Published Ahead of Print on March 28, 2011 as /JCO J Clin Oncol by American Society of Clinical Oncology INTRODUCTION

Transcription

1 Published Ahead of Print on March 28, 2011 as /JCO The latest version is at JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T Phase III, Open-Label, Randomized Study Comparing Concurrent Gemcitabine Plus Cisplatin and Radiation Followed by Adjuvant Gemcitabine and Cisplatin Versus Concurrent Cisplatin and Radiation in Patients With Stage IIB to IVA Carcinoma of the Cervix Alfonso Dueñas-González, Juan J. Zarbá, Firuza Patel, Juan C. Alcedo, Semir Beslija, Luis Casanova, Pittayapoom Pattaranutaporn, Shahid Hameed, Julie M. Blair, Helen Barraclough, and Mauro Orlando See accompanying editorial doi: /JCO From the National Cancer Institute/Institute of Biomedical Research, Universidad Nacional AutónomadeMéxico, México City, México; Medical Center, San Roque, Tucumán; Eli Lilly Interamerica, Buenos Aires, Argentina; Post Graduate Institute of Medical Education and Research, Chandigarh, India; National Institute of Oncology, Panamá, Panamá; Institute of Oncology, Clinical Centre of Sarajevo University, Sarajevo, Bosnia and Herzegovina; Instituto Nacional de Enfermedades Neoplasticas, Lima, Peru; Siriraj Hospital, Mahidol University, Bangkok, Thailand; Shaukat Khanum Memorial Cancer Hospital and Research Centre, Lahore, Pakistan; and Eli Lilly, Sydney, Australia. Submitted October 1, 2009; accepted January 31, 2011; published online ahead of print at on March 28, Supported by Eli Lilly. Presented in part on the clinical trial registry located at ClinicalTrials.gov (identification No. NCT ), on the Lilly Clinical Trial Registry ( trial identification No. 4015), and at the 45th Annual Meeting of the American Society of Clinical Oncology, May 29-June 2, 2009, Orlando, FL. A B S T R A C T Purpose To determine whether addition of gemcitabine to concurrent cisplatin chemoradiotherapy and as adjuvant chemotherapy with cisplatin improves progression-free survival (PFS) at 3 years compared with current standard of care in locally advanced cervical cancer. Patients and Methods Eligible chemotherapy- and radiotherapy-naive patients with stage IIB to IVA disease and Karnofsky performance score 70 were randomly assigned to arm A (cisplatin 40 mg/m 2 and gemcitabine 125 mg/m 2 weekly for 6 weeks with concurrent external-beam radiotherapy [XRT] 50.4 Gy in 28 fractions, followed by brachytherapy [BCT] 30 to 35 Gy in 96 hours, and then two adjuvant 21-day cycles of cisplatin, 50 mg/m 2 on day 1, plus gemcitabine, 1,000 mg/m 2 on days 1 and 8) or to arm B (cisplatin and concurrent XRT followed by BCT only; dosing same as for arm A). Results Between May 2002 and March 2004, 515 patients were enrolled (arm A, n 259; arm B, n 256). PFS at 3 years was significantly improved in arm A versus arm B (74.4% v 65.0%, respectively; P.029), as were overall PFS (log-rank P.0227; hazard ratio [HR], 0.68; 95% CI, 0.49 to 0.95), overall survival (log-rank P.0224; HR, 0.68; 95% CI, 0.49 to 0.95), and time to progressive disease (log-rank P.0012; HR, 0.54; 95% CI, 0.37 to 0.79). Grade 3 and 4 toxicities were more frequent in arm A than in arm B (86.5% v 46.3%, respectively; P.001), including two deaths possibly related to treatment toxicity in arm A. Conclusion Gemcitabine plus cisplatin chemoradiotherapy followed by BCT and adjuvant gemcitabine/cisplatin chemotherapy improved survival outcomes with increased but clinically manageable toxicity when compared with standard treatment. J Clin Oncol by American Society of Clinical Oncology Authors disclosures of potential conflicts of interest and author contributions are found at the end of this article. Clinical Trials repository link available on JCO.org. Corresponding author: Mauro Orlando, MD, Eli Lilly Interamerica, Tronador 4890, Piso 12, Buenos Aires 1430, Argentina; mauro@lilly.com by American Society of Clinical Oncology X/11/2999-1/$20.00 DOI: /JCO INTRODUCTION Cervical carcinoma is the second leading cause of cancer-related death among women worldwide, with an estimated 273,505 deaths from cervical cancer in mainly as a result of patients in the developing world presenting with locally advanced cervical cancer (LACC). 2 Five randomized clinical trials showed a 30% to 50% decrease in risk of death in LACC with addition of concurrent platinumbased chemoradiotherapy when compared with radiotherapy alone. 3-7 Mechanisms to account for improved efficacy are thought to include direct cytotoxicity, radiosensitization of tumor cells, and control of subclinical metastases. 8 Gemcitabine synergizes with radiation and cisplatin, as shown in preclinical and clinical studies. 9,10 A phase I/II trial in patients with International Federation of Gynecology and Obstetrics stage IIA to IVA cervical cancer established gemcitabine 125 mg/m 2 as the dose for combination with the standard regimen of weekly cisplatin 40 mg/m 2 and radiation. 11 A randomized phase II trial comparing this regimen with cisplatin, both 2011 by American Society of Clinical Oncology 1 Copyright 2011 by American Society of Clinical Oncology

