Patients

Patients with stage IIIB or IV NSCLC who were aged 18 years or older and who had stable disease or an objective clinical response after first-line treatment, which consisted of either chemotherapy alone or chemotherapy and radiotherapy (completed at least 3 weeks before study entry with no upper limit for time after completion), were eligible for enrollment. Eligibility requirements included an Eastern Cooperative Oncology Group (ECOG) performance status of ≤ 2, neutrophil count ≥ 1.5 × 10⁹/L, platelet count ≥ 100 × 10⁹/L, WBC ≥ 2.5 × 10⁹/L, hemoglobin ≥ 90 g/L, and an expected survival of at least 4 months. Patients were excluded for surgery or immunotherapy within 4 weeks before study entry, receipt of immunosuppressive drugs including systemic corticosteroids within 3 weeks before study entry, past or current history of neoplasm other than lung carcinoma, autoimmune disease, recognized immunodeficiency disease, clinically significant hepatic or renal dysfunction, significant cardiac disease, active infection, or splenectomy.

The study was conducted in accordance with the Declaration of Helsinki. The protocol was approved by institutional review boards with jurisdiction over the specific sites that registered patients onto the study. Each patient gave written informed consent before enrollment.

Pretreatment evaluations included a complete history, physical examination, and clinical laboratory studies. Identical radiologic or clinical evaluations used before first-line treatment were used to evaluate the response after first-line treatment. Evaluations of other potential disease sites were conducted, if clinically warranted, to rule out progressive disease in other areas. Women of childbearing potential were required to have a negative pregnancy (human chorionic gonadotropin) test before treatment.

Cyclophosphamide Pretreatment

A single, low, intravenous dose (300 mg/m² to a maximum of 600 mg) of cyclophosphamide was administered 3 days before immunotherapy. Previous studies demonstrated that a low dose of cyclophosphamide enhanced the effect of immunotherapy.^20-22 In various animal models, cyclophosphamide has demonstrated its ability to augment delayed-type hypersensitivity responses, increase antibody production, abrogate tolerance, and potentiate antitumor immunity.^23,24 Cyclophosphamide does not have significant antitumor activity in NSCLC, and the dose used in this setting was well below what is generally used in conventional cytotoxic chemotherapy for those cancers in which it has clinically useful cytotoxic activity.

Vaccine

L-BLP25 is a lyophilized preparation consisting of BLP25 lipopeptide, immunoadjuvant monophosphoryl lipid A, and three lipids (cholesterol, dimyristoyl phosphatidylglycerol, and dipalmitoyl phosphatidylcholine) forming a liposomal product. To enhance the antigenic stimulation of a greater number of draining lymph nodes, the vaccine was administered to four anatomic sites. It was hypothesized that this method of administration would increase the likelihood of an effective immune response against the disease. The 1,000 μg of L-BLP25 consisted of four 0.5-mL subcutaneous injections, each containing one fourth of the total dose and administered in the deltoid or triceps region of the upper arms and the left and right anterolateral aspects of the abdomen.

Study Design

Patients were randomly assigned to either L-BLP25 plus best supportive care (BSC) or BSC alone. Random assignment took place centrally once eligibility was confirmed. The randomization was stratified by disease status (stage IIIB locoregional [LR] or stage IIIB with malignant pleural effusion and stage IV). Eligible patients randomly assigned to the L-BLP25 arm received primary treatment consisting of a single low dose of cyclophosphamide 3 days before the first L-BLP25 treatment, followed by eight weekly subcutaneous vaccinations of 1,000 μg of L-BLP25 administered at weeks 0, 1, 2, 3, 4, 5, 6, and 7. At the investigator's discretion, patients in the L-BLP25 arm continued to receive maintenance vaccinations every 6 weeks starting on week 13. The vaccine could be continued after documentation of disease progression at the discretion of the investigator.

BSC was provided at the investigator's discretion to all patients. This could include psychosocial support, analgesics, and nutritional support. Second-line chemotherapy and/or palliative radiotherapy were allowed when indicated for treatment of progressive disease.

Assessment of Efficacy and Safety

Survival time is defined as the time from date of random assignment to date of death. For patients who were alive or lost to follow-up at time of analysis, the interval between date of random assignment and date on which the patient was last known alive was calculated and used as a censored observation. All patients were observed for poststudy therapies and survival. Survival was monitored at 3-month intervals for 12 months after completion of patient accrual. An additional survival follow-up (24 months after completion of accrual) was conducted to gather further information to assist in future development of the vaccine. T-cell proliferation assays were performed as a measure of immune response to MUC1 antigen before and after vaccinations.

