Phase II Clinical Trial of Amatuximab, a Chimeric Antimesothelin Antibody with Pemetrexed and Cisplatin in Advanced Unresectable Pleural Mesothelioma

Malignant pleural mesothelioma (MPM) is an aggressive disease with poor prognosis. Although patients with a limited tumor burden may benefit from surgical resection, most patients have advanced disease at diagnosis and are not candidates for surgery (1). For patients who are not eligible for curative surgery, the median survival with supportive care alone is approximately 6 months, whereas with the current standard treatment, a combination of cisplatin and pemetrexed, the median survival is 12 months (2, 3).

Mesothelin is a glycosylphosphatidyl inositol (GPI)–anchored membrane glycoprotein, which is present in a restricted set of normal adult tissues such as the mesothelium (4). In contrast, mesothelin is highly expressed in many epithelial cancers. More than half of all the ovarian cancers and lung adenocarcinomas and nearly all epithelial mesotheliomas and pancreatic ductal adenocarcinomas express mesothelin (5–9). Although the normal biologic function of mesothelin is unknown, growing evidence suggests that it may play a role in tumorigenesis and metastasis in mesothelioma (10). Its limited expression in normal human tissue and high expression in tumor make mesothelin an excellent target antigen for antibody-based immunotherapy (11).

The mesothelin gene encodes a 71-kDa precursor protein that is cleaved into a soluble 31-kDa fraction, megakaryocyte potentiating factor (MPF), and the 40-kDa mesothelin (12). Mesothelin binds to CA125, a specific epitope expressed on MUC16, a transmembrane mucin. The interaction between CA125, which is present on a majority of mesothelioma cells, and mesothelin has been suggested to facilitate implantation and metastasis of mesothelioma (13–15). Serum mesothelin, MPF, and CA125 could be potentially useful as biomarkers for mesothelioma (16–20).

Amatuximab (MORAb-009) is a chimeric high-affinity monoclonal IgG1/k antibody targeting mesothelin (21). In vitro, amatuximab elicits antibody-dependent cellular cytotoxicity (ADCC) against mesothelin-expressing tumor cell lines and inhibits heterotypic cell adhesion of mesothelin-positive tumor cells to CA125-expressing tumor cells. In tumor xenograft studies, combination treatment with amatuximab plus chemotherapy led to a greater reduction in the growth of mesothelin-expressing tumors than either amatuximab or chemotherapy alone. In a phase I study of patients with mesothelin-expressing cancers, weekly infusions of amatuximab were well tolerated, and the MTD was identified as 200 mg/m² (22). Dose-limiting toxicities were grade 4 transaminitis and grade 3 serum sickness. Other adverse events at least possibly related to amatuximab included grade 1 or 2 drug hypersensitivity. In the phase I study, amatuximab treatment resulted in an increase in serum CA125, possibly due to inhibition of binding of tumor shed CA125 to mesothelin present on the serosal lining of pleural and peritoneal cavities (23). On the basis of its safety in the phase I study and preclinical studies showing synergy with chemotherapy, amatuximab was combined with pemetrexed and cisplatin in a single-arm phase II study in patients with unresectable MPM.

Patients with histologically confirmed, chemotherapy-naive MPM who were not candidates for curative surgery were assessed for eligibility. The study was approved by the Institutional Review Boards of participating institutions, and informed consent was obtained before enrolment. The trial was registered at clinicaltrials.gov (identifier NCT00738582).

Eligibility criteria included age ≥18 years, epithelial type or biphasic (mixed) MPM with low sarcomatous content, radiographically measurable disease, Karnofsky performance status (KPS) score of ≥70, adequate bone marrow reserve [absolute neutrophil count (ANC) ≥1.5 × 10⁹/L; platelet count ≥100 × 10⁹/L; hemoglobin ≥ 9 g/dL], hepatic function [bilirubin ≤1.5 times the upper limit of normal (ULN); alanine transaminase and aspartate transaminase ≤2.5 × ULN; alkaline phosphatase ≤3.0 × ULN], and normal renal function (serum creatinine ≤2.0 mg/dL and a calculated creatinine clearance ≥60 mL/min) based on the standard Cockcroft and Gault formula. Pregnant women were not eligible, and all men and women of reproductive potential were required to use an approved method of birth control. Patients were excluded if they had predominantly mesothelioma of the sarcomatous type, disease located primarily in the peritoneum, prior systemic therapy or radiotherapy, known central nervous system tumor involvement, treatment within three months of the start of the trial with other immunomodulatory therapy, and known hypersensitivity to any of the following: monoclonal antibodies or biologic therapy, pemetrexed, cisplatin, or other platinum-containing compounds.