2 Dueñas-González et al arms concurrent with external-beam radiation (XRT) followed by radical hysterectomy, showed a statistically significant difference in pathologic complete response rates, a potential surrogate marker of survival. 12 Adjuvant chemotherapy may be beneficial because 35% of patients experience disease progression after chemoradiotherapy. 13 Although a 54% reduction in the risk of death was observed when concurrent chemoradiotherapy was followed by adjuvant chemotherapy, 8,13 one trial was performed in patients with postoperative earlystage disease, 7 sample size was small in both trials, and neither addressed the specific contribution of adjuvant treatment. This multicenter trial aimed to improve survival outcomes for patients with LACC, using the synergy between gemcitabine, cisplatin, and XRT and the potential value of adjuvant chemotherapy. The primary objective was to compare progression-free survival (PFS) at 3 years of patients with LACC treated with gemcitabine plus cisplatin and XRT followed by brachytherapy (BCT) and then two additional 21-day cycles of adjuvant gemcitabine plus cisplatin (arm A) versus patients treated with cisplatin and XRT followed by BCT only (arm B). The primary objective was changed from overall survival (OS) to PFS at 3 years in March 2007, after enrollment was completed (further details regarding this change can be found in the Data Supplement). The secondary objectives were to compare OS, time to progressive disease (TtPD), local failure rate (LFR), tumor response rate (TRR), and toxicity between treatment arms. PATIENTS AND METHODS Eligibility Criteria Inclusion criteria for Study B9E-MC-JHQS included a histologic diagnosis of International Federation of Gynecology and Obstetrics stage IIB to IVA cervical cancer 14 with at least one measurable lesion and Karnofsky performance score of 70. Patients were excluded if there was evidence of cancer on fine-needle aspiration of an enlarged ( 1 cm) para-aortic lymph node or if they had previously received chemotherapy or radiotherapy. A more comprehensive list of inclusion and exclusion criteria is available in the Data Supplement. The study was conducted in Argentina, Bosnia and Herzegovina, India, Mexico, Pakistan, Panama, Peru, and Thailand according to International Conference on Harmonisation Good Clinical Practice Guidelines, including obtaining written informed consent from all patients, and was approved by local ethics committees. Treatment Schedule The study CONSORT diagram is presented in Figure 1. Patients were randomly assigned to arm A or arm B at week 0 (Fig 2) using the Pocock and Assessed for eligibility (N = 617) Excluded (n = 102) Screen failure (n = 102) Random assignment (n = 515) Allocated to Arm A (n = 259) Received allocated intervention (n = 260) Did not receive allocated intervention (n = 0) Number missing (n = 0) Discontinued intervention (n = 42) Lost to follow-up/unable to contact patient (n = 2) Adverse event (n = 18) Personal conflict/patient decision (n = 9) Protocol entry criteria not met (n = 2) Clinical relapse, PD after confirmed CR/PR (n = 2) Lack of efficacy, PD Lack of efficacy, SD Physician decision (n = 3) Protocol violation Death from study treatment toxicity (n = 2) Death from other causes Allocated to Arm B (n = 256) Received allocated intervention (n = 255) Did not receive allocated intervention Number missing Discontinued intervention (n = 11) Adverse event Patient moved away Personal conflict/patient decision (n = 3) Protocol entry criteria not met Clinical relapse, PD after confirmed CR/PR Lack of efficacy, PD (n = 2) Physician decision Protocol violation Fig 1. CONSORT diagram of study design. Arm A treatment consisted of gemcitabine plus cisplatin chemoradiotherapy followed by brachytherapy and adjuvant gemcitabine and cisplatin. Arm B treatment consisted of cisplatin chemoradiotherapy followed by brachytherapy only. PD, progressive disease; CR, complete response; PR, partial response; SD, stable disease; ITT, intent to treat; LTFU, lost to follow-up. Efficacy analysis Analyzed (ITT; n = 259) Excluded from analysis (n = 0) Safety analysis Analyzed (safety; n = 260) Excluded from analysis (n = 0) Efficacy analysis Analyzed (ITT; n = 256) Excluded from analysis (n = 0) Safety analysis Analyzed (safety; n = 255) Excluded from analysis (n = 0) LTFU by end of study (n = 31) LTFU by end of study (n = 25) by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