The Functional Assessment of Cancer Therapy–Lung (FACT-L)²⁵ questionnaire was administered to all patients before study random assignment, at weeks 4 and 8, and then every 12 weeks, commencing at week 19, until withdrawal or discontinuation from the study. Data from all patients who completed the FACT-L at the pretreatment evaluation and at least one subsequent visit were included in the QOL analysis.

All patients received a treatment evaluation (physical examination, ECOG status, vital signs, treatment site inspection for the L-BLP25 arm, and adverse event [AE] assessment) and had blood samples drawn at week 8. The samples were analyzed for standard safety and immune response. In addition, patients in the L-BLP25 arm had treatment evaluation and safety and immunology blood work performed at week 4. Patients in the L-BLP25 arm had vital signs assessed and previous injection sites inspected before each L-BLP25 treatment. Vital signs were also monitored 1 hour after each L-BLP25 treatment. Patients were given diary cards after each vaccination to record any AE. Toxicity was graded according to the Cancer and Leukemia Group B Expanded Common Toxicity Criteria.

All patients who continued on maintenance treatment received treatment evaluations every 12 weeks, commencing at week 19. In addition, patients in the L-BLP25 arm had treatment evaluation and safety blood work performed at each maintenance vaccination as well as an immunology profile examination 1 week after the first maintenance vaccination. Those patients in the BSC arm who continued treatment evaluations and QOL questionnaires were considered to be on study.

Statistical Analysis

Sample size calculations required a total of 108 observed events from 150 assessable patients (75 patients in each arm) enrolled onto the study. On the basis of previous study results, the dropout rate was estimated to be approximately 10%; therefore, the total study enrollment was targeted for 166 patients. Given the assumption that 90% of patients enrolled will have stage IIIB or IV disease and 10% will have stage IIIB LR disease, the projected median survival time was 12 months for the L-BLP25 arm and 7 months for the BSC arm. The study was powered to detect a difference in survival of 5 months (hazard ratio [HR] = 0.583), with a power of 80% and a one-sided P < .025.

Safety analyses were based on the clinical and laboratory AEs experienced by patients in the study. The incidence of treatment-related AEs in both study arms was classified and summarized by MedDRA terminology (MedDRA MSSO, Reston, VA).

The efficacy analyses were performed using an intent-to-treat approach, which included data from all patients entered onto the study. The primary analysis of the efficacy variable of survival duration was performed using the Cox proportional hazard regression model, with treatment assignment, response to first-line treatment, and disease status at study entry as covariates. Test statistics were based on the regression parameter for the variable indicating treatment assignment. As a secondary analysis, a test for a treatment difference in survival duration was evaluated using the log-rank test. The P values for 12- and 24-month survival were calculated using differences in Kaplan-Meier survival estimates, with the SEs calculated using the Greenwood's formula. All reported P values are two sided.

Compliance rates for FACT-L questionnaire completion at each time point for both study arms were collected. The QOL analysis included evaluation of mean FACT-L individual change scores from baseline to week 4 and week 8, graphic representation of QOL scores over time, and area under the curve analyses for total and subscale scores. Results of the total FACT-L and Trial Outcome Index (TOI)²⁵ change scores are presented as the primary QOL analysis. The effect size between treatment arms was determined from baseline. An effect size of 0.2 to less than 0.49 indicates a small effect, a size of 0.5 to 0.79 indicates a moderate effect, and a size of 0.8 or greater indicates a large effect. The clinically meaningful change was investigated for the TOI using the cut point of 7.²⁵ Immune response was measured and evaluated in an exploratory manner.

Patient Characteristics

Between August 2000 and December 2002, 171 patients (88 in the L-BLP25 arm and 83 in the BSC arm) were enrolled at 17 investigation sites in Canada and the United Kingdom (Table 1, Fig 1). The median time from completion of primary therapy to study entry was 6.4 weeks (7.0 weeks in the BLP-25 arm and 6.1 weeks in the BSC arm). Prior chemotherapy was platinum based for all but six patients; the majority of patients received platinum-based doublets (86.3% in the BLP-25 arm and 80.7% in the BSC arm). After initial therapy, 45% of patients had stable disease, 51% had a partial response, and 4% had a complete response. Four of 171 patients were stratified incorrectly, and four of 171 patients had eligibility infractions (three patients received second-line chemotherapy before entry, and one patient was determined to have stage IIIA disease on review). All patients received the correct treatment per the randomization schedule. All patients in the L-BLP25 arm received at least five vaccinations; 96.6% of patients completed the primary phase of the treatment plan, and 69.3% continued on to the maintenance phase of the treatment plan. Second-line therapy while on study consisted mostly of chemotherapy (second or third line), radiotherapy, and surgery. During the primary treatment period of the study, five patients (9.1%) in the L-BLP25 arm and 10 patients (12%) in the BSC arm received second-line therapy. Of the 134 (78.4%) of 171 patients who continued on to the maintenance period of the study, 43 (65.2%) of 66 patients in the L-BLP25 arm and 45 (66.2%) of 68 patients in the BSC arm received second-line therapy. As of the primary analysis in March 2004, 36 patients are still actively on study (18 patients in the L-BLP25 arm and 18 patients in the BSC arm). These patients are being observed for safety, ancillary treatment receipt, and survival. Twenty-one off-study patients are alive and being observed for survival only.