Amatuximab (5 mg/kg) by intravenous infusion over 1 hour was administered on days 1 and 8 with pemetrexed (500 mg/m² by intravenous infusion over 10 minutes) and cisplatin (75 mg/m² by intravenous infusion over 2 hours) administered on day 1 of 21-day cycles for up to six cycles. Patients with objective response or stable disease continued to receive amatuximab maintenance (5 mg/kg on days 1 and 8 of a 21-day cycle) until disease progression. Supportive treatment included premedication with acetaminophen (650 mg) and diphenhydramine (25 to 50 mg) 30 minutes before amatuximab infusion; folic acid (350 μg to 1 mg orally daily) starting at least 5 days before the first dose of pemetrexed; vitamin B12 (1 mg approximately every 9 weeks); dexamethasone (4 mg orally twice daily) on the day before, the day of, and the day following pemetrexed. Dose adjustments for adverse events and management of infusion-related adverse events are described in Supplementary Appendix A.

Tumor measurements were performed at baseline and thereafter on day 8 of every third cycle starting with cycle 3 until disease progression or treatment discontinuation. Response was assessed using the modified RECIST for the assessment of response in malignant pleural mesothelioma (24). A complete blood count and a comprehensive metabolic panel were performed on days 1 and 8 of each cycle during the combination therapy phase and on day 1 of each cycle during the amatuximab maintenance phase. Safety was assessed using the Common Terminology Criteria for Adverse Events scale, version 3.0.

Pharmacokinetic assessments were performed and will be reported as a separate article. Serum mesothelin and MPF were measured using Morphotek proprietary assays. CA125 levels (U/mL) were measured using an automated commercial assay. All assays were run according to the manufacturer's instructions, blinded to patient data. Relative changes in MPF and CA125 levels were compared with the patients best overall radiologic response during treatment. Changes in serum mesothelin levels were not assessed because amatuximab is known to interfere with mesothelin assays.

This was a multicenter, single-arm phase II study. The primary objective was to determine the 6-month progression-free survival (PFS) of amatuximab plus pemetrexed and cisplatin. In the phase III trial of cisplatin and pemetrexed in MPM, the median time to progression was 5.7 months, which corresponded to a 6-month PFS response rate of 48.2% (2). By adding amatuximab to the combination, we sought to demonstrate an improvement in median PFS to 8.7 months corresponding to 6-month PFS response rate of 62%. The study used Simon optimal 2-stage design (25) and set the probability for accepting a poor drug to 10% (α = 0.10) and the probability of rejecting a good drug at 15% (β = 0.15). The Kaplan–Meier (K-M) methodology was used to estimate the median of time-to-event endpoints (26). A two-sided 95% confidence interval (CI) was constructed using the methodology of Brookmeyer and Crowley (27). K-M estimates at selected time points (e.g., 3, 6, 9, and 12 months) were determined, and corresponding 95% CIs were constructed using the log–log transformation methodology.

The secondary endpoints were objective response rate (ORR), overall survival (OS), overall PFS, and safety and tolerability of the combination.

The ORR was based on an independent assessment of total tumor measurement. An exact two-sided 95% CI for the ORR was constructed. Simple descriptive statistics (i.e., number and percentage of patients) were used to summarize safety data. All patients who received at least one dose of amatuximab were included in the safety analysis and OS efficacy assessment. The population for the PFS analysis was defined as all patients who received treatment and underwent at least one postbaseline imaging assessment, or who had died.

A post hoc exploratory objective was to determine optimal thresholds for biomarker levels to predict survival. These optimal thresholds were determined using maximally selected χ² statistics (28). Other post hoc exploratory investigations included an analysis of OS by European Organization for Research and Treatment of Cancer (EORTC) prognostic score category (low-risk vs. high-risk; ref. 29) using the K-M methodology, and a stepwise multivariate Cox regression analysis of OS using the categorized biomarker data (above or below the optimal threshold for baseline CA125, mesothelin, and MPF) and three categorized elements of the EORTC prognostic score [baseline white blood cell count (WBC) above or below 8.3 × 10⁹/L, baseline Eastern Cooperative Oncology Group (ECOG) score (0 vs. 1 or 2), and gender (male vs. female)]. A 0.10 level of significance was used for selection and retention of factors in the Cox model. EORTC prognostic variables of histology and probability of diagnosis were not included in the model because all patients had confirmed diagnosis of epithelioid mesothelioma.

