Globally, lung cancer is the most common cause of cancer

Transcription

1 ORIGINAL ARTICLE Survival without Common Criteria Grade 3/4 for Compared with in Previously Treated Patients with Advanced Non-small Cell Lung Cancer (NSCLC): A Risk-Benefit Analysis Jean-Louis Pujol, MD,* Sofia Paul, PhD, Nadia Chouaki, MD, Patrick Peterson, PhD, Patti Moore, MS, Donald A. Berry, PhD, and Marc Salzberg, MD Background: In a recent large phase III study, previously treated patients with advanced non-small cell lung cancer who received pemetrexed demonstrated a survival time similar to patients who received docetaxel (median, 8.3 months with pemetrexed versus 7.9 months with docetaxel), with a more favorable toxicity profile, and significantly fewer Common Criteria grade 3/4 toxicities. This is a retrospective risk-benefit analysis of survival without grade 3/4 toxicity, defined as the time to the first occurrence of Common Criteria grade 3 or 4 toxicity or death, in the prospective phase III study comparing pemetrexed with docetaxel. Methods: A total of 541 patients (of 571 randomized) received either pemetrexed (500 mg/m 2 intravenously [IV]) supplemented with vitamin B 12 injections and oral folic acid or docetaxel (75 mg/m 2 IV) on day 1 of 21-day cycles. Survival without grade 3/4 toxicity was analyzed using Kaplan-Meier and Cox methods. Results: demonstrated a statistically significantly longer survival without grade 3/4 toxicity compared with docetaxel (hazard ratio 0.60, 95% confidence interval: ; p ). A supportive analysis based on selected grade 3/4 toxicities (neutropenia lasting 5 days, febrile neutropenia, infection with neutropenia, anemia, thrombocytopenia, fatigue, nausea, vomiting, diarrhea, stomatitis, and neurosensory events) also demonstrated an advantage for pemetrexed (hazard ratio 0.53; 95% confidence interval: ; p ). Conclusion: This analysis of survival without grade 3/4 toxicity suggests a benefit-to-risk profile that favors pemetrexed over docetaxel in the second-line treatment of patients with non-small cell lung cancer. *Institut Universitaire de Recherche Clinique, Montpellier, Cedex, France; Eli Lilly and Company, Indianapolis, Indiana (during the conduct of the study) and Novartis, East Hanover, New Jersey (currently); Lilly France, Suresnes, Cedex, France; Eli Lilly and Company, Indianapolis, Indiana; M.D. Anderson Cancer Center, Houston, Texas; University Hospital, Basel, Switzerland. Disclosure: This study was funded by Eli Lilly and Company. Address for correspondence: Jean-Louis Pujol, MD; Departement de Biostatistiques, Epidemiologie et Recherche Clinique, Institut Universitaire de Recherche Clinique, Rue de la Cardonille, Montpellier, Cedex 5, France. jl-pujol@chu-montpellier.fr Copyright 2007 by the International Association for the Study of Lung Cancer ISSN: /07/ Key Words:, Survival,, Non-small cell lung cancer. (J Thorac Oncol. 2007;2: ) Globally, lung cancer is the most common cause of cancer deaths. Non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases. 1 Most patients with NSCLC present with advanced, inoperable disease (stage IIIB or IV). 2 Combination chemotherapy is recommended for patients with a good performance status, yet the majority of patients will develop disease progression after first-line therapy. 2,3 In the past, best supportive care was given to patients with relapsed advanced NSCLC for symptom palliation. Currently, however, there are second-line chemotherapeutic agents that have demonstrated efficacy, including improved survival, in patients with a good performance status. Two such agents are pemetrexed and docetaxel. has shown improved survival compared with best supportive care, 4,5 whereas pemetrexed has shown efficacy comparable with that of docetaxel with a more favorable toxicity profile in a head-to-head phase III comparison. 6 Both drugs have been approved as single agents for second-line treatment of advanced NSCLC. is also approved for the treatment of malignant pleural mesothelioma in combination with cisplatin. 7, a novel multitargeted antifolate, targets several folate-dependent enzymatic pathways and inhibits multiple enzymes involved in purine and pyrimidine synthesis Conventional cytotoxic agents that amplify targeted pathways are often associated with reduced efficacy. The mechanisms of action of pemetrexed are thought to explain its greater potency and broader spectrum of antitumor activity observed preclinically relative to other antimetabolites such as 5-fluorouracil, methotrexate, or raltitrexed. 11 In a large phase III study of second-line treatment for advanced NSCLC, 571 previously treated patients were randomized to pemetrexed or docetaxel; those treated with pemetrexed demonstrated a similar survival time (median of 8.3 months with pemetrexed versus 7.9 months with do- Journal of Thoracic Oncology Volume 2, Number 5, May

