- © 2008 by American Society of Clinical Oncology
Induction of Immune Responses and Clinical Efficacy in a Phase II Trial of IDM-2101, a 10-Epitope Cytotoxic T-Lymphocyte Vaccine, in Metastatic Non–Small-Cell Lung Cancer
- Minal Barve,
- James Bender,
- Neil Senzer,
- Casey Cunningham,
- F. Anthony Greco,
- David McCune,
- Ronald Steis,
- Hung Khong,
- Donald Richards,
- Joe Stephenson,
- Prasanthi Ganesa,
- Jackie Nemunaitis,
- Glenn Ishioka,
- Beena Pappen,
- Michael Nemunaitis,
- Michael Morse,
- Bonnie Mills,
- Phillip B. Maples,
- Jeffrey Sherman and
- John J. Nemunaitis
- From the Mary Crowley Cancer Research Centers; Baylor Sammons Cancer Center; Gradalis Inc; and Texas Oncology Physicians Association, Dallas; Tyler Cancer Center, Tyler, TX; IDM Pharma Inc, Irvine; Pharmexa-Epimmune, San Diego, CA; Sarah Cannon Cancer Center, Nashville, TN; Madigan Army Medical Center, Tacoma WA; Atlanta Cancer Care, Roswell, GA; University of South Alabama, Mitchell Cancer Institute, Mobile, AL; Cancer Center of the Carolinas, Greenville, SC; and Duke University Medical Center, Durham, NC
- Corresponding author: John J. Nemunaitis, MD, 1700 Pacific Ave, Suite 1100, Dallas, TX 75201; e-mail: jnemunaitis{at}marycrowley.org
Abstract
Purpose Generation of broad cytotoxic T-lymphocyte responses against multiple epitopes and tumor-associated antigens (TAAs) may provide effective immunotherapy in patients with cancer. We evaluated a single-vial peptide vaccine consisting of nine HLA-A2 supertype-binding epitopes (two native and seven analog epitopes modified for optimal HLA binding or T-cell receptor stimulation) covering five TAAs and the universal helper pan-DR epitope, formulated as a stable emulsion with incomplete Freund's adjuvant (Montanide ISA 51; Seppic SA, Paris, France). The clinical efficacy, safety, and multiepitope immunogenicity of IDM-2101 was evaluated in patients with stage IIIB or IV non–small-cell lung cancer (NSCLC).
Patients and Methods A total of 63 patients were enrolled who were positive for HLA-A2. End points included survival, safety, and immune response. IDM-2101 (previously EP-2101) was administered every 3 weeks for the first 15 weeks, then every 2 months through year 1, then quarterly through year 2, for a total of 13 doses. Epitope-specific cytotoxic and helper T-lymphocyte immunogenic responses were measured by the interferon gamma enzyme-linked immunosorbent spot assay.
Results No significant adverse events were noted. Low-grade erythema and pain at the injection site were the most common adverse effects. One-year survival in the treated patients was 60%, and median survival was 17.3 months. One complete and one partial response were identified. Survival was longer in patients demonstrating an immune response to epitope peptides (P < .001).
Conclusion IDM-2101 was well tolerated, and evidence of efficacy was suggested.
INTRODUCTION
Non–small-cell lung cancer (NSCLC) is the most common cancer diagnosed in the United States. With the recent addition of targeted therapies to our armamentarium, survival of patients initially diagnosed with advanced disease (stages IIIB and IV) has improved. Unfortunately, virtually all patients experience recurrence, and 5-year survival remains less than 5%.
Peptide epitopes have been used for the induction of cytotoxic T-lymphocyte (CTL) responses in patients with cancer in numerous clinical studies, with some encouraging results.1-3 The heterogeneity of tumor-associated antigen (TAA) expression, the potential for TAA loss, and variability of the human T-cell repertoire suggest that effective cancer vaccines require induction of a wide breadth of CTL specificities. This can best be achieved with vaccines targeting multiple TAAs.
