Randomized Comparison of Weekly Cisplatin or Protracted Venous Infusion of Fluorouracil in Combination With Pelvic Radiation in Advanced Cer
http://www.100md.com
《临床肿瘤学》
the Department of Radiation Oncology, Delaware County Memorial Hospital, Drexel Hill, PA
Radiation Oncologist, St Joseph's Hospital/Cancer Institute, Tampa, FL
Gynecologic Oncology Group Statistical and Data Center, Roswell Park Cancer Institute, Buffalo, NY
University of Oklahoma, Oklahoma City, OK
Department of Gynecologic Oncology, University of Texas, Galveston Medical Branch, Galveston, TX
Department of Obstetrics/Gynecology, Indiana University School of Medicine, Indianapolis, IN
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of California-Irvine Medical Center, Orange, CA
Norton Healthcare Inc, Clinical Pathology Associates, Louisville, KY
ABSTRACT
PURPOSE: Concurrent chemoradiotherapy is the standard of care for locally advanced cervix cancer; the optimal chemotherapy regimen is not yet defined. This trial was designed to compare the outcome of protracted venous infusion (PVI) fluorouracil (FU) with standard weekly cisplatin and concurrent radiation therapy (RT).
PATIENTS AND METHODS: Patients with stage IIB, IIIB, and IVA cervical cancer with clinically negative aortic nodes were eligible. Pelvic RT dose was 45 Gy with a parametrial boost to involved sides of 5.4 to 9 Gy, and high- or low-dose rate intracavitary brachytherapy. Standard therapy was weekly cisplatin 40 mg/m2, and experimental therapy was PVI FU 225 mg/m2/d for 5 d/wk for six cycles during RT.
RESULTS: The study was closed prematurely when a planned interim futility analysis indicated that PVI FU/RT had a higher treatment failure rate (35% higher) and would, most likely, not result in an improvement in progression-free survival compared with weekly cisplatin/RT. The PVI FU/RT arm continues to show a higher risk of treatment failure (relative risk [RR] unadjusted, 1.29) and a higher mortality rate (RR unadjusted, 1.37). There was no difference in pelvic treatment failure between regimens, but there was an increase in the failure rate at distant sites in the PVI FU arm.
CONCLUSION: In this study, PVI FU does not show improved outcome over weekly cisplatin. Future research should explore combinations of FU with cisplatin, new radiosensitizers, and active drugs combined with RT to reduce the high rate of pelvic and distant treatment failure still seen in advanced cervix cancer.
INTRODUCTION
Five randomized trials have demonstrated the superiority of combined chemotherapy with radiation in the treatment of advanced cervix cancer.1-5 The Gynecologic Oncology Group (GOG) has studied both cisplatin/fluorouracil (FU) and weekly cisplatin concurrent with radiation therapy (RT) in two sequential randomized trials and both trials demonstrated a survival benefit for combined chemoradiotherapy compared with hydroxyurea and RT.1,2 Weekly cisplatin was directly compared with FU/cisplatin/hydroxyurea in combination with RT in the most recent GOG trial and there was no difference in progression-free survival (PFS) between the two drug regimens; however, a significant difference in adverse events for the three-drug regimen was demonstrated.2 Therefore, the regimen of weekly cisplatin concurrent with RT was chosen as the standard treatment approach in the GOG for locally advanced cervix cancer with which new combination treatments should be compared.
FU as a single-agent RT sensitizer had not been evaluated for cervix cancer in the GOG. Various combined-modality programs have used infusional FU with RT, especially for GI cancers.6-8 A phase III randomized trial of adjuvant postoperative chemoradiotherapy for high-risk rectal cancer demonstrated an improvement in relapse-free and overall survival for protracted venous infusion (PVI) FU compared to bolus FU in combination with pelvic RT.9 The only previous randomized trial of FU concurrent with RT for cervix cancer found a benefit to FU with improved disease-free survival in subset analysis only (stage IB/IIB or medial stage IIB). Overall, survival and pelvic control were not statistically improved.10 In combination with RT, PVI FU is also appealing for the treatment of cervix cancer because increased drug-RT interactions with tumor cells should occur with potentially increased RT sensitivity and pelvic control. In adenocarcinoma tissue cell lines, cells exposed continuously to FU for 48 hours after RT had maximum RT sensitivity.11 The disadvantage of this approach is the need for central access and specialized equipment to deliver continuous FU, and the potential for increased acute and chronic bowel toxicity. Because it has been a strategy for the GOG to investigate the benefit of chemoradiotherapy and to develop improved regimens, standard weekly cisplatin in combination with RT was compared with the experimental arm of PVI FU in this phase III trial for advanced cervix cancer (GOG 165). This report details the results of that study.
PATIENTS AND METHODS
Patients with primary, previously untreated, histologically confirmed invasive squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma of the uterine cervix stage IIB, IIIB, and IVA were eligible for GOG 165. Histologic diagnoses were confirmed by centralized review by the GOG Pathology Committee. Patients with stage IIIA disease or lower-third vaginal involvement were ineligible because of the individualized implantation techniques required for these patients. Patients must have had negative para-aortic lymph nodes as determined by lymphangiogram, computed tomography (CT), magnetic resonance imaging, or lymphadenectomy. However, this is the first GOG trial for locally advanced cervix cancer that did not require surgical staging of the para-aortic lymph nodes. As is standard for all previous and current cervix studies, patients were required to have adequate hematologic, renal, and hepatic function, as well as performance status, for study entry.