3 Gemcitabine and Cisplatin Chemoradiation in Cervical Carcinoma R Week 0 Arm A (n = 259) Arm B (n = 256) Chemoradiotherapy 6 Preadjuvant phase Simon algorithm 15 according to disease stage (IIB v III to IVA), tumor diameter ( v 5 cm), investigational center, radiation equipment (cobalt-60 v linear accelerator), and age ( v 55 years). The random allocation sequence was generated by central telephone (Eli Lilly, Indianapolis, IN) and was concealed until interventions were assigned. All patients received cisplatin 40 mg/m 2 by intravenous infusion over 60 minutes, followed immediately in arm A by gemcitabine 125 mg/m 2 administered by intravenous infusion over 30 to 60 minutes once weekly for 6 weeks. Both drugs were administered 1 to 2 hours before XRT, for which patients received 50.4 Gy of XRT to the entire pelvic region in 28 fractions of 1.8 Gy/d, 5 days a week, over the 6 weeks of chemotherapy. XRT was administered using a standard four-field box technique (anteroposterior, posteroanterior, and two lateral) for energy sources of 6 MV or using two parallel opposing fields (anteroposterior and posteroanterior) for energy sources of less than 6 MV. Further details of radiation field margins are provided in the Data Supplement. Local quality assurance of radiation procedures was performed by a radiation oncologist at each site; there was no required centralized review of the treatment plans. Immediately after completion of the chemoradiotherapy schedule, patients underwent low- or intermediate-dose rate BCT with cesium-137, using standard Fletcher-Suit applicators or their modification. A BCT dose of 30 to 35 Gy was delivered to point A to result in a cumulative dose of 80 to 85 Gy combining XRT and BCT. The cumulative XRT and BCT dose to point B (the pelvic wall) was 55 to 65 Gy. After BCT and 2 weeks of rest, arm A patients received two consecutive 21-day cycles of adjuvant chemotherapy with cisplatin 50 mg/m 2 on day 1 and gemcitabine 1,000 mg/m 2 on days 1 and 8. Baseline and Treatment Assessments At baseline, all patients underwent a physical examination, chest x-ray, computed tomography measurement of existing lesions, and clinical laboratory tests. During study treatment, all patients underwent weekly hematology and blood chemistry testing for safety and dose adjustments and physical examinations, including assessment of tumor response by physical examination during the chemoradiotherapy phase (all patients) and on day 1 of each adjuvant chemotherapy cycle (arm A only). After BCT, all patients underwent abdominal and pelvic computed tomography scans. In the poststudy follow-up phase, all patients had evaluations for efficacy and safety end points approximately 30 days after completion of either adjuvant chemotherapy for arm A or BCT for arm B, and then approximately every 4 months for No adjuvant chemotherapy BCT Rest Adjuvant chemotherapy Adjuvant phase Fig 2. Study treatment schedule. Arm A treatment consisted of gemcitabine plus cisplatin chemoradiotherapy for 6 weeks. Arm B treatment consisted of cisplatin chemoradiotherapy for 6 weeks. All patients received 28 fractions of 1.8 Gy per day, 5 days per week, over the 6 weeks of chemoradiotherapy. After chemoradiotherapy, all patients were scheduled to receive 30 to 35 Gy of brachytherapy (BCT) in week 7. After BCT and a subsequent 2-week rest period, patients randomly assigned to arm A received adjuvant chemotherapy (cisplatin 50 mg/m 2 on day 1 plus gemcitabine 1,000 mg/m 2 on days 1 and 8, every 3 weeks for two cycles). R, random assignment. 15 months. Survival, disease progression, and post-therapy treatment were monitored every 6 months up to 48 months thereafter until death or study end. The primary end point of the study was PFS at 3 years. The secondary efficacy measures were OS, TtPD, TRR, and LFR. Definitions of survival outcomes and tumor response are provided in the Data Supplement. Safety was evaluated in all patients who had received one or more doses of study drug according to the treatment received (safety population) by recording clinical adverse events (AEs) using National Cancer Institute Common Toxicity Criteria (version 2.0) laboratory and nonlaboratory toxicity grading for acute toxicity and Radiotherapy Oncology Group/European Organisation for Research and Treatment of Cancer for late toxicity. Statistical Methods PFS at 3 years was chosen as the new primary end point because PFS is not confounded by postdiscontinuation therapy and the 3-year fixed time point did not require a prespecified number of events for analysis, as described in the Data Supplement. The sample size of approximately 500 evaluable patients was based on the original OS primary end point; 436 deaths would have provided 80% power at a two-sided.05 level to show superiority of arm A over arm B, assuming a true hazard ratio (HR) of Sample size was not adjusted after amendment of the primary end point to PFS at 3 years because the protocol was amended 3 years after the last patient was enrolled. However, the study retained 80% power to detect a significant treatment difference between the arms in PFS at a two-sided.05 level, assuming 100 to 150 events would occur after all patients had completed at least 3 years of follow-up and a true PFS HR of 0.56 to PFS at 3 years was estimated in each arm using the Kaplan-Meier method and compared between arms using a Z statistic. Time-to-event end points were compared between treatment arms using the log-rank test. Supporting analyses were carried out using the Kaplan-Meier method to estimate median survival time and survival probability at fixed time points. A univariate Cox proportional hazards model was used to estimate the unadjusted HR. LFR, TRR, and all safety end points were compared between treatment arms using Fisher s exact test. Median follow-up time was calculated using the reverse Kaplan-Meier method. 