Safety

Eighty-seven (98.9%) of 88 patients in the L-BLP25 arm and 79 (95.2%) of 83 patients in the BSC arm reported AEs. Most AEs were disease related and unrelated to the study drug. Grade 1 flu-like symptoms were the most common AEs related to study drug. Of patients in the L-BLP25 arm, 25 (28.4%) had AEs related to cyclophosphamide, primarily nausea. Injection site reactions (ISRs) have been reported in 45 (51.1%) of 88 patients in the L-BLP25 arm (Fig 2). All ISRs but one were grade 1.

Serious AEs were reported in 23 (26.1%) of 88 patients in the L-BLP25 arm and in 30 (36.1%) of 83 patients in the BSC arm. One severe AE (pneumonia, not otherwise specified) was classified by the investigator as possibly related to study drug. All abnormal laboratory data were considered to be consistent within the context of the patients' disease status.

Survival

Survival results indicate that the median overall survival time for patients in the L-BLP25 arm was 17.4 months compared with 13 months for patients in the BSC arm (Fig 3). The difference was not statistically significant (P = .066, unadjusted Cox). Adjusting for the stratification variables (response to first-line treatment and disease stage at study entry) reduces the survival effect for the L-BLP25 arm (adjusted HR = 0.739; 95% CI, 0.509 to 1.073; P = .112, Cox). The 2-year survival rate was 43.2% for the L-BLP25 arm v 28.9% for the BSC arm.

The greatest difference in survival was observed in stage IIIB LR patients (adjusted HR = 0.524; 95% CI, 0.261 to 1.052; P = .069, Cox; Figs 4 and 5). The observed 2-year survival rate for patients with stage IIIB LR disease was 60% for the L-BLP25 arm (median survival time not yet reached) v 36.7% for the BSC arm (median survival time, 13.3 months). The median follow-up time (from date of random assignment to time of data analysis or last contact if still alive) of patients as of the primary analysis in March 2004 is 26 months for both the L-BLP25 arm and BSC arm.

QOL

Of the 171 patients enrolled onto the study, 166 (97%) were considered eligible for inclusion in the QOL assessment (88 of 88 patients in the L-BLP25 arm and 78 of 83 patients in the BSC arm). The five patients excluded from the QOL analysis were all in the BSC arm and had completed only the baseline FACT-L. One of these patients had died 4 weeks after random assignment, and another patient died 8 weeks after random assignment. The other three patients also died on study, ranging from 14 to 38 weeks after random assignment; however, this does not explain why there were no QOL assessments completed after baseline. Overall, a clear advantage for the L-BLP25 arm compared with the BSC arm was demonstrated. A negative-change score indicates a worsening in QOL, whereas a positive-change score indicates an improvement. More patients in the L-BLP25 arm either demonstrated a clinically meaningful improvement or did not change compared with patients in the BSC arm, where more patients demonstrated a clinically meaningful worsening in the TOI (Table 2). Patients receiving L-BLP25 maintained QOL as measured by the FACT-L total score and the TOI.

Immune Response: T-Cell Proliferation

T-cell proliferation assays were performed on 78 (88.6%) of 88 patients in the L-BLP25 arm before immunization and at several time points after immunization. Sixteen patients were determined to have a positive MUC1-specific T-cell proliferative response that was induced by the L-BLP25 vaccine. Of these 16 patients, only two had stage IIIB LR disease. Because the greatest survival difference was seen in the stage IIIB LR patients, no assumptions correlating immune response with survival can be produced by this data. The median survival time of those patients on the L-BLP25 arm with a positive proliferative response was 27.6 months, whereas those patients with a negative proliferative response had a median survival time of 16.7 months.