Two biologic correlative studies were done in GOG 165 that required additional laboratory analysis. The first study sought to determine prospectively if smoking behavior is a confounding variable in the effectiveness of RT for locally advanced cervix cancer. This was accomplished through the use of a standard smoking questionnaire, with biochemical validation of smoking status confirmed through urine cotinine analysis, as has been reported recently.12 The second study sought to determine the prognostic value of hemoglobin levels for advanced cervix cancer, with data regarding hemoglobin levels and transfusion history before and during RT collected prospectively and correlated with outcome. No attempt was made to maintain patients at any specific hemoglobin level as part of this protocol. Both correlative studies will be reported separately. This study received local institution board approval and all patients signed an approved, informed consent.
Study Modality: RT
The RT protocol for advanced cervix cancer was updated for GOG 165 by the GOG Radiation Oncology Committee. This included a higher point A dose (85 Gy) with combined external and intracavitary RT, shorter overall treatment time (< 10 weeks), and expanded external-beam field definitions. For the first time in GOG trials, high-dose rate (HDR) brachytherapy was accepted. Physicians planning to treat a patient with HDR brachytherapy were required to complete HDR certification before the patient's enrollment. This required submitting RT dosimetry of two previous cervix cancer patients treated with HDR brachytherapy, to be evaluated by the HDR Radiation Oncology Subcommittee and the Radiologic Physics Center. Whole pelvic field definitions were modified in this protocol to avoid possible marginal misses posteriorly; this was accomplished by placing the posterior border on the lateral field behind the anterior bony sacrum. CT slices through the tumor volume were required and reviewed by the Radiation Oncology Committee in conjunction with simulation films at the semiannual GOG meetings to analyze coverage of the gross tumor volume. The treating physician was required to draw the cervical mass, uterus, and involved pelvic lymph nodes, if any, on the simulation films submitted for review.
External RT dose prescription to the whole pelvis was 45 Gy in 25 fractions at the isocenter. A four-field box arrangement was recommended. Intracavitary RT could include either low-dose rate (LDR; 40 Gy to point A in one to two fractions) or HDR (30 Gy to point A in five fractions) brachytherapy. The type of brachytherapy (HDR v LDR) was chosen at the time of patient random assignment. Tandem and ovoid applicators were recommended for both LDR and HDR brachytherapy, and HDR brachytherapy started the fourth week of external-beam therapy with at least one fraction per week. After completion of external-beam therapy, two fractions of HDR were given per week. A parametrial boost of 5.4 to 9.0 Gy was given to the involved parametrium at the completion of whole pelvic RT.
All charts/simulation films/portal films and pretreatment CT scans were reviewed at biannual GOG meetings by a team of radiation oncologists. A major variation was scored if the dose was 11% to 20% less/more than protocol prescription dose or if elapsed time in days was greater than 20% longer compared with expected time in days for protocol prescription. All CT scans were reviewed and a major variation was scored if field margins were tight on the tumor. If tumor was not satisfactorily included in treatment volume, the fields were deemed unacceptable.
Modifications to RT treatment were described in the protocol and included appropriate delays for neutrophil count less than 1,000 K/μL, platelet count less than 50,000 K/μL or grade 3 to 4 GI or genitourinary toxicity.
Study Modality: Chemotherapy
The standard (arm I) treatment consisted of weekly cisplatin (40 mg/m2 to a maximum of 70 mg) started on day 1 of external RT, preferably 4 hours before RT initiation. Cisplatin could be given during parametrial boost as well as pelvic RT, for a total of six cycles. Cisplatin was reduced by 25% for grade 4 nausea/vomiting or grade 2 neurotoxicity and was discontinued for creatinine greater than 2.0 mg/100 mL or grade 3 to 4 neurotoxicity. Cisplatin was withheld for a WBC less than 3,000 K/μL or a platelet count less than 100,000 K/μL.
The experimental (arm II) treatment consisted of continuous infusion FU 225 mg/m2/d, 5 days per week throughout external-beam whole pelvis and parametrial boost, for a total of 6 weeks. All patients on this arm were required to have central access, and anticoagulation was suggested. Chemotherapy was delayed for WBC less than 2,000 K/μL, platelet less than 50,000 K/μL, or grade 3 diarrhea, and was resumed at full dose when toxicity resolved. For grade 2 mucositis, stomatitis, or esophagitis, or grade 2 weight loss, FU was to be temporarily discontinued and restarted at a dose reduction of 50 mg/m2/d when symptoms subsided.
The GOG Statistical Center randomly assigned the treatment regimen using a fixed block with an equal number of each regimen for each combination of stratification factors: International Federation of Gynecology and Obstetrics stage (IIB, III, IVA), type of brachytherapy (LDR or HDR), and whether patient had undergone para-aortic lymph nodes sampling. A sample size of 416 patients was set, with an estimated 26 months of follow-up, to observe 150 recurrences before testing the primary hypothesis, which was based on the research question of whether concurrent prolonged infusion of FU and RT decreases the risk of progression when compared with concurrent weekly cisplatin and RT. This was a one-sided test with the probability of a type I error set at .05 and the statistical power of 0.80 to detect a hazard ratio (FU to cisplatin) of 0.67. Survival was also a primary end point but no adjustment was made to control the type I error due to the strong correlation between PFS and survival. PFS was defined as the date from protocol registration to the date of reappearance of disease, progression of existing disease, or death, whichever comes first. Survival was defined as the length of life from entry onto protocol to death, or to the date of last contact. Recurrence site was considered local if it was within the pelvic field and distant if it was outside the pelvic field. Distant treatment failure was categorized by site of failure, such as para-aortic or supraclavicular lymph nodes, lung, liver, bone, and so on.
All eligible patients were included in the analysis of survival and PFS (intent-to-treat principle for eligible patients). All causes of death were used to calculate survival, and the estimates of the cumulative proportions of survival were based on Kaplan-Meier procedures.13 Relative risk (RR) estimates and CIs of treatment effects on progression and death, with adjustments for prognostic factors, were accomplished using the Cox model.14 The cumulative incidence curve was used to evaluate the rate of site-specific recurrences between the regimens.15 This method properly handles censored patients and determines the incidence rate of local failure in the presence of failure at distant sites (with/without local relapse), and death due to intercurrent disease.