2 Pujol et al. Journal of Thoracic Oncology Volume 2, Number 5, May 2007 cetaxel; hazard ratio [HR] 0.99) and progression-free survival time (median of 2.9 months for both arms; HR 0.97) compared with those treated with docetaxel. 6 However, treatment with pemetrexed was associated with a more favorable toxicity profile, as demonstrated by significantly fewer grade 3 and 4 toxicities and hospitalizations related to adverse events and fewer toxic deaths. The results of the standard efficacy and safety measures such as overall survival, progression-free survival, time to disease progression, time to treatment failure, response rate, toxicity, and quality of life (using the Lung Cancer Symptom Scale) have been published. 6 In a study for which there is neither statistically different efficacy between therapies nor even a numerical trend toward a difference, it can be difficult to evaluate the risks and benefits of the therapies. Moreover, a treatment that can reduce the risk of toxicity is important, especially in a palliative setting. The retrospective analysis presented here compares the time from randomization to the first occurrence of grade 3 or grade 4 toxicity or death in the prospective phase III study of pemetrexed and docetaxel. 6 We considered this analysis a measure of benefit relative to risk with pemetrexed compared with docetaxel in that overall survival time (i.e., clinical benefit) relative to the first occurrence of grade 3/4 toxicity (i.e., clinical risk) was compared. The risk of all grade 3/4 toxicity was considered in the interest of avoiding any potential bias that may result from attempting to select or weight specific toxicities. However, because not all grade 3/4 toxicities have the same clinical implications, we conducted an additional analysis including only specific grade 3/4 toxicity that we regarded as the most clinically important. PATIENTS AND METHODS The study on which this analysis is based 6 included patients 18 years and older with measurable or assessable stage III or IV NSCLC and good performance status (Eastern Cooperative Oncology Group performance status 0 2). 12 Patients were previously treated with only one regimen for metastatic disease, but were allowed one additional previous regimen for neoadjuvant, adjuvant, or neoadjuvant plus adjuvant treatment. Patients with symptomatic brain metastasis, grade 3 or 4 peripheral neuropathy, weight loss of 10% or more during the previous 6 weeks, uncontrolled pleural effusions, previous docetaxel or pemetrexed therapy, or an inability to interrupt nonsteroidal anti-inflammatory drugs were excluded from the study. Patients were randomized to receive either pemetrexed (500 mg/m 2 as a 10-minute intravenous [IV] infusion) plus vitamin B 12 injections and oral folic acid supplementation or docetaxel (75 mg/m 2 as a 1-hour IV infusion) on day 1 of 21-day cycles. Randomized patients were stratified according to performance status (0/1 versus 2), disease stage (III versus IV), number of previous chemotherapy regimens (one or two), time since last chemotherapy ( 3 versus 3 months), best response to last chemotherapy (objective tumor response/stable disease versus progressive disease/unknown), baseline plasma homocysteine level ( 12 mol/liter versus 12 mol/liter), previous platinum-containing chemotherapy (yes versus no), previous paclitaxel-containing chemotherapy (yes versus no), and center. Cycles were repeated until unacceptable toxicity or disease progression occurred or until the patient or the investigator requested therapy discontinuation. Dexamethasone 4 mg orally twice daily (the day before, the day of, and the day after treatment) was provided prophylactically to patients receiving pemetrexed for the prevention of skin rash. Dexamethasone 8 mg twice daily (the day before, the day of, and 2 days after treatment) was provided prophylactically to patients receiving docetaxel for the prevention of hypersensitivity reactions and fluid retention. Granulocyte colony stimulating factor support was not provided routinely, but was allowed to treat neutropenia or as prophylaxis in patients who had neutropenia in a previous cycle. Patients received comprehensive baseline assessments including laboratory tests and imaging. Thereafter, hematologic laboratory values were evaluated