The IDM-2101 vaccine was designed to induce CTLs against five TAAs frequently overexpressed in NSCLC (ie, carcinoembryonic antigen [CEA],4 p53,5,6 HER-2/neu,7,8 and melanoma antigens [MAGE] 2 and 39). These TAAs have been used in previous vaccine studies involving patients with NSCLC1-3,10-26 and have been extensively characterized in the literature. IDM-2101 is composed of 10 synthetic peptides from these TAAs, nine of the peptides representing CTL epitopes. Each CTL epitope is restricted by HLA-A2.1 and at least one other member of the HLA-A2 superfamily of major histocompatibility complex class I molecules, providing coverage of approximately 45% of the general population. The tenth synthetic peptide is the pan-DR epitope (PADRE), a rationally designed helper T-lymphocyte (HTL) epitope included to augment the magnitude and duration of CTL responses.27
It was previously demonstrated that all of the epitopes selected were immunogenic.28-30 The utility of the analog peptides was further documented, as the CTLs generated were capable of recognizing wild-type epitopes expressed on tumor cell lines.28,31,32 Moreover, others have demonstrated measurable recall or postvaccination CTL responses against most of the nine vaccine CTL epitopes (including the wild-type versions of analogs in the current vaccine).2,33-39 The immunogenicity of IDM-2101 after in vivo immunization was also demonstrated in HLA-A2.1/Kb transgenic mice, which express the human HLA-A2.1 molecule.27,40-43
Two phase I clinical trials previously demonstrated the safety and immunogenicity of the IDM-2101 vaccine in patients with colon cancer or NSCLC who were rendered disease-free by standard therapy. A total of 24 patients were enrolled, 16 of whom completed six injections over 18 weeks.44 No significant adverse events were observed, and a strong CTL response in the majority of patients tested was observed.2,24,45,46
We report here the efficacy, safety, and immunogenicity results of IDM-2101 in HLA-A2–positive patients with advanced NSCLC.
PATIENTS AND METHODS
Study Design
This was an open-label, multicenter, single-arm, phase II study designed to evaluate the safety, efficacy (response and survival), and immunogenicity of IDM-2101 in patients with advanced NSCLC who were HLA-A2 positive. The vaccine was administered subcutaneously every 3 weeks for the first 15 weeks, then every 2 months through year 1, then quarterly through year 2, for a total of 13 doses.
Patients enrolled had tissue typing positive for HLA-A2. Tumor staging was performed at baseline with reassessment at weeks 9 and 18 and at months 6, 9, and 12. Leukapheresis (5 L) was performed before vaccination (at screening) and at weeks 9 and 18 to obtain sufficient cells to conduct the immunogenicity assays. Hematology, electrolytes, liver, other organ functions, urinalysis, and antinuclear antibody titer were assessed. Toxicity was monitored and graded according to the National Cancer Institute Common Toxicity Criteria. All patients signed the protocol-specific local institutional review board approved informed consent form. Response Evaluation Criteria in Solid Tumors were used to evaluate response.
Study Population
Eligible patients for this study were 18 years or older with histologic confirmation of stage IIIB or IV or recurrent NSCLC. Patients had to have an Eastern Cooperative Oncology Group performance status of 0 or 1, an absolute granulocyte count ≥ 1,500/μL, platelet count ≥ 100,000/μL, hemoglobin ≥ 10 g/dL, total bilirubin ≤ 2 mg/dL, AST and ALT ≤ 2.5× the upper limit of normal, and serum creatinine ≤ 2× the upper limit of normal. An estimated tumor volume of ≤ 125 mL was required (not including lymph nodes and bone disease). Patients with brain metastases were eligible if the disease was clinically stable for at least 2 months before study entry.
Patients were excluded if they had received prior cancer vaccine therapy or systemic corticosteroids, interferons, or interleukins within 1 month of study entry. Patients with history of HIV or Hepatitis B or C were not eligible to participate in the study.
A second group of patients who were HLA-A2 negative when screened for the study were retrospectively identified to use as a comparator group.
Study Drug
IDM-2101 vaccine was provided by IDM Pharma Inc (Irvine, CA), in 2-mL glass vials to each study site. The IDM-2101 drug product is a preservative-free, sterile emulsion of 10 synthetically manufactured peptides and an aqueous/dimethyl sulfoxide buffer system and is emulsified with incomplete Freund's adjuvant (Montanide ISA 51; Seppic SA, Paris, France).47 The product is formulated with 0.5 mg/mL of each peptide (5.0 mg/mL total peptide) and refrigerated at 2° to 8°C. The specific peptide IDs and sequences are listed in Table 1. Product release testing included appearance, endotoxin, sterility, viscosity, particle size, peptide concentration of each peptide, volume, pH, and potency, and all results were acceptable.