16 All efficacy and safety analyses were carried out on the intent-to-treat and safety populations, respectively (Fig 1). All tests were carried out at a two-sided.05 level. RESULTS Patient Disposition From May 2002 to March 2004, 515 women age 18 to 70 years were enrolled and randomly assigned to arm A (n 259) or arm B (n 256); these patients composed the intent-to-treat population. The follow-up phase lasted until April 2008; median follow-up time was 46.9 months. One patient assigned to arm B received the arm A regimen; therefore, the safety population comprised 260 patients in arm A and 255 patients in arm B (Fig 1). Clinical characteristics were well balanced at baseline between treatment arms (Table 1). Efficacy There was statistically significant improvement for arm A over arm B in the three main survival end points of PFS at 3 years, overall PFS, and OS. PFS at 3 years was 74.4% (95% CI, 68.0% to 79.8%) in arm A and 65.0% (95% CI, 58.5% to 70.7%) in arm B (P.029). Arm A was superior compared with arm B for overall PFS (log-rank P.0227; HR, 0.68; 95% CI, 0.49 to 0.95; Fig 3A), OS (log-rank P.0224; HR, 0.68; 95% CI, 0.49 to 0.95; Fig 3B), and TtPD (log-rank P.0012; HR, 0.54; 95% CI, 0.37 to 0.79; Data Supplement). TRR was 95.8% (95% CI, 92.5% to 97.9%) in arm A and 93.4% (95% CI, 89.6% to 96.1%) in arm B, but the between-treatment difference was not statistically significant (P.249; data not shown). Sensitivity analyses of PFS and by American Society of Clinical Oncology 3

4 Dueñas-González et al Characteristic Table 1. Baseline Patient Demographics and Clinical Characteristics for All Randomly Assigned Patients ArmA(n 259) ArmB(n 256) All Patients (N 515) Age, years Median Range KPS Median Range Pathological diagnosis Adenocarcinoma Nonadenocarcinoma Disease stage IIB IIIA IIIB IVA Maximum diameter of the largest bidimensional lesion, cm Median Range Hemoglobin at baseline, g/dl Median Range Had 1 enlarged PA LN ( 1 cm) Radiation equipment Co Linear accelerator Race/ethnicity Western Asian White East/Southeast Asian Hispanic Abbreviations: KPS, Karnofsky performance score; PA, para-aortic; LN, lymph node; Co60, cobalt-60. Therapy in arm A was as follows: chemoradiotherapy, cisplatin 40 mg/m 2 followed by gemcitabine 125 mg/m 2 on day 1 of each week for 6 weeks; and adjuvant chemotherapy, cisplatin 50 mg/m 2 on day 1 followed by gemcitabine 1,000 mg/m 2 on day 1 and gemcitabine 1,000 mg/m 2 on day 8 of each of two 21-day cycles. Therapy in arm B was as follows: chemoradiotherapy, cisplatin 40 mg/m 2 on day 1 of each week for 6 weeks; and no adjuvant chemotherapy. Nonadenocarcinoma included squamous cell/poorly differentiated carcinoma and adeno/squamous carcinoma. TtPD using different end point definitions and patient populations supported these results (data not shown). LFR was decreased in arm A (11.2%) versus arm B (16.4%), but this difference was not statistically significant (P.097). Exploratory analyses showed that the distant failure rate was statistically significantly lower in arm A (8.1%) than in arm B (16.4%; P.005; Data Supplement). Two patients had local and distant sites of progressive disease, and one patient had unknown site of progressive disease; results were similar when these patients were classified as having local or distant progressive disease (data not shown). Treatment and Compliance Table 2 lists chemotherapy and radiation treatment. During chemoradiotherapy, the median number of gemcitabine and cisplatin doses in arm A was five doses (range, one to six doses); the median number of cisplatin doses in arm B was six doses (range, one to six doses). In arm A only, 224 patients (86.2%) received at least one cycle of adjuvant chemotherapy, and 199 patients (76.5%) received both cycles. The median XRT dose was 50.4 Gy in both arms (arm A range, 1.8 to 70.4 Gy; arm B range, 1.8 to 72.0 Gy). More than 90% of patients in each arm received BCT; the median dose administered to point A was 30 Gy (arm A interquartile range [IQ] range, 28.2 to 32.0 Gy; arm B IQ range, 28.5 to 31.6 Gy). The median doses were similar in both treatment arms for XRT and BCT, but radiotherapy duration was longer for arm A (49 days; IQ range, 44 to 55 days) than for arm B (45 days; IQ range, 42 to 50 days; P.001). AEs Three deaths occurred in arm A during or within 30 days of study treatment. Two of the deaths were considered possibly related to study treatment toxicity (one death was associated with disseminated intravascular coagulation, bowel perforation, and septicemia, and the other death was associated with acute leukoencephalopathy). Table 3 lists incidences and categories of AEs. Overall, arm A had more grade 3 or 4 AEs than arm B (P.001); however, grade 4 toxicities were generally infrequent. Hematologic grade 3 or by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