A formal futility analysis was included in the design of this trial and called for an analysis when 50 recurrences (one third of the final goal) were observed. The stopping rule was to consider study closure if there was any degree of excess rate of recurrence in the PVI FU group using the RR estimate. The RR of progression was 1.35 (90% CI, 0.87 to 2.09). The RR of mortality was 1.02.
RESULTS
The study opened for accrual in October 1997. It was initially a three-arm trial with pelvic irradiation and intracavitary brachytherapy alone as the third arm. When data became available from five randomized trials testing the benefit of concurrent chemoradiotherapy for cervix cancer, the RT-only arm, with 24 patients accrued, was closed in August 1998. The results reported here represent mature data of the two chemoradiotherapy arms. In July 2000, the Data Monitoring Committee (DMC) met and reviewed the interim analysis that included 54 disease progressions and four deaths without progression. The DMC decided to close the trial prematurely because of the increased rate of progression for patients on arm II. The current report is based on 316 eligible patients, of whom 159 were assigned to arm I and 157 were assigned to arm II. Median follow-up of those still alive at the date of analysis is 40.4 months, with quartiles of 33.8, 40.4, and 49.0 months.
Patient characteristics are listed in Table 1. Only 18% of patients had surgically staged para-aortic lymph nodes, which is a dramatic change from prior GOG studies where surgical staging was required. Approximately 14% of patients had either radiographically or surgically determined positive pelvic lymph nodes.
More patients on arm II (72%) than arm I (56%) completed six cycles of chemotherapy (Table 2). However, at least five cycles were received by 74% of patients on the cisplatin arm versus 86% on the PVI FU arm. Among patients treated using HDR, the median dose to point A was 79.8 Gy for the cisplatin arm and 75.0 Gy for the PVI FU arm (prescription protocol dose was 75 Gy). Two patients (5%) treated with HDR brachytherapy were considered to have a major violation but in only one patient was this because of a low dose. Among patients treated using LDR, the median dose to point A was 86.4 Gy for the cisplatin arm and 84.8 Gy for PVI FU arm (prescription protocol dose was 85 Gy). Thirty-eight (15%) patients treated with LDR brachytherapy were considered to have a major violation, but in only eight patients (3%) were this because of low dose. Fourteen patients (4%) failed to complete protocol treatment and were coded as a protocol violation because of low dose. Median total elapsed time for HDR patients was 7.1 weeks and the maximum time was 9.6 weeks to complete RT treatment. Median total elapsed time for LDR patients was 7.3 weeks, with only 10 patients (4%) requiring more than 10 weeks to complete treatment (Table 2).
Interestingly, despite the additional chemotherapy received in the PVI FU arm, grade 3 to 4 toxicities occurred less frequently, including hematologic (10% v 33%) and GI (19% v 25%). Other grade 3 to 4 adverse events were reported in less than 10% of patients. Combining all toxicity classifications, 45% of patients had grade 3 to 4 toxicity during either treatment or follow-up, with a higher rate seen for the cisplatin group (58% v 32%). If reversible hematologic toxicity was excluded from this combined analysis, grade 3/4 toxicity was still higher with cisplatin compared with FU (43% v 27%).
There have been 123 (39%) disease progressions, with an additional 20 deaths without a documented progression, which were included in the computation of PFS. Figure 1 displays the PFS rate through 4 years for the cisplatin group compared with the PVI FU group (57% v 50%; not significant). The unadjusted RR for recurrence was 1.29 (95% CI; 0.93 to 1.80) and 1.25 when adjusting for stage, brachytherapy type, performance of surgical staging, and tumor size (95% CI, 0.90 to 1.74).
There have been 121 deaths reported. There was a difference in OS at 4 years with 36% of patients dead of disease on arm I compared with 45% of patients on arm II (Fig 2) The unadjusted relative risk for death was 1.37 (95% CI, 0.96 to 1.97) and 1.33 when adjusting for stage, brachytherapy type, performance of surgical staging, and tumor size.
Failure rate (cumulative incidence rate) within 4 years confined to the pelvis alone was 16% and 14% in the cisplatin and PVI FU arm, respectively. In the same time frame, distant failure (including abdominal, para-aortic region, bone, liver, and lung) was higher in the PVI FU group (29% v 18%). The higher rate among the PVI FU group was consistent for specific distant sites: lung metastases (9% v 5%) and abdominal failures (11% v 3%). Para-aortic failure occurred in only 7% and 5% of patients in the PVI FU and cisplatin arms, respectively, despite the fact that only 18% of patients were surgically staged in the para-aortic region.
The study was closed by the GOG DMC based on its review of results of the planned interim analysis, which indicated that there was increased risk of progression of 35% in the PVI FU group (ie, unadjusted RR of 1.35 for the PVI FU group compared with the cisplatin group). Under the rules of the interim analysis plan (specifically, futility analysis), the null hypothesis (chemoradiotherapy with PVI FU is not more efficacious than chemoradiotherapy with weekly cisplatin) was accepted. In short, a statistically significant improvement in PFS in the PVI FU group compared with the weekly cisplatin group would not, in all likelihood, be achieved if the study were to be completed; such a deficit was observed after one third of the PFS information (ie, one third of the total progression/deaths required for the completed study analysis) was available.