Previous cancer trials from other laboratories have tested escalating doses of peptide, ranging from 0.1 to 10 mg of peptide per injection dose, emulsified in incomplete Freund's adjuvant.1,48-52 At all doses tested, the peptide/incomplete Freund's adjuvant vaccine was deemed to be safe and well tolerated, with no severe dose-related systemic toxicities being reported.
Statistical and Analytic Plan
Demographic and baseline information for all patients was presented and summarized. Continuous variables were summarized using mean, median, standard deviation (SD), and range of values. All categoric data were summarized by the frequency and percentage in each category. All patients who were exposed to study drug were included in the safety and survival analyses.
Overall survival was illustrated using the Kaplan-Meier estimator. The HLA-A2–negative group was not prospectively observed after determination of negative HLA type. Immune response relationship to survival was done by comparing the number of epitopes with measurable enzyme-linked immunosorbent spot assay (ELISPOT) responses in relation to survival using the log-rank statistic.
Immune Function Testing
To measure CTL responses, 2 × 106 peripheral-blood mononuclear cells (PBMCs)/well (three to four wells per epitope) were stimulated in vitro with each vaccine peptide (10 μg/mL). Ten U/mL of rIL-2 was added after 24 hours. After 10 days of culture, the in vitro–expanded PBMCs were tested for epitope-specific (vaccine epitope and wild-type epitope of vaccine analog) CTL responses, measured by an 18-hour interferon gamma ELISPOT assay.44 Briefly, 5 × 104 PBMCs were stimulated in each assay well with 10 μg/mL of peptide (in vitro expansion peptide and wild-type peptide if testing a vaccine analog) and 105 irradiated autologous PBMCs as antigen-presenting cells. Irrelevant HLA-A2–binding hepatitis B virus core antigen peptide was used as a negative control. Spot-forming cells (SFCs) were enumerated with a computer-assisted ELISPOT reader (Carl Zeiss MicroImaging Inc, Thornwood, NJ).
HTL responses were measured from PBMCs without an in vitro expansion step. PBMCs were thawed, rested overnight in medium, and 2 × 105 PBMCs/well were stimulated with 10 μg/mL PADRE or irrelevant malaria peptide in the interferon gamma ELISPOT assay.
A positive response (number of cells that respond to antigen by secreting gamma interferon) to an epitope was defined as a response that met all three of the following criteria: (1) net number of spots (after subtracting background) is greater than 5 spots/50,000 cells, (2) net number of spots exceeds the background spots plus two SDs, and (3) doubling occurs of the net spots at postvaccination compared with the baseline spots, plus two SDs.
Data are expressed as net SFCs/5 × 104 cells induced by the vaccine epitope after subtracting background (SFC against the irrelevant hepatitis B virus or malaria epitope peptide) and as fold of epitope-induced SFC over background SFC.
RESULTS
Sixty-three patients were found to be HLA-A2 positive and were enrolled onto the study for treatment with IDM-2101 between February 2005 and March 2006. A total of 351 vaccinations were administered. Mean age of the HLA-A2–positive patients was 64 years (range, 26 to 87 years), and 35 patients (56%) were male. Survival and toxicity were assessable in all HLA-A2–positive patients. Demographics of the HLA-A2–positive patients are shown in Table 2.
Safety
One grade 3 fever/chills was observed related to study medication. Other treatment related grade 1 to 2 adverse events (total, 63) included injection site erythema (21%), fatigue (16%), injection site pain (14%), and fever (14%). Overall treatment was well tolerated and without significant clinical sequela. There was no adverse affect on hematopoietic, gastrointestinal, ocular, or neurologic function.
Survival
One-year survival for the treated group was 60%, and the median survival (Fig 1) was 520 days (17.3 months). One patient had a complete response (Fig 2), and one patient had a partial response. Stable disease of 3 months or greater by Response Evaluation Criteria in Solid Tumors was observed in 54 additional patients (86%) in the IDM-2101–treated group. Benefit as measured by stable disease or partial/complete response was achieved in 89% of patients. Seventeen patients (27%) continued treatment for 1 year and demonstrated no evidence of progressive disease. Fourteen patients (22%) completed 2 years of dosing and remain without evidence of progressive disease.