An independent data monitoring committee (IDMC) was established to review safety data after 8 patients had completed one cycle, and after 17 and 33 patients had completed 6 cycles. Because of the rapid enrollment of this study, this initial safety review occurred after 17 patients had completed one cycle. In addition, the IDMC reviewed all safety data on a quarterly basis.

Between February 2009 and October 2010, 89 patients from 26 sites in the North America and Europe were enrolled. Twelve patients were enrolled over the design-specified target of 77, under the a priori assumption of an approximate 10% loss to follow-up and nonevaluability for tumor assessment. All patients received at least one dose of amatuximab. Eighty-three (93%) patients had at least one postbaseline imaging assessment and were evaluable for efficacy. One patient withdrew consent after their third cycle.

Patient demographics and baseline disease characteristics are shown in Table 1. The median age was 67 years (range, 46–80 years), 78% were male, 89% were Caucasian, and 93% had a KPS score ≥80. Eighty-eight percent had stage III/IV disease and 89% had epithelial histology. Sixty-three percent reported prior exposure to asbestos.

Forty patients (45%) completed 6 cycles of the combination therapy phase. Reasons for discontinuation of combination therapy included adverse events (35%), progressive disease (12%), investigator discretion (3%), and other (5%). Fifty-six patients (63%) entered the amatuximab maintenance phase. Reasons for discontinuation of maintenance therapy were progressive disease (73%), adverse events (5%), and other (4%). A median of five cycles of combination therapy (range, 1–6) and six cycles of maintenance therapy (range, 1–30) were administered. Sixteen patients received ≥10 cycles, and eight patients received ≥20 cycles of amatuximab maintenance, the longest being 52 cycles.

Efficacy data are shown in Table 2. The study did not meet the prespecified design criterion of 43 responders in the first 77 patients, and the study did not meet its primary endpoint. The K-M estimate of 6-month PFS was 51.3% (95% CI, 39.1–62.3; Fig. 1A), median PFS was 6.1 months (95% CI, 5.8–6.4), and median OS was 14.8 months (95% CI, 12.4–19.2; Fig. 1B). There were no complete responses; 33 of 83 patients had partial responses [ORR, 39.8% (95% CI, 29.2–51.1)]. Forty-two patients (51%) had stable disease as the best response. The disease control rate was 90%. As of April 23, 2012, 29 patients are alive and 5 remain on amatuximab maintenance. A subgroup of patients who received at least 4 cycles of combination chemotherapy (n = 60) had a median OS of 19.2 months (95% CI, 13.5–20.8). The posthoc exploratory analysis of the median OS by EORTC prognostic score category showed that those subjects with a score indicating a low risk (n = 25) had a median OS of 20.7 months (95% CI, 16.0–28.6) as compared with those with a score indicating a high risk (n = 64) having a median OS of 12.6 months (95% CI, 11.3–17.1).

Treatment-emergent adverse events are shown in Table 3. Among the most common adverse events seen in ≥15% of patients during the combination therapy phase were nausea (71%), fatigue (61%), anorexia (43%), vomiting (32%), constipation (30%), anemia (29%), neutropenia (29%), diarrhea (28%), and weight decrease (20%). Hypersensitivity reactions (12%) as well as infusion-related reactions (9%) were also seen. The serious adverse events (SAE) during the combination therapy phase were hypersensitivity reactions (4 patients), neutropenia (4 patients), atrial fibrillation (3 patients), hyponatremia (3 patients), anemia (2 patients), dehydration (2 patients), and pneumothorax (2 patients). Adverse events which led to discontinuation of treatment were hypersensitivity reactions to amatuximab (8 patients), increasing serum creatinine (6 patients), fatigue (3 patients), neutropenia (2 patients), worsening in the patients' general condition (2 patients), nausea and vomiting (2 patients), development of a pneumothorax (2 patients), and one patient each for anemia, thrombocytopenia, dyspnea, pericarditis, and pericardial effusion. One additional patient was discontinued to undergo a thoracotomy.

During the amatuximab maintenance phase, the most common adverse events (>15%) seen were dyspnea (23%), nausea (20%), peripheral neuropathy (18%), fatigue (18%), and non–cardiac chest pain (16%). Grade 3 and 4 adverse events were dyspnea (3 patients), fatigue (3 patients), abdominal pain (2 patients), and flank pain (2 patients). SAEs seen during the maintenance phase were dyspnea (4 patients) and fatigue (2 patients). Adverse events which led to discontinuation of treatment were hyperbilirubinemia (1 patient), peritonitis (1 patient), and overdose/cardiopulmonary arrest (1 patient). The latter was reported in a 58-year-old man who suffered a cardiac arrest related to cocaine use, one week after the first dose of maintenance amatuximab, from which he was successfully resuscitated.