DISCUSSION
Concurrent chemoradiotherapy is the standard of care for locally advanced cervix cancer. It is employed as definitive treatment for stage IB bulky and higher disease, and as postoperative treatment when nodes or margins are positive or parametrial extension is noted pathologically. This approach was initially reported as a National Cancer Institute consensus statement in 1999, and has gained widespread acceptance.16 The optimal chemotherapy regimen is not yet defined; however, weekly cisplatin2,3 and FU plus cisplatin every 3 weeks1,4,5 are the most popular regimens. The value of weekly cisplatin in conjunction with external pelvic RT as used in this study has been reported previously by Keys et al3 for stage IB bulky cervix cancer. Pearcey et al17 could not confirm this; their contrary result has been discussed extensively in an editorial by Rose and Bundy.18 Rose et al2 compared weekly cisplatin versus cisplatin/FU/hydroxyurea versus hydroxyurea alone, concurrent with pelvic irradiation, and found the cisplatin arms to be comparable and superior to hydroxyurea alone. The weekly cisplatin was better tolerated and therefore became the GOG standard.
There appears to be no benefit to PVI FU as administered in this study over weekly cisplatin for advanced cervix cancer. This lack of benefit was also shown with an alternate FU schedule of 1 g/m2/d during the first and last weeks of external RT compared with RT alone.10 Because this study was closed short of its expected accrual, it is not powered to detect a statistically significant difference between the treatment arms in PFS or OS.
This was the first GOG study to accept clinical staging and HDR brachytherapy for advanced cervix cancer populations. Given that less than 20% of patients on this study had surgical staging and pathologically negative para-aortic lymph nodes, this group of patients should have a higher risk of microscopic involvement of pelvic and para-aortic lymph nodes, and subsequently lower survival rates compared with previous GOG trials. Despite these facts, the outcome for the cisplatin arm in this study is comparable to results of other studies published recently (Table 3; Fig 3). One hypothesis is that the improvement is secondary to the higher RT doses and shorter overall treatment times used in the current study compared with previous GOG trials. Because only 13% of patients in this study received HDR brachytherapy, this technology and its outcome cannot be evaluated.
As administered in this study, PVI FU did not demonstrate statistical improvement over weekly cisplatin as a radiosensitizer. Future research investigating new radiosensitizers and active drugs in combination with radiotherapy to improve the significant pelvic and distant failure rates seen in locally advanced cervix cancer is warranted.
Appendix
The following member institutions participated in this study: University of Alabama at Birmingham, Duke University Medical Center, Abington Memorial Hospital, Walter Reed Army Medical Center, Wayne State University, University of Mississippi Medical Center, Colorado Gynecologic Oncology Group, PC, University of California at Los Angeles, University of Washington, Milton S. Hershey Medical Center, University of Cincinnati, University of North Carolina School of Medicine, University of Iowa Hospitals and Clinics, University of Texas/Southwestern Medical Center at Dallas, Indiana University Medical Center, Wake Forest University School of Medicine, University of California Medical Center at Irvine, Rush-Presbyterian-St Luke's Medical Center, State University of New York Downstate Medical Center, University of Kentucky, The Cleveland Clinic Foundation, State University of New York at Stony Brook, Washington University School of Medicine, Columbus Cancer Council, University of Massachusetts Medical Center, Fox Chase Cancer Center, Medical University of South Carolina, Women's Cancer Center, University of Oklahoma, University of Virginia, University of Chicago, Tacoma General Hospital, Mayo Clinic, Case Western Reserve University, Tampa Bay Cancer Consortium, North Shore University Hospital, Brookview Research Inc, and Ellis Fischel Cancer Center.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. 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.
NOTES
Supported by National Cancer Institute grants to the Gynecologic Oncology Group Administrative Office (Grant No. CA 27469) and the Gynecologic Oncology Group Statistical Office (Grant No. CA 37517).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Whitney CW, Sause W, Bundy BM, et al: Randomized comparison of fluorouracil plus cisplatin vs. 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 17:1339-1348, 1999
Rose P, Bundy B, Watkins E, et al: Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 340:1144-1153, 1999
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 340:1154-1161, 1999
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) 90-01. J Clin Oncol 22:872-880, 2004
Peters WA, Liu PY, Barrett RJ, 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 18:1606-1613, 2000
Minsky B, Pajak T, Ginsberg R, et al: Phase III trial of combined modality therapy for esophageal cancer: High dose versus standard dose radiation therapy. J Clin Oncol 20:1167-1174, 2002
Flam M, John M, Pajak T, et al: Role of mitomycin-C in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: Results of a phase III randomized Intergroup study. J Clin Oncol 14:2527-2539, 1996
Macdonald J, Smalley S, Benedetti J, et al: Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 345:725-730, 2001
O'Connell MJ, Martenson JA, Wieland HS, et al: Improving adjuvant therapy for rectal cancer by combining protracted infusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 331:502-507, 1994
Thomas G, Dembo A, Ackerman I, et al: A randomized trial of standard versus partially hyperfractionated radiation with or without concurrent 5-fluorouracil in locally advanced cervical cancer. Gynecol Oncol 69:137-145, 1998
Byfield JE, Calabro-Jones P, Klisik I, et al: Pharmacologic requirements for obtaining sensitization of human tumor cells in vitro to combine 5-fluorouracil or ftorafur and x-rays. Int J Radiat Oncol Biol Phys 8:1923-1933, 1982
Waggoner SD, Fuhrman B, Darcy K, et al: Influence of smoking on progression-free and overall survival in women with stage II-B, III-B and IV-A cervical carcinoma: A Gynecologic Oncology Group (GOG) study. Gynecol Oncol 88:190, 2003 (abstr 71)
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression model and life tables (with discussion). J R Stat Soc B 34:187-220, 1972
Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley & Sons, 1980
National Cancer Institutes Clinical Announcement: Concurrent Chemoradiation for Cervical Cancer. Bethesda, MD, National Cancer Institute, February 1999
Pearcey R, Brundage M, Drouin P, et al: Phase III trial comparing radical radiotherapy with and without cisplatin chemotherapy in patients with advanced squamous cell cancer of the cervix. J Clin Oncol 20:966-972, 2002
Rose PG, Bundy BN: Chemoradiation for locally advanced cervical cancer: Does it help? J Clin Oncol 20:891-893, 2002sdfsfasfasf(Rachelle Lanciano, Alison)
Radiation Oncologist, St Joseph's Hospital/Cancer Institute, Tampa, FL
Gynecologic Oncology Group Statistical and Data Center, Roswell Park Cancer Institute, Buffalo, NY
University of Oklahoma, Oklahoma City, OK
Department of Gynecologic Oncology, University of Texas, Galveston Medical Branch, Galveston, TX
Department of Obstetrics/Gynecology, Indiana University School of Medicine, Indianapolis, IN
Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of California-Irvine Medical Center, Orange, CA
Norton Healthcare Inc, Clinical Pathology Associates, Louisville, KY
ABSTRACT
PURPOSE: Concurrent chemoradiotherapy is the standard of care for locally advanced cervix cancer; the optimal chemotherapy regimen is not yet defined. This trial was designed to compare the outcome of protracted venous infusion (PVI) fluorouracil (FU) with standard weekly cisplatin and concurrent radiation therapy (RT).