Description of the patients achieving response is as follows. The first achieved a complete response. He was a 65-year-old white male patient with a 40-packs-per-year history of smoking who presented with stage IV moderately differentiated squamous cell carcinoma. He demonstrated partial response to carboplatin and paclitaxel. Recurrent lesions before vaccination included a 1.6 × 1.6 cm right lower lobe nodule and three subcentimeter left lower lobe nodules. Five months after IDM-2101 administration, computed tomography (CT) scan showed complete resolution of all disease (Fig 2). Follow-up CT scans and positron emission tomography scans performed 1 month later and at 10, 20, 29, and 32 months after treatment all demonstrated no evidence of disease. Evaluation of the immune responses in this patient showed robust responses to seven of nine vaccine epitopes at both weeks 9 and 18. The other patient who achieved a partial response had progressive/recurrent large-cell lung carcinoma diagnosed in March 2005 involving his mediastinal nodes, with left pleural effusion after a left upper lobe lobectomy in 2004 (T2N0). He received radiation therapy and low-dose weekly paclitaxel/carboplatin for 6 weeks. On July 7 (5 weeks after radiation therapy/weekly chemotherapy), 2005, his chest CT showed the mediastinal nodal mass to be 3.8 × 1.3 cm, compared with 6.1 × 2.6 cm in April 2005. Left pleural effusion persisted. Clinically it was felt that there was substantial residual tumor owing to the nodal mass and the pleural effusion. He was then started on IDM-2101. His CT chest on September 27, 2005, showed the mediastinal tumor to be 1.9 × 1.1 cm and revealed unchanged left pleural effusion. His measurable mass met criteria for partial response. Unfortunately, disease progressed several months later, and the patient died of progressive disease in his lungs.
Immune Response
Samples for evaluation of immune response were available from 33 patients. Responses ranged between 10 and 50 spots/5 × 104 PBMCs, with stronger responses of more than 100 spots observed. The first 11 patients were tested for a response to all of the peptides, and the remaining 22 patients were tested for response to the five strongest epitopes (CEA24, CEA605, HER2.369, MAGE3.112 [with wild-type cross-over], and MAGE2.157 [wild-type]). The percent of patients responding to each of the CTL epitopes is shown in Figure 3. All nine vaccine epitopes were immunogenic in at least one patient. Four analog epitopes showed strong responses that cross-reacted with wild-type peptide. Ninety-one percent of patients demonstrated CTL response to at least one of the five strongest epitopes. Eighty-five percent demonstrated responses to two or more, 64% responded to three or more, 39% responded to four or more, and 18% responded to five or more epitopes. In three of four patients tested at 12 months, responses were observed to four of five epitopes tested. Longer survival was associated with a higher number of positive responses. If a zero- or one-epitope response occurred, survival was 406 days (95% CI, 292 to 520 days). By comparison, a two- to three-epitope response was associated with 778 days (95% CI, 637 to 919 days) of survival, and patients achieving a four- to five-epitope response had 875 days (95% CI, 743 to 1,007 days) of survival (P < .001). Interferon gamma–producing helper T cells against PADRE were detected in 18 (55%) of 33 patients tested in a direct ELISPOT assay (Fig 4).
DISCUSSION
These results suggest remarkable safety and intriguing activity of IDM-2101 with respect to clinical response, proportion of patients achieving initial stable disease, and overall survival in comparison with historical patients who received conventional treatment. Moreover, this study reports the induction of broad and durable immune responses to the vaccine and a favorable safety profile, with treatment-associated adverse events limited to injection site reactions.
Immune responses in 33 patients collectively showed induction of CTLs to all of the vaccine epitopes after vaccination. Although patient to patient variability was observed in the frequency and magnitude of the CTL responses, 85% of tested patients responded to at least two epitopes. These data are consistent with reported results from an earlier phase I trial.44 Moreover, longer survival was shown in patients achieving responses to two or more epitopes (P < .001).
Remarkably, induction of responses to the HTL epitope PADRE was observed without in vitro amplification in 55% of the patients, indicating a greater immunogenicity of this peptide and providing further support for its value as a source of T-cell help in sustaining the responses to the CTL peptides.