As depicted in Fig. 2A, a strong correlation was observed between pretreatment serum mesothelin levels and serum MPF levels [r = 0.77 (P < 0.0001)]. On the basis of the current dataset, we estimated optimal thresholds of baseline serum mesothelin, MPF, and CA125 to predict OS as 33.14 ng/mL, 4.7 ng/mL, and 6 U/mL, respectively. Patients with low pretreatment mesothelin levels had a significantly longer OS compared with patients with high baseline mesothelin levels [18.5 months (95% CI, 13.2–not reached) vs. 12.5 months (95% CI, 10.5–16.7); Fig. 2B]. A similar association was observed with pretreatment MPF wherein patients with low levels at baseline had superior OS compared with those with high levels [18.5 months (95% CI, 13.2–not reached) vs. 12.8 months (95% CI, 11.3–16.0); Fig. 2C] and CA125 [20.7 months (95% CI, 10.7–not reached) vs. 13.3 months (95% CI, 11.6–18.2); Fig. 2D]. The stepwise multivariate Cox regression analysis using these three categorized biomarkers and three categorized elements of the EORTC prognostic score resulted in a final model that retained the following factors: baseline mesothelin (P = 0.004), baseline ECOG status (P = 0.046), and baseline CA125 (P = 0.070). The final model was based on the 77 subjects for whom the values of all six candidate factors were known. Baseline MPF was not selected because, although associated with OS in a univariate setting, it strongly correlated with baseline mesothelin.

Serum MPF data from before and after treatment were available for 59 patients who were also evaluable for response. For each patient, the relative change in serum MPF levels at the time of best overall radiologic response (stable disease, partial response, or progressive disease) was displayed in a waterfall plot (Fig. 3). The 2 patients with progressive disease experienced an increase in serum MPF, whereas 17 of 21 (81%) patients with a confirmed partial response had a decrease in MPF levels from baseline. Among 32 patients with stable disease as the best response, 10 (31%) had an increase, whereas 22 (69%) had a decrease in serum MPF with treatment.

This multicenter phase II study demonstrated that amatuximab in combination with pemetrexed and cisplatin was well tolerated and resulted in a disease control rate of 90% and median PFS of 6.1 months by independent radiologic review in the primary efficacy population.

Response rate of 40% by independent radiologic review is comparable with the 45.5% observed with cisplatin and pemetrexed alone (2). Reproducibly measuring tumor response is challenging in MPM, and to avoid investigator bias in response assessment, we used independent radiologic review. The study did not meet the primary endpoint of 3-month improvement in PFS over historical controls. Nevertheless, with all the caveats associated with a cross trial comparison, the median OS of 14.8 months compares favorably with 13.3 months in the fully supplemented subset of patients in the phase III study of cisplatin and pemetrexed (2). However, direct comparison of results of this trial with the phase III study of cisplatin and pemetrexed is difficult due to differences in patient populations studied and frequency of response assessment. For example, the phase III trial of cisplatin and pemetrexed had more patients with poor performance status (48% patients with KPS 70/80 vs. 30% in our trial) and unfavorable histology (22% with mixed or sarcomatoid histology vs. 11% mixed histology in our trial). However, our trial had more patients with advanced stages of disease (87% patients with stage III/IV vs. 77% in the phase III trial of cisplatin and pemetrexed). The improved median OS with 33% of patients alive at the time of analysis also exceeds the literature-based expectations of outcomes for MPM (30, 31). Postprogression second-line treatment, which may influence the duration of OS, is not known for our patients.

The extended final plateau of the K-M curve of OS of patients suggests that the combination may be particularly effective in a subgroup of patients. Hypothetically, the prolongation of OS with no improvement in PFS may be a function of the mechanism of action of amatuximab whereby the immune system may be modulated to control tumor growth. Although the full extent of the mechanism of action of amatuximab and its synergy with chemotherapy is not known, amatuximab elicits ADCC against mesothelin-expressing tumor cell lines and inhibits heterotypic cell adhesion of mesothelin-positive tumor cells to CA125-expressing tumor cells (21).

The combination chemotherapy was well tolerated with no overlapping toxicities. Hypersensitivity reactions as well as infusion-related reactions were the most common amatuximab-related adverse events. These are expected given the fact that amatuximab is a chimeric monoclonal antibody, with human constant regions and murine variable regions, which contain nonself epitopes that can stimulate immune responses. The hypersensitivity and infusion reactions were not life threatening and responded to supportive care.