weekly, with chemistry laboratory values evaluated after days 1 and 8 of each cycle. Tumor measurements were assessed using Southwest Oncology Group criteria after every other cycle. Lung Cancer Symptom Scale evaluations were conducted weekly. Toxicities were rated before each cycle using the National Cancer Institute Common Criteria (CTC), version 2. In the prospective analyses of the large phase III study of pemetrexed and docetaxel, 6 comparisons of tumor response rates and toxicity event rates between arms were performed using Fisher s exact test. The Kaplan-Meier 13 and Cox 14 methods were used for time-to-event analyses. Analysis of Survival without Grade 3/4 Survival without grade 3/4 toxicity was defined as the time from the date of randomization to the first date of any grade 3 or 4 toxicity (regardless of drug causality) or death due to any cause. Times were censored at the date of the last follow-up visit for patients who were still alive and who did not have any grade 3/4 toxicity. Times for all 541 randomized patients who received study drug were analyzed using the Kaplan-Meier method 13 ; treatment-arm comparisons were based on a Cox estimate 14 for the HR. This analysis was repeated using a stricter definition of risk, which included only the most clinically important grade 3/4 toxicities: neutropenia lasting longer than 5 days, febrile neutropenia, documented infections related to neutropenia, anemia, thrombocytopenia, fatigue, nausea, vomiting, diarrhea, stomatitis, and neurosensory events. Survival without these selected, clinically important grade 3/4 toxicities was defined as the time from the date of randomization to the first date of any selected grade 3 or 4 toxicity (regardless of drug causality) or death due to any cause. Censoring and analytical methods were applied in the same manner as in the analysis of all grade 3/4 toxicity. RESULTS Patient Characteristics and Dose Administration Of the 571 randomized patients, 541 received either pemetrexed (n 265) or docetaxel (n 276) between March 2001 and February 2002 and were included in both analyses 398 Copyright 2007 by the International Association for the Study of Lung Cancer

3 Journal of Thoracic Oncology Volume 2, Number 5, May 2007 RRH TABLE 1. Patient Characteristics (n 571 a ) (n 283) (n 288) TABLE 2. NCI-CTC Hematologic Toxicities, Regardless of Drug Causality % of Patients with Grade 3/4 Median age, yr (range) 59 (22 81) 57 (28 87) Gender, % Female Male ECOG PS 0 or 1, % Stage IV disease, % Homocysteine levels 12 mol/liter, % Histology, % Adenocarcinoma Squamous cell carcinoma Best response to previous chemotherapy, % CR/PR Time since last chemotherapy, % 3 mo mo Previous therapy with paclitaxel, % Previous therapy with platinum, % a All randomized patients. of survival without grade 3/4 toxicity (all grade 3/4 and selected grade 3/4 toxicity). Baseline characteristics were similar and well balanced across treatment arms, with no statistically significant differences between arms (Table 1). 6 Patients received a median of four cycles of pemetrexed (range, 1 20 cycles) and a median of four cycles of docetaxel (range, 1 14 cycles), with relative dose intensities of 96.6% and 94.4%, respectively. There was no significant difference between arms in the number of dose delays (19.8% versus 17.8%); however, there were significantly fewer dose reductions for the patients treated with pemetrexed (1.2% versus 5.6%, p 0.001). Efficacy Efficacy measures were similar for each arm, with no statistically significant differences found between pemetrexed and docetaxel in overall response or stable disease rates, progression-free survival time, time to disease progression, time to treatment failure, time to response, duration of response, and duration of stable disease. 6 The overall survival time was also comparable (a median of 8.3 months with pemetrexed versus 7.9 months with docetaxel; HR 0.99), with a 1-year overall survival rate of 29.7% for each arm. Although pemetrexed and docetaxel were found to be similar with respect to all major efficacy endpoints, the toxicity and safety profiles of pemetrexed were found to be significantly better than those of docetaxel (Tables 2 and 3). Patients treated with pemetrexed had statistically significantly fewer CTC grade 3/4 toxicities (p 0.001), with lower rates of neutropenia (5.3% versus 40.2%, p 0.001), febrile neutropenia (1.9% versus 12.7%, p.001), and infection associated with grade 3/4 neutropenia (0% versus 5.8%, p (n 265) (n 276) p a Neutropenia Febrile neutropenia Infection with grade 3/4 neutropenia Anemia