PATIENTS AND METHODS: Patients with stage IIB, IIIB, and IVA cervical cancer with clinically negative aortic nodes were eligible. Pelvic RT dose was 45 Gy with a parametrial boost to involved sides of 5.4 to 9 Gy, and high- or low-dose rate intracavitary brachytherapy. Standard therapy was weekly cisplatin 40 mg/m2, and experimental therapy was PVI FU 225 mg/m2/d for 5 d/wk for six cycles during RT.
RESULTS: The study was closed prematurely when a planned interim futility analysis indicated that PVI FU/RT had a higher treatment failure rate (35% higher) and would, most likely, not result in an improvement in progression-free survival compared with weekly cisplatin/RT. The PVI FU/RT arm continues to show a higher risk of treatment failure (relative risk [RR] unadjusted, 1.29) and a higher mortality rate (RR unadjusted, 1.37). There was no difference in pelvic treatment failure between regimens, but there was an increase in the failure rate at distant sites in the PVI FU arm.
CONCLUSION: In this study, PVI FU does not show improved outcome over weekly cisplatin. Future research should explore combinations of FU with cisplatin, new radiosensitizers, and active drugs combined with RT to reduce the high rate of pelvic and distant treatment failure still seen in advanced cervix cancer.
INTRODUCTION
Five randomized trials have demonstrated the superiority of combined chemotherapy with radiation in the treatment of advanced cervix cancer.1-5 The Gynecologic Oncology Group (GOG) has studied both cisplatin/fluorouracil (FU) and weekly cisplatin concurrent with radiation therapy (RT) in two sequential randomized trials and both trials demonstrated a survival benefit for combined chemoradiotherapy compared with hydroxyurea and RT.1,2 Weekly cisplatin was directly compared with FU/cisplatin/hydroxyurea in combination with RT in the most recent GOG trial and there was no difference in progression-free survival (PFS) between the two drug regimens; however, a significant difference in adverse events for the three-drug regimen was demonstrated.2 Therefore, the regimen of weekly cisplatin concurrent with RT was chosen as the standard treatment approach in the GOG for locally advanced cervix cancer with which new combination treatments should be compared.
FU as a single-agent RT sensitizer had not been evaluated for cervix cancer in the GOG. Various combined-modality programs have used infusional FU with RT, especially for GI cancers.6-8 A phase III randomized trial of adjuvant postoperative chemoradiotherapy for high-risk rectal cancer demonstrated an improvement in relapse-free and overall survival for protracted venous infusion (PVI) FU compared to bolus FU in combination with pelvic RT.9 The only previous randomized trial of FU concurrent with RT for cervix cancer found a benefit to FU with improved disease-free survival in subset analysis only (stage IB/IIB or medial stage IIB). Overall, survival and pelvic control were not statistically improved.10 In combination with RT, PVI FU is also appealing for the treatment of cervix cancer because increased drug-RT interactions with tumor cells should occur with potentially increased RT sensitivity and pelvic control. In adenocarcinoma tissue cell lines, cells exposed continuously to FU for 48 hours after RT had maximum RT sensitivity.11 The disadvantage of this approach is the need for central access and specialized equipment to deliver continuous FU, and the potential for increased acute and chronic bowel toxicity. Because it has been a strategy for the GOG to investigate the benefit of chemoradiotherapy and to develop improved regimens, standard weekly cisplatin in combination with RT was compared with the experimental arm of PVI FU in this phase III trial for advanced cervix cancer (GOG 165). This report details the results of that study.
PATIENTS AND METHODS
Patients with primary, previously untreated, histologically confirmed invasive squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma of the uterine cervix stage IIB, IIIB, and IVA were eligible for GOG 165. Histologic diagnoses were confirmed by centralized review by the GOG Pathology Committee. Patients with stage IIIA disease or lower-third vaginal involvement were ineligible because of the individualized implantation techniques required for these patients. Patients must have had negative para-aortic lymph nodes as determined by lymphangiogram, computed tomography (CT), magnetic resonance imaging, or lymphadenectomy. However, this is the first GOG trial for locally advanced cervix cancer that did not require surgical staging of the para-aortic lymph nodes. As is standard for all previous and current cervix studies, patients were required to have adequate hematologic, renal, and hepatic function, as well as performance status, for study entry.
Two biologic correlative studies were done in GOG 165 that required additional laboratory analysis. The first study sought to determine prospectively if smoking behavior is a confounding variable in the effectiveness of RT for locally advanced cervix cancer. This was accomplished through the use of a standard smoking questionnaire, with biochemical validation of smoking status confirmed through urine cotinine analysis, as has been reported recently.12 The second study sought to determine the prognostic value of hemoglobin levels for advanced cervix cancer, with data regarding hemoglobin levels and transfusion history before and during RT collected prospectively and correlated with outcome. No attempt was made to maintain patients at any specific hemoglobin level as part of this protocol. Both correlative studies will be reported separately. This study received local institution board approval and all patients signed an approved, informed consent.