5 Gemcitabine and Cisplatin Chemoradiation in Cervical Carcinoma A Progression-Free Survival (probability) B Overall Survival (probability) Arm A Arm B Log-rank P =.023 HR, 0.68; 95% CI, 0.49 to Arm A Arm B Log-rank P =.022 HR, 0.68; 95% CI, 0.49 to 0.95 Time (months) Time (months) Fig 3. Kaplan-Meier estimates of (A) progression-free survival (PFS) and (B) overall survival for patients who were randomly assigned to arm A or arm B. PFS at 3 years is shown by the dotted black lines and was 74.4% for arm A and 65.0% for arm B (P.029). HR, hazard ratio. AEs were predominant (arm A, 71.9%; arm B, 23.9%), most commonly neutropenia. Arm A had more grade 3 or 4 nonhematologic toxicities, including vomiting and diarrhea, than arm B (P.002). Similarly, exploratory analyses, in which toxicity was assessed separately for preadjuvant and adjuvant treatment (Fig 2), showed that arm A had more toxicity during the preadjuvant phase than arm B (Data Supplement) and that incidence of new toxicity in the adjuvant treatment phase was generally low (Data Supplement). There were more discontinuations in arm A than in arm B (P.001), most commonly as a result of AEs, including decreased creatinine clearance, neutropenia, and diarrhea. Although 30 arm A patients were hospitalized as a result of study drug compared with 11 arm B patients (P.003), the median number of hospitalizations as a result of study drug was zero for both arms (arm A range, zero to 10 hospitalizations; arm B range, zero to five hospitalizations). Up to 30 days from last study drug dose, 198 patients (arm A: 128 patients, 49.2%; arm B: 70 patients, 27.5%; P.001) received one transfusion (median of zero transfusions per arm; arm A range, zero to eight transfusions; arm B range, zero to five transfusions). Further details and exploratory analyses examining these factors during preadjuvant and adjuvant phases can be found in the Data Supplement. Treatment Table 2. Treatment Received Arm A (n 260) Arm B (n 255) All Patients (N 515) Chemotherapy Chemoradiotherapy phase Gemcitabine, cycles Median 5 Range 1-6 Cisplatin, cycles Median Range Adjuvant chemotherapy phase Received cycle 1 patients % Received cycle 2 patients % Radiation therapy Chemoradiotherapy phase patients % Duration of XRT, days Median IQ range Total XRT dose, Gy Median IQ range BCT patients % Total point A dose, Gy Median IQ range Total point B dose, Gy Median IQ range Overall study Duration of all RT, days Median IQ range Cumulative BCT point A XRT dose, Gy Median IQ range Cumulative BCT point B XRT dose, Gy Median IQ range NOTE. As-treated data are presented. Because of rounding, not all percentages total 100%. All patients received at least one dose of study drug and at least one dose of RT during the chemoradiotherapy phase of the study. Arm A treatment consisted of gemcitabine plus cisplatin with concurrent RT, followed by BCT, and then by adjuvant gemcitabine and cisplatin. Arm B treatment consisted of cisplatin with concurrent RT, followed by BCT and no adjuvant chemotherapy. Abbreviations: XRT, external-beam radiation; IQ, interquartile; BCT, brachytherapy; RT, radiotherapy. One dose per treatment cycle. Two doses of gemcitabine per treatment cycle; one dose of cisplatin per treatment cycle. Incidence of late toxicities was generally low. Incidence of grade 3 or 4 late toxicities was numerically higher in arm A versus arm B, although the difference was not statistically significant, mainly because of increased incidence of grade 4 intestinal toxicity (arm A: 2.3%; arm B: 0%; Table 4) by American Society of Clinical Oncology 5

6 Dueñas-González et al Table 3. Acute Toxicities Related to Study Drug ArmA(n 260) Arm B (n 255) Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Toxicity Neutropenia Diarrhea Nausea Anemia Vomiting Fatigue Radiation dermatitis Thrombocytopenia Abdominal pain/ cramping Anorexia Proctitis AST ALT Creatinine Febrile neutropenia NOTE. Some patients had more than one adverse effect. Other grade 3 and 4 adverse effects for both treatment arms are listed in the Data Supplement. Because of rounding, not all percentages add up to 100%. Toxicity was measured weekly while on study and again 30 days after last study treatment. Dose adjustments for toxicity are described in the Data Supplement. Arm A treatment consisted of gemcitabine plus cisplatin with concurrent radiation therapy, followed by brachytherapy, and then by adjuvant gemcitabine and cisplatin. Arm B treatment consisted of cisplatin with concurrent radiation therapy, followed by brachytherapy and no adjuvant chemotherapy. DISCUSSION Cisplatin-based chemoradiotherapy confers a survival advantage over radiation alone, 3-7,17 but improvements in patient outcomes have not yet been shown when multiagent modalities that included cisplatinbased chemoradiotherapy were compared with single-agent cisplatin chemoradiotherapy. 