Thrombocytopenia NCI-CTC, National Cancer Institute Common Criteria. a p value calculated using Fisher s exact test. TABLE 3. NCI-CTC Nonhematologic Toxicities, Regardless of Drug Causality % of Patients with Grade 3/4 (n 265) (n 276) p a Alopecia b ALT Neurosensory Fatigue Nausea Diarrhea Vomiting NS Stomatitis NS NCI-CTC, National Cancer Institute Common Criteria; ALT, alanine transaminase. a p value calculated using Fisher s exact test. b All grades ). Patients receiving pemetrexed also had significantly less grade 3/4 diarrhea (0.4% versus 4.0%, p 0.006) and alopecia (all grades) (11.3% versus 42.4%, p 0.001), but had a greater number of reversible grade 3/4 alanine transaminase elevations (2.6% versus 0.4%, p 0.034). Significantly more patients treated with docetaxel required hospitalization due to any drug-related event than patients treated with pemetrexed (21.7% versus 7.2%, p 0.001); of note, hospitalization for febrile neutropenia was also higher for patients treated with docetaxel (13.4% versus 1.5%, p 0.001). Patients treated with docetaxel also required significantly more supportive care measures than those treated with pemetrexed, with 19.2% versus 2.6% of patients (p 0.001) requiring the use of granulocyte colony stimulating factor. In addition, there were more drug-related deaths (five versus three) and drug-related serious adverse events on the docetaxel arm compared with the pemetrexed arm (23.9% versus 10.2%, p 0.001). There were no statistically significant differences between treatment arms in the occurrence of anemia, thrombocytopenia, most of the nonhematologic toxicities, red blood cell transfusions, and the use of erythropoietin. Copyright 2007 by the International Association for the Study of Lung Cancer 399

4 Pujol et al. Journal of Thoracic Oncology Volume 2, Number 5, May 2007 FIGURE 1. Survival without the first occurrence of all grade 3/4 toxicity. is solid line; docetaxel is dashed line. FIGURE 2. Survival without the first occurrence of selected grade 3/4 toxicity: neutropenia lasting longer than 5 days, febrile neutropenia, documented infections related to neutropenia, anemia, thrombocytopenia, fatigue, nausea, vomiting, diarrhea, stomatitis, and neurosensory events. is solid line; docetaxel is dashed line. Survival without Grade 3/4 In the analysis of all grade 3/4 toxicity, patients randomized to the pemetrexed arm had statistically significantly longer survival time without grade 3/4 toxicity or death (HR 0.60; 95% CI: ; p ). The median survival time without grade 3/4 toxicity for patients treated with pemetrexed was 1.2 months versus 0.4 months for patients treated with docetaxel. The 6-month rate of survival without grade 3/4 toxicity was 25.7% for pemetrexed compared with 10.1% for docetaxel; the respective 12-month rates of survival without grade 3/4 toxicity were 12.2% and 6.1%. The Kaplan-Meier results are illustrated in Figure Copyright 2007 by the International Association for the Study of Lung Cancer

5 Journal of Thoracic Oncology Volume 2, Number 5, May 2007 RRH In the additional analysis that considered only selected grade 3/4 toxicity, significantly longer survival time without grade 3/4 toxicity or death was again observed for the pemetrexed arm (HR 0.53; 95% CI: ; p ). The median survival time without selected grade 3/4 toxicity for patients treated with pemetrexed was 5.8 months versus 1.2 months for patients treated with docetaxel. The 6-month rate of survival without selected grade 3/4 toxicity was 49.0% for pemetrexed compared with 22.1% for docetaxel; the respective 12-month rates of survival without selected grade 3/4 toxicity were 23.8% and 12.2%. The Kaplan-Meier results are illustrated in Figure 2. DISCUSSION The previously published results of this large phase III randomized study indicate that pemetrexed and docetaxel have similar efficacy. 6 They also describe a more favorable safety profile for pemetrexed. This retrospective analysis of these data 6 further characterizes the benefit-to-risk profile of pemetrexed compared with docetaxel. By comparing these chemotherapies in terms of the time to the first occurrence of grade 3/4 toxicity or death, we are essentially comparing overall survival time (i.e., clinical benefit) relative to the first occurrence of CTC grade 3/4 toxicity (i.e., clinical risk), thus providing a measure of benefit relative to risk. The results of this analysis are especially important for the treatment of relapsed NSCLC for which therapy remains primarily palliative; as such, the ultimate goal is to improve survival with minimal toxicity. Although