Study Modality: RT
The RT protocol for advanced cervix cancer was updated for GOG 165 by the GOG Radiation Oncology Committee. This included a higher point A dose (85 Gy) with combined external and intracavitary RT, shorter overall treatment time (< 10 weeks), and expanded external-beam field definitions. For the first time in GOG trials, high-dose rate (HDR) brachytherapy was accepted. Physicians planning to treat a patient with HDR brachytherapy were required to complete HDR certification before the patient's enrollment. This required submitting RT dosimetry of two previous cervix cancer patients treated with HDR brachytherapy, to be evaluated by the HDR Radiation Oncology Subcommittee and the Radiologic Physics Center. Whole pelvic field definitions were modified in this protocol to avoid possible marginal misses posteriorly; this was accomplished by placing the posterior border on the lateral field behind the anterior bony sacrum. CT slices through the tumor volume were required and reviewed by the Radiation Oncology Committee in conjunction with simulation films at the semiannual GOG meetings to analyze coverage of the gross tumor volume. The treating physician was required to draw the cervical mass, uterus, and involved pelvic lymph nodes, if any, on the simulation films submitted for review.
External RT dose prescription to the whole pelvis was 45 Gy in 25 fractions at the isocenter. A four-field box arrangement was recommended. Intracavitary RT could include either low-dose rate (LDR; 40 Gy to point A in one to two fractions) or HDR (30 Gy to point A in five fractions) brachytherapy. The type of brachytherapy (HDR v LDR) was chosen at the time of patient random assignment. Tandem and ovoid applicators were recommended for both LDR and HDR brachytherapy, and HDR brachytherapy started the fourth week of external-beam therapy with at least one fraction per week. After completion of external-beam therapy, two fractions of HDR were given per week. A parametrial boost of 5.4 to 9.0 Gy was given to the involved parametrium at the completion of whole pelvic RT.
All charts/simulation films/portal films and pretreatment CT scans were reviewed at biannual GOG meetings by a team of radiation oncologists. A major variation was scored if the dose was 11% to 20% less/more than protocol prescription dose or if elapsed time in days was greater than 20% longer compared with expected time in days for protocol prescription. All CT scans were reviewed and a major variation was scored if field margins were tight on the tumor. If tumor was not satisfactorily included in treatment volume, the fields were deemed unacceptable.
Modifications to RT treatment were described in the protocol and included appropriate delays for neutrophil count less than 1,000 K/μL, platelet count less than 50,000 K/μL or grade 3 to 4 GI or genitourinary toxicity.
Study Modality: Chemotherapy
The standard (arm I) treatment consisted of weekly cisplatin (40 mg/m2 to a maximum of 70 mg) started on day 1 of external RT, preferably 4 hours before RT initiation. Cisplatin could be given during parametrial boost as well as pelvic RT, for a total of six cycles. Cisplatin was reduced by 25% for grade 4 nausea/vomiting or grade 2 neurotoxicity and was discontinued for creatinine greater than 2.0 mg/100 mL or grade 3 to 4 neurotoxicity. Cisplatin was withheld for a WBC less than 3,000 K/μL or a platelet count less than 100,000 K/μL.
The experimental (arm II) treatment consisted of continuous infusion FU 225 mg/m2/d, 5 days per week throughout external-beam whole pelvis and parametrial boost, for a total of 6 weeks. All patients on this arm were required to have central access, and anticoagulation was suggested. Chemotherapy was delayed for WBC less than 2,000 K/μL, platelet less than 50,000 K/μL, or grade 3 diarrhea, and was resumed at full dose when toxicity resolved. For grade 2 mucositis, stomatitis, or esophagitis, or grade 2 weight loss, FU was to be temporarily discontinued and restarted at a dose reduction of 50 mg/m2/d when symptoms subsided.
The GOG Statistical Center randomly assigned the treatment regimen using a fixed block with an equal number of each regimen for each combination of stratification factors: International Federation of Gynecology and Obstetrics stage (IIB, III, IVA), type of brachytherapy (LDR or HDR), and whether patient had undergone para-aortic lymph nodes sampling. A sample size of 416 patients was set, with an estimated 26 months of follow-up, to observe 150 recurrences before testing the primary hypothesis, which was based on the research question of whether concurrent prolonged infusion of FU and RT decreases the risk of progression when compared with concurrent weekly cisplatin and RT. This was a one-sided test with the probability of a type I error set at .05 and the statistical power of 0.80 to detect a hazard ratio (FU to cisplatin) of 0.67. Survival was also a primary end point but no adjustment was made to control the type I error due to the strong correlation between PFS and survival. PFS was defined as the date from protocol registration to the date of reappearance of disease, progression of existing disease, or death, whichever comes first. Survival was defined as the length of life from entry onto protocol to death, or to the date of last contact. Recurrence site was considered local if it was within the pelvic field and distant if it was outside the pelvic field. Distant treatment failure was categorized by site of failure, such as para-aortic or supraclavicular lymph nodes, lung, liver, bone, and so on.
All eligible patients were included in the analysis of survival and PFS (intent-to-treat principle for eligible patients). All causes of death were used to calculate survival, and the estimates of the cumulative proportions of survival were based on Kaplan-Meier procedures.13 Relative risk (RR) estimates and CIs of treatment effects on progression and death, with adjustments for prognostic factors, were accomplished using the Cox model.14 The cumulative incidence curve was used to evaluate the rate of site-specific recurrences between the regimens.15 This method properly handles censored patients and determines the incidence rate of local failure in the presence of failure at distant sites (with/without local relapse), and death due to intercurrent disease.