8 Therefore, to our knowledge, the data from this trial represent the first finding of significantly improved survival outcomes (PFS at 3 years, overall PFS, and OS) for LACC with combination multimodality therapy compared with singleagent cisplatin chemoradiotherapy. The magnitude of improvement in survival outcomes in arm A over arm B is similar to that observed when standard concurrent chemoradiotherapy is compared with radiotherapy alone. 5,18 The 3-year probabilities for PFS (65.0%) and OS (69.1%) in the standard chemoradiotherapy arm of this study are consistent with those Table 4. Late Toxicities ArmA(n 260) Arm B (n 255) Maximum Toxicity Grade Maximum Toxicity Grade Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Grade 0 Grade 1 Grade 2 Grade 3 Grade 4 Toxicity Patients Patients Any Skin Subcutaneous tissue Mucous membrane Spinal cord Small/large intestine Bladder NOTE. Late toxicity was measured 30 days after last study drug dose and again every 4 months until 1 year after completion of study drug treatment. Because of rounding, not all percentages add up to 100%. Incidence of late toxicity was compared between treatment arms using the Fisher s exact test. Supplementary information on late toxicities is included in Data Supplement. Arm A treatment consisted of gemcitabine plus cisplatin with concurrent radiation therapy, followed by brachytherapy, and then by adjuvant gemcitabine and cisplatin. Arm B treatment consisted of cisplatin with concurrent radiation therapy, followed by brachytherapy and no adjuvant chemotherapy. Some data were missing by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

7 Gemcitabine and Cisplatin Chemoradiation in Cervical Carcinoma reported in similar patient populations, 18 which suggests that the improvements seen in arm A are not related to underperformance in arm B. Moreover, the TtPD curves show improvement from as early as 6 months after treatment initiation, suggesting a rapid and sustained effect of the gemcitabine combination over standard chemoradiotherapy. Inclusion of gemcitabine in this regimen has the convenience of short-infusion, same-day administration with cisplatin. Although single-agent weekly gemcitabine 125 mg/m 2 has limited systemic activity, this dose synergizes effectively with cisplatin and radiation in cervical cancer, 11,19 producing higher complete pathologic response compared with cisplatin alone. 12 The relative LFR improvement ( 30%) in arm A compared with arm B was not statistically significant (P.097) but was directionally consistent with the statistically significant distant failure rate improvement. These failure rates were similar to those in a recent meta-analysis 20 (when analyzed as a proportion of total randomly assigned patients; Data Supplement). Although results from a comparative phase II study 12 support this study s findings of improved efficacy with arm A, absence of detailed pathologic response analyses and use of physical examination to assess tumor response in this study limit our ability to draw definitive conclusions regarding specific effects of the combination therapy on the local tumor, whereby metastatic potential could have been inhibited, or on subclinical (occult) metastases. The lower local and distant failure rates observed in arm A might be attributable to the addition of gemcitabine to the concurrent chemoradiotherapy phase because a recent meta-analysis found significant improvements of both local and distant relapses when concurrent chemotherapy was added to radiation in the absence of adjuvant chemotherapy. 2 The adjuvant chemotherapy phase may also have contributed. However, this is not part of routine practice in cervical cancer; there are only limited data from controlled studies with small numbers of patients, and no study addresses the value of maintenance versus no maintenance. 13 Although limited evidence suggests a potential benefit, interpretation of the contribution of adjuvant chemotherapy in this study is challenging because the study design did not control for the adjuvant chemotherapy phase and the patients previously received six or fewer cycles of systemically active cisplatin. Finally, although patients in this study were prospectively randomly assigned to balance for several known clinicopathologic factors between treatment arms 21,22 and post hoc analyses showed that hemoglobin levels and other factors were similar between treatment arms (data not shown), other unknown prognostic or predictive factors may have been unevenly distributed between treatment arms and may have affected response to study treatment. Increased toxicity of multidrug combinations is of concern because of direct cytotoxic effects of the drugs, as well as potentiation of radiation-induced toxicity. Others have shown that treatment with multidrug regimens results in increased toxicity compared with single-drug regimens, often with no increase in efficacy. 