and VEGFR2 in a majority of mesothelioma tumor specimens.^31–33 Moreover, VEGF has been found to stimulate the proliferation of mesothelioma cells in vitro, and this growth can be inhibited by the use of purified rabbit polyclonal antibodies targeting VEGF.³³

Bevacizumab, a humanized monoclonal antibody directed against VEGF, has been shown to increase survival when used in combination therapy in patients with metastatic colon cancer³⁴ and non–small-cell lung cancer,³⁵ and it has been shown to prolong TTP in patients with metastatic breast cancer.³⁶ To study its use in mesothelioma, a phase II trial randomized previously untreated patients to receive cisplatin plus gemcitabine, with or without bevacizumab.²² The addition of bevacizumab did not result in significant improvements in either response (25% vs. 22%), median progression free survival (PFS) (6.9 vs. 6.0 months), or median OS (15.6 vs. 14.7 months) compared with chemotherapy alone. Subset analysis did show a correlation between higher baseline plasma VEGF levels and shorter PFS and OS, suggesting a basis for further inquiry of bevacizumab in a more selected population.

Cediranib, an oral small molecule inhibitor of VEGF, has also been studied in mesothelioma. In a phase II trial of cediranib in 47 evaluable patients with mesothelioma that had progressed after first-line chemotherapy, four patients had partial responses and 16 patients had stable disease, (disease control rate 42%), with median PFS of 2.6 months and median OS of 9.5 months. ³⁷ A phase I/II trial of cediranib in combination with cisplatin plus pemetrexed as first-line systemic therapy in patients with mesothelioma is currently underway (NCT01064648).

Multitargeted small molecule tyrosine kinase inhibitors (TKIs) with anti-VEGF activity also have been investigated in mesothelioma. Vatalanib, an oral anilinophthalazine that targets VEGFR1 and 2, c-KIT, platelet-derived growth factor receptor (PDGFR), and c-Fms, has been studied in a small phase II trial in previously untreated patients with mesothelioma (n = 47).³⁸ Although there were five objective responses (RR 11%) and median PFS was 4.1 months, the 3-month PFS (55%) did not meet a prespecified primary endpoint of 75%.

Sorafenib, an oral inhibitor of VEGFR2 and 3, ref, PDGFR, and c-KIT, has been studied in a phase II trial for both chemotherapy-naive and previously treated patients with mesothelioma.³⁹ Although the response rate in this trial was only 4%, the 3-month PFS was 78% and median PFS was 3.7 months, with a median OS of 10.7 months.

In a phase II trial of sunitinib, patients who had progressed after 1 prior therapy received second-line treatment with this oral inhibitor of VEGFR1–3, PDGFR, and c-Kit.⁴⁰ Among 22 assessable patients, there were three responses (RR 15%), with a median TTP of 3.5 months and a median OS of 5.9 months. In the face of limited activity for these agents, there are no current plans to develop these agents further in mesothelioma.

Histone Deacetylase Inhibition

Histones serve as a protein spool around which DNA is wound; the wrapping and unwrapping of DNA about these central histones are regulated by histone acetyltransferases and histone deacetylases (HDACs). Inhibitors of HDAC can alter access of transcription factors to DNA, thereby causing increased expression of some genes but repression of others.^41–43 In a phase I trial of vorinostat, an oral HDAC inhibitor, there were 13 patients with mesothelioma. Of these, two patients had objective response to vorinostat monotherapy (RR 15%), and six patients remained on therapy for more than 4 months.⁴⁴ From signals of activity for this agent in mesothelioma, an international phase III trial is underway in which patients who have progressed after prior pemetrexed-containing therapy are randomized to receive vorinostat or placebo (NCT00128102).

Anti-Mesothelin Antibodies

Mesothelin is a glycosyl-phosphatidyl inositol-linked (GPI-anchor) cytoplasmic membrane glycoprotein thought to be involved in cell adhesion and is tightly associated with a range of cancers, including mesothelioma.⁴⁵ Three anti-mesothelin agents are in clinical development: Morab-009 is a monoclonal anti-mesothelin antibody,⁴⁶ SS1P is an anti-mesothelin antibody with a conjugated toxin (Pseudomonas exotoxin A),⁴⁷ whereas CRS-207 is a mesothelin vaccine derived from attenuated Listeria monocytogenes and intended to elicit an antibody response against surface mesothelin.⁴⁸ By targeting surface mesothelin, investigators hope to block cell adhesion as well as elicit an antibody-dependent cytotoxicity response against mesothelin-positive tumor cells.^46,49 Separate phase II trials of MORAb-009 (NCT00738582) and SS1P (NCT01445392) in combination with cisplatin plus pemetrexed have completed accrual, but no results have yet been released. A phase I trial of CRS-207 monotherapy has completed accrual, though it appears that its sponsor will be focusing its development in pancreatic cancer at this time.

Other Targets of Interest

In addition to the work mentioned previously, there are ongoing efforts in mesothelioma exploring the efficacy of mammalian target of rapamycin (mTOR) inhibition (NCT00770120), G2 checkpoint abrogation (NCT00700336), human tumor necrosis factor-α (hTNF-α) (NCT01098266), and vaccines targeting the Wilms tumor suppressor gene (WT1) (NCT01265433). Although it is too early to determine the role that these agents will have in the standard treatment of patients with mesothelioma, it is encouraging to see the investment in clinical drug development and infrastructure in this disease.

Although current systemic treatment options for mesothelioma are limited, many novel agents are being investigated. At this time, first-line therapy with cisplatin plus pemetrexed remains the standard of care. In addition, both gemcitabine and vinorelbine have demonstrated activity in this disease, alone or in combination with cisplatin, and their use is worthy of consideration in patients with mesothelioma who have progressed after pemetrexed-based therapy. Newer agents aimed at the inhibition of targets such as VEGF or HDAC are undergoing further investigation. Despite the modest early clinical results, the use of clinical or biological markers to select a more appropriate treatment group may ultimately result in better outcomes in such refined populations. Additional clinical trials of these and other agents are needed to provide more and better treatment options for patients with mesothelioma, and a greater commitment is required to identify and enroll patients with this disease into appropriate clinical trials.

The study of mesothelioma in specialized medical centers in the 1980s initiated the search for effective chemotherapy for this diffuse disease of the pleura. The search for a systemic drug has been hampered by the infrequency of the disease (only 3000 cases per year in the United States) and by the inability to accrue homogeneous groups in timely fashion to prospective trials (i.e., groups with mixed histologies or mixed stages). Finally, the variability of staging systems also has hampered the collection of large groups for data analysis. The take-home message of this chapter regarding chemotherapy and mesothelioma is disappointingly simple: few chemotherapeutic agents are effective at this time. Only one drug, the antifolate, pemetrexed, has received FDA approval for malignant mesothelioma. First-line therapy is cis-platinum plus pemetrexed. Although other investigational drugs have been disappointing, to date, many novel agents are being investigated. Historically, chemotherapy was anticipated to have the best probability of success for malignant mesothelioma. The diffuse nature of the disease was thought to preclude surgical therapy. Today, the diffuse nature of the disease requires combinations of multiple modalities of treatment.