The development of targeted agents to which only a subset of patients responds depends on the identification of robust predictive biomarkers. In this study, we investigated the effect of pretreatment levels of serum CA125, mesothelin, and MPF levels on survival. Previous reports which identified a significant correlation between serum mesothelin and MPF levels in MPM have involved a relatively small number of patients (32, 33).Our finding of a strong correlation between pretreatment serum mesothelin levels and serum MPF levels [r = 0.77 (P < 0.0001)] confirms these findings in a large cohort. This correlation may be attributable to the release of MPF by physiologic cleavage at the furin cleavage site of the mesothelin precursor protein (19). Univariate analyses using optimal biomarker thresholds identified from this dataset showed that patients with low baseline mesothelin, MPF, and CA125 levels had a significantly longer OS compared with patients with high levels. A multivariate analysis, which included three categorized biomarkers (baseline mesothelin, MPF, and CA125 levels) and three elements of the EORTC prognostic score (baseline WBC above or below 8.3 × 109/L, baseline ECOG 0 vs. ≥ 1, and gender) showed that baseline mesothelin, baseline ECOG performance status, and baseline CA125 were prognostic of OS. It is to be noted that the number of patients included in this analysis was limited. Future studies should explore the changes in these circulating biomarkers with treatment and their potential to provide an early assessment of treatment efficacy.

In mesothelioma, objective assessment of radiologic response to treatment is difficult, and markers of response could complement radiologic assessment to discriminate between effective and ineffective treatments. Serum mesothelin has previously been reported as a potential biomarker of response in mesothelioma (34). However, amatuximab binds to the same epitope on mesothelin as one of the antibodies used in the mesothelin assay (MESOMARK Assay; Fujirebio Diagnostics, Inc.) and interferes with the measurement of serum mesothelin. This precludes the use of serum mesothelin as a biomarker of response in patients treated with amatuximab. In this study, we investigated the relative change in serum MPF levels with treatment as a biomarker of response. Although data were limited, we observed that patients with progressive disease had increases in serum MPF with treatment, whereas a majority of patients with partial responses had decreases in MPF levels.

We conclude that amatuximab plus pemetrexed and cisplatin has activity in pleural mesothelioma. The single-arm design and the limited patient numbers preclude a definitive conclusion about its survival advantage over cisplatin plus pemetrexed. However, a median OS of 14.8 months with a third of patients alive at the time of analysis is suggestive of antitumor activity of the combination of amatuximab plus pemetrexed/cisplatin. A randomized, placebo-controlled study is planned to investigate the survival benefit of this combination. Discovery of tissue or serum biomarkers predictive of response to therapy will be a priority in any further investigation.

H.L. Kindler is a consultant/advisory board member for Morphotek. T. Jahan reports receiving a commercial research grant from Morphotek; other research grants from Aduro, Boehringer Ingellheim, Genetech/OSI, Johnson and Johnson, Medimmune, Merck, Novartis, and Verastem; and is a consultant/advisory board member for Clovis. D.J. O'Shannessy, P. Fatato, J.D. Maltzman, and B.A. Wallin are employees of Morphotek. No potential conflicts of interest were disclosed by the other authors.

Conception and design: R. Hassan, H.L. Kindler, T. Jahan, I. Pastan, J.D. Maltzman, B.A. Wallin

Development of methodology: R. Hassan, D.J. O'Shannessy, J.D. Maltzman, B.A. Wallin

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R. Hassan, H.L. Kindler, T. Jahan, L. Bazhenova, M. Reck, A. Thomas, D.J. O'Shannessy, B.A. Wallin

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Hassan, T. Jahan, L. Bazhenova, M. Reck, A. Thomas, J. Parno, D.J. O'Shannessy, J.D. Maltzman, B.A. Wallin

Writing, review, and/or revision of the manuscript: R. Hassan, H.L. Kindler, T. Jahan, L. Bazhenova, M. Reck, A. Thomas, I. Pastan, J. Parno, D.J. O'Shannessy, P. Fatato, J.D. Maltzman, B.A. Wallin

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R. Hassan, A. Thomas, I. Pastan, B.A. Wallin

Study supervision: R. Hassan, M. Reck, P. Fatato, J.D. Maltzman, B.A. Wallin

Other (study execution and management): P. Fatato

We thank all the study investigators, staff at the participating centers, the referring clinicians, and especially the patients for their participation. We would also like to thank Jason Wustner and his staff of the Bioanalytical Development Group at Morphotek.

This research was supported in part by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research, and in part by Morphotek, Inc., under a Cooperative Research and Development Agreement with the NCI.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.