A formal futility analysis was included in the design of this trial and called for an analysis when 50 recurrences (one third of the final goal) were observed. The stopping rule was to consider study closure if there was any degree of excess rate of recurrence in the PVI FU group using the RR estimate. The RR of progression was 1.35 (90% CI, 0.87 to 2.09). The RR of mortality was 1.02.
RESULTS
The study opened for accrual in October 1997. It was initially a three-arm trial with pelvic irradiation and intracavitary brachytherapy alone as the third arm. When data became available from five randomized trials testing the benefit of concurrent chemoradiotherapy for cervix cancer, the RT-only arm, with 24 patients accrued, was closed in August 1998. The results reported here represent mature data of the two chemoradiotherapy arms. In July 2000, the Data Monitoring Committee (DMC) met and reviewed the interim analysis that included 54 disease progressions and four deaths without progression. The DMC decided to close the trial prematurely because of the increased rate of progression for patients on arm II. The current report is based on 316 eligible patients, of whom 159 were assigned to arm I and 157 were assigned to arm II. Median follow-up of those still alive at the date of analysis is 40.4 months, with quartiles of 33.8, 40.4, and 49.0 months.
Patient characteristics are listed in Table 1. Only 18% of patients had surgically staged para-aortic lymph nodes, which is a dramatic change from prior GOG studies where surgical staging was required. Approximately 14% of patients had either radiographically or surgically determined positive pelvic lymph nodes.
More patients on arm II (72%) than arm I (56%) completed six cycles of chemotherapy (Table 2). However, at least five cycles were received by 74% of patients on the cisplatin arm versus 86% on the PVI FU arm. Among patients treated using HDR, the median dose to point A was 79.8 Gy for the cisplatin arm and 75.0 Gy for the PVI FU arm (prescription protocol dose was 75 Gy). Two patients (5%) treated with HDR brachytherapy were considered to have a major violation but in only one patient was this because of a low dose. Among patients treated using LDR, the median dose to point A was 86.4 Gy for the cisplatin arm and 84.8 Gy for PVI FU arm (prescription protocol dose was 85 Gy). Thirty-eight (15%) patients treated with LDR brachytherapy were considered to have a major violation, but in only eight patients (3%) were this because of low dose. Fourteen patients (4%) failed to complete protocol treatment and were coded as a protocol violation because of low dose. Median total elapsed time for HDR patients was 7.1 weeks and the maximum time was 9.6 weeks to complete RT treatment. Median total elapsed time for LDR patients was 7.3 weeks, with only 10 patients (4%) requiring more than 10 weeks to complete treatment (Table 2).
Interestingly, despite the additional chemotherapy received in the PVI FU arm, grade 3 to 4 toxicities occurred less frequently, including hematologic (10% v 33%) and GI (19% v 25%). Other grade 3 to 4 adverse events were reported in less than 10% of patients. Combining all toxicity classifications, 45% of patients had grade 3 to 4 toxicity during either treatment or follow-up, with a higher rate seen for the cisplatin group (58% v 32%). If reversible hematologic toxicity was excluded from this combined analysis, grade 3/4 toxicity was still higher with cisplatin compared with FU (43% v 27%).
There have been 123 (39%) disease progressions, with an additional 20 deaths without a documented progression, which were included in the computation of PFS. Figure 1 displays the PFS rate through 4 years for the cisplatin group compared with the PVI FU group (57% v 50%; not significant). The unadjusted RR for recurrence was 1.29 (95% CI; 0.93 to 1.80) and 1.25 when adjusting for stage, brachytherapy type, performance of surgical staging, and tumor size (95% CI, 0.90 to 1.74).
There have been 121 deaths reported. There was a difference in OS at 4 years with 36% of patients dead of disease on arm I compared with 45% of patients on arm II (Fig 2) The unadjusted relative risk for death was 1.37 (95% CI, 0.96 to 1.97) and 1.33 when adjusting for stage, brachytherapy type, performance of surgical staging, and tumor size.
Failure rate (cumulative incidence rate) within 4 years confined to the pelvis alone was 16% and 14% in the cisplatin and PVI FU arm, respectively. In the same time frame, distant failure (including abdominal, para-aortic region, bone, liver, and lung) was higher in the PVI FU group (29% v 18%). The higher rate among the PVI FU group was consistent for specific distant sites: lung metastases (9% v 5%) and abdominal failures (11% v 3%). Para-aortic failure occurred in only 7% and 5% of patients in the PVI FU and cisplatin arms, respectively, despite the fact that only 18% of patients were surgically staged in the para-aortic region.
The study was closed by the GOG DMC based on its review of results of the planned interim analysis, which indicated that there was increased risk of progression of 35% in the PVI FU group (ie, unadjusted RR of 1.35 for the PVI FU group compared with the cisplatin group). Under the rules of the interim analysis plan (specifically, futility analysis), the null hypothesis (chemoradiotherapy with PVI FU is not more efficacious than chemoradiotherapy with weekly cisplatin) was accepted. In short, a statistically significant improvement in PFS in the PVI FU group compared with the weekly cisplatin group would not, in all likelihood, be achieved if the study were to be completed; such a deficit was observed after one third of the PFS information (ie, one third of the total progression/deaths required for the completed study analysis) was available.
DISCUSSION
Concurrent chemoradiotherapy is the standard of care for locally advanced cervix cancer. It is employed as definitive treatment for stage IB bulky and higher disease, and as postoperative treatment when nodes or margins are positive or parametrial extension is noted pathologically. This approach was initially reported as a National Cancer Institute consensus statement in 1999, and has gained widespread acceptance.16 The optimal chemotherapy regimen is not yet defined; however, weekly cisplatin2,3 and FU plus cisplatin every 3 weeks1,4,5 are the most popular regimens. The value of weekly cisplatin in conjunction with external pelvic RT as used in this study has been reported previously by Keys et al3 for stage IB bulky cervix cancer. Pearcey et al17 could not confirm this; their contrary result has been discussed extensively in an editorial by Rose and Bundy.18 Rose et al2 compared weekly cisplatin versus cisplatin/FU/hydroxyurea versus hydroxyurea alone, concurrent with pelvic irradiation, and found the cisplatin arms to be comparable and superior to hydroxyurea alone. The weekly cisplatin was better tolerated and therefore became the GOG standard.