5,6 Two previous phase I studies showed unacceptable systemic toxicity levels when gemcitabine and cisplatin chemoradiotherapy were used in combination 23,24 ; however, sample sizes were small (nine and 13 patients, respectively), and gemcitabine treatment preceded cisplatin treatment. As found in other phase I and II studies, 11,12,19 toxicity, particularly hematologic toxicity, was increased with the gemcitabine combination in this study, but grade 3 and 4 toxicity rates were low overall and were considered clinically acceptable, given that there was significantly increased efficacy and that similar changes in toxicity were seen with the addition of chemotherapy to radiotherapy in a recent cervical cancer meta-analysis. 2 Specifically, there were low levels of grade 4 severity, few patients discontinued treatment as a result of toxicity or AEs, and less than 1% of patients died during study treatment or within 30 days of last study drug dose. Moreover, late toxicity incidence was low in both treatment arms, suggesting acceptable post-therapy quality of life. However, reliable assessment of late toxicity was not possible because it was not mandatory to collect late toxicities beyond 1 year of treatment completion. Although the incidence of late toxicity was similar to the incidence reported in recent meta-analyses, the data in those studies were also incomplete. 13,18 More late toxicity details are provided in the Data Supplement. We cannot conclusively attribute increased toxicity in the gemcitabine-containing arm to the preadjuvant or adjuvant phases because the study was not designed to do this. However, post hoc analyses implied that most of the toxicities and discontinuations caused by AEs occurred during the preadjuvant phase, suggesting that adjuvant treatment had a relatively minor contribution to toxicity. Radiotherapy dosing was similar across both arms, and most patients completed their treatment within 8 weeks; however, median radiotherapy duration was a few days longer for arm A than for arm B. Other studies suggest that increased radiotherapy duration has detrimental effects on progression and survival. 25,26 In this study, it is likely that increased toxicity experienced by patients in the gemcitabine-containing arm delayed administration of radiation, thereby extending overall duration of radiation in this treatment arm. In conclusion, this study demonstrates that, within study limitations and despite increased but clinically manageable toxicity, the gemcitabine combination improves survival outcomes for patients with LACC compared with current standard of care. Further studies are required to define optimal patient selection for this combination and to delineate the specific contributions of multiagent chemoradiotherapy and adjuvant chemotherapy phases to survival outcomes. AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a U are those for which no compensation was received; those relationships marked with a C were compensated. For a detailed description of the disclosure categories, or for more information about ASCO s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment or Leadership Position: Julie M. Blair, Eli Lilly (C); Helen Barraclough, Eli Lilly (C); Mauro Orlando, Eli Lilly (C) Consultant or Advisory Role: None Stock Ownership: Julie M. Blair, Eli Lilly; Helen Barraclough, Eli Lilly; Mauro Orlando, Eli Lilly Honoraria: None Research Funding: Alfonso Dueñas-González, Eli Lilly; Juan J. Zarbá, Eli by American Society of Clinical Oncology 7

8 Dueñas-González et al Lilly; Firuza Patel, Eli Lilly; Juan C. Alcedo, Eli Lilly; Semir Beslija, Eli Lilly; Luis Casanova, Eli Lilly; Pittayapoom Pattaranutaporn, Eli Lilly; Shahid Hameed, Eli Lilly; Julie M. Blair, Eli Lilly; Helen Barraclough, Eli Lilly; Mauro Orlando, Eli Lilly Expert Testimony: None Other Remuneration: None AUTHOR CONTRIBUTIONS Conception and design: Alfonso Dueñas-González, Juan J. Zarbá, Shahid Hameed, Mauro Orlando Provision of study materials or patients: Alfonso Dueñas-González, Juan J. Zarbá, Firuza Patel, Juan C. Alcedo, Semir Beslija, Luis Casanova, Pittayapoom Pattaranutaporn, Shahid Hameed Collection and assembly of data: Alfonso Dueñas-González, Firuza Patel, Semir Beslija, Pittayapoom Pattaranutaporn, Shahid Hameed, Mauro Orlando Data analysis and interpretation: Alfonso Dueñas-González, Juan J. Zarbá, Firuza Patel, Juan C. Alcedo, Semir Beslija, Luis Casanova, Pittayapoom Pattaranutaporn, Shahid Hameed, Julie M. Blair, Helen Barraclough, Mauro Orlando Manuscript writing: Alfonso Dueñas-González, Juan J. Zarbá, Firuza Patel, Juan C. Alcedo, Semir Beslija, Luis Casanova, Pittayapoom Pattaranutaporn, Shahid Hameed, Julie M. Blair, Helen Barraclough, Mauro Orlando Final approval of manuscript: Alfonso Dueñas-González, Juan J. Zarbá, Firuza Patel, Juan C. Alcedo, Semir Beslija, Luis Casanova, Pittayapoom Pattaranutaporn, Shahid Hameed, Julie