The treatment regimen of patients with locally advanced or metastatic NSCLC after experiencing treatment failure with first-line chemotherapy is variable but generally involves docetaxel or pemetrexed.53,54 Hanna et al55 reported overall survival values in a similar population as ours with either pemetrexed or docetaxel of 8.3 months or 7.9 months, respectively. One-year survival was approximately 30% for both groups. Similarly, Shepherd et al56 reported docetaxel (75 mg/mm2) versus best supportive care and showed a median survival of 7.0 months versus 4.6 months, respectively, and a 1-year survival rate of 37% versus 11%, respectively. A more recent randomized, two-dose efficacy trial of gefitinib, again in similar patients, also revealed response rates of approximately 19% and median overall survival of approximately 8 months in both dose groups.57 The 1-year survival rate was 35% and 29% for the high- and low-dose groups, respectively. In a second randomized trial, slightly lower overall survival values were reported for gefitinib-treated patients who had been more heavily exposed to prior chemotherapy. In this study, an overall survival of approximately 6 months and a 1-year survival rate of 25% were reported.58
The populations in these different studies were similar to each other, and the overall survival times of approximately 8 months seem remarkably similar. The 1-year survival rate for patients in different stages of therapy was also similar, ranging from 25% to 37%. Thus our observation in similar patients of a median survival of better than 17 months and a 60% 1-year survival rate in HLA-A2–positive patients treated with IDM-2101 was encouraging. However, patient selection issues within the relatively small number of patients treated and observed may in part account for survival.
Nearly 90% of the IDM-2101 patients also demonstrated stable disease or better 3 months into treatment. This was somewhat of a surprise given the delay generally expected in induction of an immune response after vaccination, as demonstrated in testing of two gene-based vaccines, granulocyte-macrophage colony stimulating factor vaccine (GVAX; Cell Genesys, South San Francisco, CA)59 and belagenpumatucel (Lucanix; NovaRx, San Diego, CA),60 with which we have extensive experience. The treated patient population demographics were similar between GVAX, Lucanix, and IDM-2101. Education of the immune system through the endogenous dendritic cell process may be more rapidly activated using IDM-2101. Somewhat analogous results were suggested in another vaccine trial using the HLA-A2–positive restricted epitope E75, which is similar to the HER-2/neu antigen peptide we used in a trial61 involving high-risk early-stage breast cancer. In conclusion, these data support a recommendation that IDM-2101 move forward to a phase III validation trial in patients with advanced NSCLC who have previously experienced treatment failure with combination cytotoxic therapy.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Minal Barve, James Bender, Neil N. Senzer, Casey Cunningham, F. Anthony Greco, David McCune, Ronald Steis, Hung Khong, Donald Richards, Joe Stephenson, Glenn Ishioka, Michael Morse, Bonnie Mills, Jeffrey Sherman, John J. Nemunaitis
Administrative support: John J. Nemunaitis
Provision of study materials or patients: Jeffrey Sherman, John J. Nemunaitis
Collection and assembly of data: Minal Barve, James Bender, Neil N. Senzer, Casey Cunningham, F. Anthony Greco, David McCune, Ronald Steis, Hung Khong, Donald Richards, Joe Stephenson, Prasanthi Ganesa, Glenn Ishioka, Beena Pappen, Michael Nemunaitis, Michael Morse, Bonnie Mills, Phillip B. Maples, Jeffrey Sherman, John J. Nemunaitis
Data analysis and interpretation: Minal Barve, James Bender, Neil N. Senzer, Casey Cunningham, F. Anthony Greco, David McCune, Ronald Steis, Hung Khong, Donald Richards, Joe Stephenson, Prasanthi Ganesa, Jackie Nemunaitis, Glenn Ishioka, Beena Pappen, Michael Nemunaitis, Michael Morse, Bonnie Mills, Phillip B. Maples, Jeffrey Sherman, John J. Nemunaitis
Manuscript writing: Minal Barve, James Bender, Neil N. Senzer, Casey Cunningham, F. Anthony Greco, David McCune, Ronald Steis, Hung Khong, Donald Richards, Joe Stephenson, Prasanthi Ganesa, Jackie Nemunaitis, Glenn Ishioka, Beena Pappen, Michael Nemunaitis, Michael Morse, Bonnie Mills, Phillip B. Maples, Jeffrey Sherman, John J. Nemunaitis
Final approval of manuscript: Minal Barve, James Bender, Neil N. Senzer, Casey Cunningham, F. Anthony Greco, David McCune, Ronald Steis, Hung Khong, Donald Richards, Joe Stephenson, Prasanthi Ganesa, Jackie Nemunaitis, Glenn Ishioka, Beena Pappen, Michael Nemunaitis, Michael Morse, Bonnie Mills, Phillip B. Maples, Jeffrey Sherman, John J. Nemunaitis
Acknowledgments
We thank Susan Mill for her assistance with manuscript preparation.
Footnotes
-
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical trial information can be found for the following: NCT00054899.
- Received February 29, 2008.
- Accepted May 21, 2008.