There appears to be no benefit to PVI FU as administered in this study over weekly cisplatin for advanced cervix cancer. This lack of benefit was also shown with an alternate FU schedule of 1 g/m2/d during the first and last weeks of external RT compared with RT alone.10 Because this study was closed short of its expected accrual, it is not powered to detect a statistically significant difference between the treatment arms in PFS or OS.
This was the first GOG study to accept clinical staging and HDR brachytherapy for advanced cervix cancer populations. Given that less than 20% of patients on this study had surgical staging and pathologically negative para-aortic lymph nodes, this group of patients should have a higher risk of microscopic involvement of pelvic and para-aortic lymph nodes, and subsequently lower survival rates compared with previous GOG trials. Despite these facts, the outcome for the cisplatin arm in this study is comparable to results of other studies published recently (Table 3; Fig 3). One hypothesis is that the improvement is secondary to the higher RT doses and shorter overall treatment times used in the current study compared with previous GOG trials. Because only 13% of patients in this study received HDR brachytherapy, this technology and its outcome cannot be evaluated.
As administered in this study, PVI FU did not demonstrate statistical improvement over weekly cisplatin as a radiosensitizer. Future research investigating new radiosensitizers and active drugs in combination with radiotherapy to improve the significant pelvic and distant failure rates seen in locally advanced cervix cancer is warranted.
Appendix
The following member institutions participated in this study: University of Alabama at Birmingham, Duke University Medical Center, Abington Memorial Hospital, Walter Reed Army Medical Center, Wayne State University, University of Mississippi Medical Center, Colorado Gynecologic Oncology Group, PC, University of California at Los Angeles, University of Washington, Milton S. Hershey Medical Center, University of Cincinnati, University of North Carolina School of Medicine, University of Iowa Hospitals and Clinics, University of Texas/Southwestern Medical Center at Dallas, Indiana University Medical Center, Wake Forest University School of Medicine, University of California Medical Center at Irvine, Rush-Presbyterian-St Luke's Medical Center, State University of New York Downstate Medical Center, University of Kentucky, The Cleveland Clinic Foundation, State University of New York at Stony Brook, Washington University School of Medicine, Columbus Cancer Council, University of Massachusetts Medical Center, Fox Chase Cancer Center, Medical University of South Carolina, Women's Cancer Center, University of Oklahoma, University of Virginia, University of Chicago, Tacoma General Hospital, Mayo Clinic, Case Western Reserve University, Tampa Bay Cancer Consortium, North Shore University Hospital, Brookview Research Inc, and Ellis Fischel Cancer Center.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. 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.
NOTES
Supported by National Cancer Institute grants to the Gynecologic Oncology Group Administrative Office (Grant No. CA 27469) and the Gynecologic Oncology Group Statistical Office (Grant No. CA 37517).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Whitney CW, Sause W, Bundy BM, et al: Randomized comparison of fluorouracil plus cisplatin vs. 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 17:1339-1348, 1999
Rose P, Bundy B, Watkins E, et al: Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 340:1144-1153, 1999
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 340:1154-1161, 1999
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) 90-01. J Clin Oncol 22:872-880, 2004
Peters WA, Liu PY, Barrett RJ, 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 18:1606-1613, 2000
Minsky B, Pajak T, Ginsberg R, et al: Phase III trial of combined modality therapy for esophageal cancer: High dose versus standard dose radiation therapy. J Clin Oncol 20:1167-1174, 2002
Flam M, John M, Pajak T, et al: Role of mitomycin-C in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: Results of a phase III randomized Intergroup study. J Clin Oncol 14:2527-2539, 1996
Macdonald J, Smalley S, Benedetti J, et al: Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 345:725-730, 2001
O'Connell MJ, Martenson JA, Wieland HS, et al: Improving adjuvant therapy for rectal cancer by combining protracted infusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 331:502-507, 1994
Thomas G, Dembo A, Ackerman I, et al: A randomized trial of standard versus partially hyperfractionated radiation with or without concurrent 5-fluorouracil in locally advanced cervical cancer. Gynecol Oncol 69:137-145, 1998
Byfield JE, Calabro-Jones P, Klisik I, et al: Pharmacologic requirements for obtaining sensitization of human tumor cells in vitro to combine 5-fluorouracil or ftorafur and x-rays. Int J Radiat Oncol Biol Phys 8:1923-1933, 1982
Waggoner SD, Fuhrman B, Darcy K, et al: Influence of smoking on progression-free and overall survival in women with stage II-B, III-B and IV-A cervical carcinoma: A Gynecologic Oncology Group (GOG) study. Gynecol Oncol 88:190, 2003 (abstr 71)
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression model and life tables (with discussion). J R Stat Soc B 34:187-220, 1972
Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley & Sons, 1980
National Cancer Institutes Clinical Announcement: Concurrent Chemoradiation for Cervical Cancer. Bethesda, MD, National Cancer Institute, February 1999
Pearcey R, Brundage M, Drouin P, et al: Phase III trial comparing radical radiotherapy with and without cisplatin chemotherapy in patients with advanced squamous cell cancer of the cervix. J Clin Oncol 20:966-972, 2002
Rose PG, Bundy BN: Chemoradiation for locally advanced cervical cancer: Does it help? J Clin Oncol 20:891-893, 2002sdfsfasfasf(Rachelle Lanciano, Alison)