other definitions of benefit and risk may be appropriate, our definition allows for minimal subjective bias (by including all CTC grade 3/4 toxicity) and can be easily understood and interpreted in terms of standard methods of statistical analyses. Survival without any type of grade 3/4 toxicity was significantly longer for patients receiving pemetrexed versus docetaxel (HR 0.60; 95% CI: ; p ). Based on this analysis, the clinical benefit (survival time) relative to the clinical risk (the first occurrence of any grade 3/4 toxicity) was significantly improved for patients treated with pemetrexed compared with docetaxel. When the analysis was based on the first occurrence of selected grade 3/4 toxicity, a more narrowly defined risk, the result was a statistically significant advantage and an improved benefitto-risk profile for pemetrexed (HR 0.53; 95% CI: ; p ). The results of these two analyses are robust in that there is a strong and consistent separation between treatments for the full patient population across the entire study timeline. For both analyses, median survival without grade 3/4 toxicity was at least threefold higher for patients treated with pemetrexed than the median for docetaxel, and the rates of survival without grade 3/4 toxicity for pemetrexed at 6 and 12 months were at least twofold higher than the respective rates for docetaxel. These results help to further characterize pemetrexed as an effective treatment compared with docetaxel in patients with advanced NSCLC who have received previous chemotherapy. In the current study, we proposed analyses that incorporate two important endpoints, survival and toxicity (grade 3/4), into a single measure of benefit relative to risk that we regard as having clinical relevance. These results help to further characterize the benefit-to-risk profile of pemetrexed as a treatment option for patients with advanced NSCLC who have received previous chemotherapy. ACKNOWLEDGMENTS The authors wish to thank Jeffery Vest for statistical support, Noelle Gasco for editorial support, and Donna Miller for technical support of this manuscript. REFERENCES 1. Shibuya K, Mathers CD, Boschi-Pinto C, et al. Global and regional estimates of cancer mortality and incidence by site: II. Results for the global burden of disease BMC Cancer 2002;2: Martins R, Kelly K, Socinski MA, et al. Management of metastatic non-small cell lung cancer in 2006: innovations and unique clinical scenarios. In Govindan R, Perry MC, Slowinski FH. (Eds). American Society of Clinical Oncology 2006 Educational Book. Alexandria, VA: American Society of Clinical Oncology; 2006: Pfister DG, Johnson DH, Azzoli CG, et al. American Society of Clinical Oncology treatment of unresectable non-small-cell lung cancer guideline: update J Clin Oncol 2004;22: Shepherd FA, Dancey J, Ramlau R, et al. Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 2000;18: Fossella FV, Lynch T, Shepherd FA. Second line chemotherapy for NSCLC: establishing a gold standard. Lung Cancer 2002;38(Suppl 4): Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of pemetrexed versus docetaxel in patients with non-small-cell lung cancer previously treated with chemotherapy. J Clin Oncol 2004;22: U.S. Food and Drug Administration, Department of Health and Human Services, Center for Drug Evaluation and Research. Alimta (pemetrexed for injection). February 5, Available at: drug/infopage/alimta/. Accessed November 10, Shih C, Habeck LL, Mendelsohn LG, et al. Multiple folate enzyme inhibition: mechanism of a novel pyrrolopyrimidine-based antifolate LY (MTA). Adv Enzyme Regul 1998;38: Taylor EC, Kuhnt D, Shih C, et al. A dideazatetrahydrofolate analogue lacking a chiral center at C-6, N-[4-[2-(2-amino-3,4-dihydro-4-oxo-7Hpyrrolo[2,3-d]pyrimidin-5-yl)ethyl]benzoyl]-L-glutamic acid, is an inhibitor of thymidylate synthase. J Med Chem 1992;35: Schultz RM, Patel VF, Worzalla JF, et al. Role of thymidylate synthase in the antitumor activity of the multitargeted antifolate, LY Anticancer Res 1999;19: Chen VJ, Bewley JR, Andis SL, et al. Preclinical cellular pharmacology of LY (MTA): a comparison with methotrexate, LY and raltitrexed for their effects on intracellular folate and nucleoside triphosphate pools in CCRF-CEM cells. Br J Cancer 1998;78(Suppl 3): Oken MM, Creech RH, Tormey DC, et al. and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982;5: Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53: Cox DR. Regression models and life tables (with discussion). J R Stat Soc B 1972;34: Copyright 2007 by the International Association for the Study of Lung Cancer 401