M. Blair, Helen Barraclough, Mauro Orlando REFERENCES 1. Parkin DM, Bray F, Ferlay J, et al: Global cancer statistics, CA Cancer J Clin 55:74-108, Green JA, Kirwan JM, Tierney JF, et al: Survival and recurrence after concomitant chemotherapy and radiotherapy for cancer of the uterine cervix: A systematic review and meta-analysis. Lancet 358: , Keys HM, Bundy BN, Stehman FB, et al: Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 340: , Morris M, Eifel PJ, Lu J, et al: Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N Engl J Med 340: , Rose PG, Bundy BN, Watkins EB, et al: Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 340: , Whitney CW, Sause W, Bundy BN, et al: Randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stage IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes: A Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 17: , Peters WA 3rd, Liu PY, Barrett RJ 2nd, et al: Concurrent chemotherapy and pelvic radiation therapy compared with pelvic radiation therapy alone as adjuvant therapy after radical surgery in high-risk early-stage cancer of the cervix. J Clin Oncol 18: , Rose PG: Chemoradiotherapy for cervical cancer. Eur J Cancer 38: , Hernández P, Olivera P, Dueñas-Gonzalez A, et al: Gemcitabine activity in cervical cancer cell lines. Cancer Chemother Pharmacol 48: , Candelaria M, Garcia-Arias A, Cetina L, et al: Radiosensitizers in cervical cancer: Cisplatin and beyond. Radiat Oncol 1:15, Zarbá JJ, Jaremtchuk AV, Gonzalez Jazey P, et al: A phase I-II study of weekly cisplatin and gemcitabine with concurrent radiotherapy in locally advanced cervical carcinoma. Ann Oncol 14: , Dueñas-González A, Cetina-Perez L, Lopez- Graniel C, et al: Pathologic response and toxicity assessment of chemoradiotherapy with cisplatin versus cisplatin plus gemcitabine in cervical cancer: A randomized phase II study. Int J Radiat Oncol Biol Phys 61: , Chemoradiotherapy for Cervical Cancer Meta-Analysis Collaboration: Reducing uncertainties about the effects of chemoradiotherapy for cervical cancer: A systematic review and metaanalysis of individual patient data from 18 randomized trials. J Clin Oncol 26: , American Joint Committee on Cancer: Cervix uteri, in: AJCC Cancer Staging Manual (ed 5). Philadelphia, PA, Lippincott-Raven, 1997, pp Pocock SJ, Simon R: Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 31: , Schemper M, Smith TL: A note on quantifying follow-up in studies of failure time. Control Clin Trials 17: , Eifel PJ, Winter K, Morris M, et al: Pelvic irradiation with concurrent chemotherapy versus pelvic and para-aortic irradiation for high-risk cervical cancer: An update of Radiation Therapy Oncology Group trial (RTOG) J Clin Oncol 22: , Green J, Kirwan J, Tierney J, et al: Concomitant chemotherapy and radiation therapy for cancer of the uterine cervix. Cochrane Database Syst Rev 3:CD002225, Umanzor J, Aguiluz M, Pineda C, et al: Concurrent cisplatin/gemcitabine chemotherapy along with radiotherapy in locally advanced cervical carcinoma: A phase II trial. Gynecol Oncol 100:70-75, Chemoradiotherapy for Cervical Cancer Meta- Analysis Collaboration (CCCMAC): Reducing uncertainties about the effects of chemoradiotherapy for cervical cancer: Individual patient data metaanalysis. Cochrane Database Syst Rev 1:CD008285, Monk BJ, Tian C, Rose PG, et al: Which clinical/pathologic factors matter in the era of chemoradiation as treatment for locally advanced cervical carcinoma? Analysis of two Gynecologic Oncology Group (GOG) trials. Gynecol Oncol 105: , Winter WE 3rd, Maxwell GL, Tian C, et al: Association of hemoglobin level with survival in cervical carcinoma patients treated with concurrent cisplatin and radiotherapy: A Gynecologic Oncology Group study. Gynecol Oncol 94: , Puget Sound Oncology Consortium, Swisher EM, Swensen RE, et al: Weekly gemcitabine and cisplatin in combination with pelvic radiation in the primary therapy of cervical cancer: A phase I trial of the Puget Sound Oncology Consortium. Gynecol Oncol 101: , Rose PG, Degeest K, McMeekin S, et al: A phase I study of gemcitabine followed by cisplatin concurrent with whole pelvic radiation therapy in locally advanced cervical cancer: A Gynecologic Oncology Group study. Gynecol Oncol 107: , Perez CA, Grigsby PW, Castro-Vita H, et al: Carcinoma of the uterine cervix: I. Impact of prolongation of overall treatment time and timing of brachytherapy on outcome of radiation therapy. Int J Radiat Oncol Biol Phys 32: , Chen SW, Liang JA, Yang SN, et al: The adverse effect of treatment prolongation in cervical cancer by high-dose-rate intracavitary brachytherapy. Radiother Oncol 67:69-76, 2003 Acknowledgment We thank the International Network for Cancer Treatment and Research Clinical Trials Office; Blue Gum Data Analysis (Australia); Yashan Zhou, Karen Donaldson, Irene Tomlin, and Victoria Reed, PhD, for their assistance; and Subhashini John, MD, and P.G. Jayaprakash, MD, (India) for their participation by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY