|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Phase III Trial Comparing Radical Radiotherapy With and Without Cisplatin Chemotherapy in Patients With Advanced Squamous Cell Cancer of the CervixByFrom the Department of Oncology, University of Alberta, Edmonton, and Department of Oncology, University of Calgary, Calgary, Alberta; National Cancer Institute of Canada Clinical Trials Group, Queen’s University, and Kingston General Hospital, Kingston, Ontario; Department of Gynecology Oncology, University of Montreal, and Department of Oncology, McGill University, Montreal; and Hamilton Regional Cancer Centre, Hamilton, Ontario, Canada. Address reprint requests to R.G. Pearcey, MD, Cross Cancer Institute, 11560 University Ave, Edmonton, Alberta, Canada, T6G 1Z2; email: robertpe{at}cancerboard.ab.ca
PURPOSE: To test the hypothesis that cisplatin (CDDP) administered concurrently with standard radiotherapy (RT) would improve pelvic control and survival in patients with advanced squamous cell cancer of the cervix.
PATIENTS AND METHODS: A total of 259 patients with International Federation of Gynecology and Obstetrics stage IB to IVA squamous cell cervical cancer with central disease RESULTS: A total of 253 patients were available for analysis. Median follow-up was 82 months. No significant difference was found in progression-free survival (P = .33). No significant difference in 3- and 5-year survival rates was found (69% v 66% and 62% v 58%, respectively; P = .42). The hazard ratio for survival (arm 2 to arm 1) was 1.10 (95% confidence interval, 0.75 to 1.62). CONCLUSION: This study did not show a benefit to either pelvic control or survival by adding concurrent weekly CDDP chemotherapy in a dose of 40 mg/m2 to radical RT as given in this trial. Careful attention to RT details is important for achieving optimum outcome for patients with this disease.
THE TRADITIONAL TREATMENT of invasive cervical cancer has been by surgery or radiotherapy (RT) or, in certain situations, a combination of both. For more locally advanced disease, with spread beyond the uterus (stage IIB to IVA according to the staging system of the International Federation of Gynecology and Obstetrics), RT is the primary modality of treatment. For these patients, cure rates decrease with advancing stage and tumor bulk.1 A significant reason for treatment failure in many patients is the inability to achieve control over the primary cancer and first-echelon lymph node metastases.2 Improvement in local control can be expected to lead to improved cure rates.3 Simply increasing the dose of RT will increase local control rates, but it does so at the expense of increased complication rates.4 A number of other strategies (eg, hyperbaric oxygen,5 heavy particle therapy,6 hypoxic cell sensitizers,7 neoadjuvant chemotherapy,8 hyperthermia9) have been investigated to determine their value in improving local control rates, but to date, these have not been adopted in routine practice. Another strategy has been the use of cytotoxic chemotherapy concomitantly with RT. Because cisplatin (CDDP) was reported to be the most active drug10 in the treatment of cervical cancer, and because it might also act as a radiosensitizer11 and hypoxic cell sensitizer,12 several phase II trials of concomitant CDDP and RT were conducted with encouraging results.13 To test the hypothesis that concomitant CDDP chemotherapy and RT are better than RT alone, the National Cancer Institute of Canada Clinical Trials Group (NCIC CTG) launched a phase III clinical trial in 1992. Patients with locally advanced squamous cell carcinoma of the cervix were randomized to standard radical RT or the same RT plus CDDP chemotherapy.
Patients with biopsy-proven squamous cell carcinoma of the cervix were entered into the study from 13 cancer centers in Canada and one center in Saudi Arabia. Eligibility was restricted to patients who had stage IB2 and IIA (  5 cm in diameter), IIB, IIIB, IIIA, and IVA disease. Patients with less massive central disease were also eligible if they had histologically positive pelvic lymph node involvement. Patients were included if they were between the ages of 18 to 75 years, had an Eastern Cooperative Oncology Group performance status of 0 to 2, and had signed a consent form approved by their local institutional ethics review board. Other eligibility criteria included the following: an absolute neutrophil count greater than 1.5 x 109/L, a platelet count greater than 100 x 109/L, hemoglobin 110 g/L (after transfusion if necessary), normal serum magnesium level, and serum creatinine less than 140 µm/L. Patients were excluded if their disease had spread outside of the pelvis, if they had a laparotomy for nodal staging, or had received previous treatment for cervical cancer. All patients were staged with a complete history and physical examination, computed tomographic scan of the abdomen and pelvis, chest x-ray, examination while under anesthesia, cystoscopy, and sigmoidoscopy. Stratification was by stage group, as listed in Table 1, and by center. Randomization was performed centrally by a computer-generated series of random numbers to RT with concurrent CDDP chemotherapy (arm 1) or RT alone (arm 2), as described below. The RT dosage was the same in both arms.
RT External-beam megavoltage RT was administered to a clinical target volume that included the primary cancer, uterus, internal iliac, presacral, upper external iliac, and lower common iliac lymph nodes. This was usually achieved by a "four-field box technique," or sometimes a parallel-opposed technique. The usual field borders for anterior and posterior fields were superior at the L5-S1 interspace, inferiorly at the bottom of the obturator foramen, and laterally 1.5 to 2.0 cm lateral to the bony pelvic wall. Lateral fields had the anterior border at the symphysis pubis and the posterior border at the S2-S3 interspace. A dose of 45 Gy was prescribed in 25 equal fractions to the isocenter. Intracavitary brachytherapy followed the external-beam RT. Three different dose rates to point A were permitted as follows: low-dose rate (LDR) (approximately 0.6 Gy per hour), medium-dose rate (MDR) (approximately 1.1 Gy per hour), and high-dose rate (HDR) (> 0.5 Gy per minute). The protocol dose prescription for LDR was 35 Gy to point A in one treatment; for MDR, the dose was 27 Gy in two treatments; and for HDR, the dose was 24 Gy in three treatments. Individual centers used one dose rate throughout the duration of the trial. Because there was stratification by center, there was, therefore, effectively stratification by dose rate also. Every attempt was made to complete treatment within 7 weeks. Central real-time review was required for the external-beam component of the treatment.
Chemotherapy
Statistics An additional five patients were recruited to compensate for five ineligible patients when the accrual goal (248 patients) was reached in the summer of 1996. An interim analysis was performed in March 1997, 6 months after the study was closed. Sixty-seven patients had died by the time of the interim analysis. The analysis was presented to the NCIC CTG Data Safety and Monitoring Committee, which found no reason for early disclosure of the results and recommended final analysis and publication when the prespecified number of deaths had occurred. In fall 1999, 103 patients had died. The decision to perform an analysis was made for two reasons. First, the required number of 110 deaths might be observed at the time when a full analysis was performed for presentation to the American Society of Clinical Oncology. Second, we would have 78% power to detect a hazard ratio of 1.61 with 103 deaths. A full analysis of the data accrued during this trial was performed in April 2000 and presented at the Annual Meeting of the American Society of Clinical Oncology (May 20-23, 2000, New Orleans, LA). At the time of analysis, 104 deaths had occurred; three of the patients who had died were ineligible. This report is based on an analysis of data updated in July 2001. Only one more death from an eligible patient was found in the update, although the medium follow-up time was increased from 64 to 82 months. Overall survival, the primary end point of this study, was defined as the time from randomization to death from any cause. Patients who survived beyond the time of analysis were censored at the time of their last follow-up visit. Progression-free survival (PFS), defined as the time from randomization to the recurrence of cancer in any site, was not a prespecified secondary end point in the protocol, but an analysis of it was performed. Overall survival and PFS of the two treatment groups were described by Kaplan-Meier curves, and the stratified log-rank test adjusting disease stage was the primary method to compare overall and PFS between two treatment groups. Fisher’s exact test was used to compare the rates of toxicity between treatment groups. All the P values reported are two-sided.
Characteristics of Patients Between April 1991 and September 1996, 259 patients were enrolled onto this study. Six patients were considered ineligible, one because of the presence of metastatic disease outside of the pelvis and five because the histology was other than squamous. Two hundred fifty-three patients were available for analysis. The median duration of follow-up was 82 months (range, 6.6 to 122.5 months). Table 2 lists the characteristics of the patients entered into the study. No significant differences in pertinent patient characteristics between the two arms of the study were detected.
Adverse Effects and Compliance With Treatment Protocols One hundred twenty-seven patients were randomized to receive RT plus weekly CDDP (arm 1). In 99 patients (70%), the full dose of CDDP was administered on time. CDDP dose was adjusted in 14 patients (11%), and 108 patients (86%) received more than 90% of the recommended dose of CDDP. Two patients who were randomized to receive CDDP subsequently refused but were included in the CDDP arm in the analysis (intent-to-treat analysis). The full dose of external-beam RT was not completed in only four of 253 patients (three in arm 1 and one in arm 2). This included one patient in arm 1 who refused all treatment, including RT, after randomization. In 18 patients (10 in arm 1 and eight in arm 2), brachytherapy was not possible and an external-beam boost was used. The median duration of external-beam RT was 35 days for patients in both arms, with 90% of the patients in arm 1 being treated between 32 and 41 days and 32 and 42 days in arm 2. The total duration of treatment depended on the brachytherapy dose rate, as illustrated in Fig 1. Most patients were treated at LDR with brachytherapy; 20 were treated at MDR and 38 with HDR. For the patients treated at LDR, the median total duration of treatment was 48 and 50 days, respectively, for arm 1 and arm 2. Ninety percent of patients were treated within the range of 41 to 61 days (arm 1) and in 49 to 62 days (arm 2).
Overall, for all patients together, the median total duration of treatment was 49 days (arm 1) and 51 days (arm 2). The doses (excluding external-beam RT) delivered to point A for patients treated at the three dose rates are listed in Table 3. The acute and delayed toxicities were measured according to the NCIC CTG expanded common toxicity criteria. Table 4 shows the significant ( grade 3) acute toxicity rates. As expected, more toxicity was present in arm 1, but this was not sufficient to cause clinically important treatment delays. The extra-acute toxicity was what would be expected from the known side effects of CDDP. Table 5 lists significant ( grade 3) late toxicity rates. No significant differences existed between the two treatment arms. The only patient whose death was attributed to treatment died as a result of small bowel perforation with peritonitis and septicemia 9 months after completing treatment.
Patient Outcomes Figures 2 and 3 illustrate that there is no significant difference in PFS or overall survival between the two treatment arms. The causes of death for patients in this study are identified in Table 6 and the sites of first recurrence in Table 7. The overall 3- and 5-year survival rates were 69% v 66% and 62% v 58% (arm 1 v arm 2), respectively. The hazard ratio for survival (arm 2 to arm 1) was 1.13 (95% confidence interval, 0.77 to 1.67).
Hemoglobin and Blood Transfusion Hemoglobin level was identified in previous studies as having prognostic importance. Hemoglobin levels were recorded at the start of treatment and at week 5 for 123 (97%) of patients in arm 1 and 107 (85%) of patients in arm 2. The mean initial hemoglobin was 125.9 g/L in arm 1 and 128.5 g/L in arm 2. Thirty-nine patients (20 in arm 1 and 19 in arm 2) required blood transfusions to meet the eligibility criteria for this study. Twenty-three patients (9%) in this study (11 in arm 1 and 12 in arm 2) received a blood transfusion during treatment. Figure 4 illustrates the change in hemoglobin during external-beam RT per treatment arm. Overall, a greater decrease was observed in arm 1. For example, 31% of patients in arm 1 experienced a decrease of hemoglobin of more than 15 g/L, compared with only 20% in arm 2.
This trial was designed to test the hypothesis that CDDP given concurrently with radical RT for the more advanced but localized stages of squamous carcinoma of the cervix would improve pelvic control rates and survival. It is the only large randomized trial to address this simple question. The results of other trials14-18 addressing similar but different questions have led many to the conclusion that CDDP and RT should be the standard treatment for this group of patients.19 Rose et al15 reported on a large three-arm study for patients with stage IIB and above disease where one of the experimental arms was radical RT with CDDP. Patients assigned to the control arm received the same RT with hydroxyurea; there was no RT-alone arm. Whitney et al14 reported on a randomized control trial that compared RT plus hydroxyurea to the same RT plus CDDP and fluorouracil. Again, there was no RT-only arm. Morris et al17 reported on a large study where patients in the experimental arm received pelvic RT plus CDDP and fluorouracil. The standard treatment arm was RT alone, but the RT was different than that provided to the experimental arm in that the treatment portals were extended to include the para-aortic lymph nodes. Keys et al16 reported on a large study involving patients with bulky stage IB2 cervical cancer. In this study, all patients were subject to an extrafascial hysterectomy after preoperative RT (standard arm) or preoperative RT plus weekly CDDP chemotherapy (experimental arm). All four of these studies were positive. A further study18 has shown a survival advantage to chemoradiotherapy as opposed to RT alone as adjuvant treatment in patients treated by radical hysterectomy but with pathologically confirmed poor prognostic factors. The negative result of our trial is therefore unexpected and requires explanation. One possible explanation for the lack of a demonstrable effect of CDDP in this study is that the addition of this drug to RT may only be effective when the RT is protracted. Particularly with respect to the studies reported by Rose et al15 and Whitney et al,14 concern has been raised about the protraction of the radiation treatment. In these studies, the median duration of each treatment course was 62 and 64 days, respectively, compared with 51 days in the study we report here. Each extra day by which the treatment is protracted may lose as much as 1.2% in local control probability, translating into worse survival.20 The benefit of adjuvant treatment can be exaggerated when the primary modality of treatment is suboptimal.21 The importance of optimal primary therapy is highlighted by the fact that in this trial, the only significant prognostic factors found on multivariate analysis were treatment duration and whether or not brachytherapy was used. In the study reported by Morris et al,17 RT was delivered without significant protraction, but the study was a positive one. This raises the possibility that the addition of fluorouracil to CDDP and RT may be the more important factor in improving survival. The importance of hemoglobin as a prognostic factor in patients treated by RT for cervical cancer has been appreciated for a number of years,22 but the magnitude of this effect and the particular importance of a decreasing hemoglobin during treatment was quantified by Grogan et al23 more recently. Although it is not possible to conclude from this study that chemotherapy-induced anemia could have the same prognostic significance, it should be noted that there was a significantly greater decrease in hemoglobin levels in the chemoradiotherapy arm of the study reported here than in the RT-only arm. This difference could have accounted for as much as an 8% reduction in survival in the experimental arm on the basis of the data of Grogan et al. In the studies reported by Rose et al15 and Whitney et al,14 the standard arm included hydroxyurea in a dose of 80 mg/kg twice weekly during the external-beam RT to a maximum dose of 6,000 mg. This was likely at least as myelosuppressive as the CDDP chemotherapy in the experimental arm. The experimental arm in the study by Morris et al17 included para-aortic RT as opposed to pelvic RT in the experimental arm. This treatment would have affected the lumbar vertebral bone marrow and might have led to equal decreases in the hemoglobin level in both arms also. Unfortunately, none of these studies reports hemoglobin levels, so we are unable to comment further on this hypothesis. In two of the positive studies reported14,15 and in a proportion of the patients reported by Morris et al,17 surgical staging was used to rule out para-aortic lymph node metastases. This is different from the study reported here, where para-aortic lymph nodes were evaluated by computed tomographic scan only and para-aortic nodal RT was not used. It could be argued that the survival advantage apparent in other studies was lost in ours because of a large number of patients with subclinical para-aortic nodal disease not effectively detected or treated. However, this is unlikely to be the case because we were unable to show any difference in pelvic control rates. Also, the survival for the patients in our control arm is better than or equivalent to that reported in other, similar studies. Other studies have included patients with adenocarcinoma and adenosquamous carcinoma. Our study was limited to patients with squamous cell carcinoma. This histology was likely the most represented in other studies (> 80%). In addition, it is unlikely that other histologies are substantially more sensitive to CDDP chemotherapy than squamous cell carcinomas. The strength of this study includes the low number of ineligible patients (six) and the low rate of protocol violations. Only 2.8% of patients were found to have a major protocol violation. We think that the real-time portal imaging review and the review of the eligibility criteria at the time of randomization were responsible for this. Finally, this trial was designed to detect a 15% improvement in 5-year survival with 80% power. As more clinical trials investigating the value of chemoradiotherapy in cervical cancer are reported, some may be negative by chance alone. This trial did not have sufficient power to reliably detect differences of less than 15%. For example, there is a 46% chance that this trial could have missed a real improvement in 5-year survival from 56% to 66% with a one-sided .05-level test. From these results, it can also be concluded that there is a 95% chance that CDDP added to RT as delivered in this trial does not improve 5-year survival by more than 12%. Certain subgroups of patients within the range of those studied in this and other trials may benefit more than others, but we did not detect this on subgroup analysis. It is inevitable that the results of this trial will be included in meta-analysis. If it is thought that this trial is negative by chance variation because of its lack of power to detect deficiencies less than 15%, then it would be entirely reasonable to pool these results with others to get the best point estimate of treatment effect. If, on the other hand, it is accepted that some other trials have suboptimal control arms (hydroxyurea on protracted treatment times), then pooling the trial results in a meta-analysis would be inappropriate. Despite the apparent negative result of this trial, we think that the balance of evidence favors the use of combined-modality treatment for the types of patients studied in this trial. The best results are certainly achieved by careful attention to RT details, including dose and overall delivery time, the use of brachytherapy whenever possible, and probably the addition of concurrent CDDP chemotherapy to RT.
Supported in part by the National Cancer Institute of Canada/Canadian Cancer Society.
1. Perez C, Breaux S, Madoc-Jones H, et al: Radiotherapy alone in the treatment of carcinoma of the cervix: Analysis of tumor recurrence. Cancer 51: 1393-1402, 1983[CrossRef][Medline] 2. Jampolis S, Andros J, Fletcher G: Analysis of sites and causes of failure of irradiation in invasive squamous cell carcinoma of the intact uterine cervix. Radiology 115: 681-685, 1975[Abstract] 3. Suit H: Assessment of the impact of local control on clinical outcome. Front Radiat Ther Oncol 29: 17-23, 1996[Medline] 4. Sinistrero G, Symandi P. Rumore A, et al: Analysis of complications of cervix carcinoma treated by radiotherapy using the Franco-Italian glossary. Radiother Oncol 26: 203-211, 1993[CrossRef][Medline]
5.
Watson ER, Halnan KE, Dische S, et al: Hyperbaric oxygen and radiotherapy: A medical research council trial in carcinoma of the cervix. Br J Radiol 51: 879-887, 1978 6. Marales P, Hassey DH, Maor MH, et al: Preliminary report of the M.D. Anderson hospital randomized trial of neutron and photon irradiation for locally advanced carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 7: 1533-1540, 1981[Medline] 7. Bauer M, Liebel S, Wasserman T, et al: Effect of misonidazole dose on survival in patients with stage IIIB–IVA squamous cell carcinoma of the uterine cervix on RTOG randomized trial. Int J Radiat Oncol Biol Phys 12: 1101-1103, 1986[Medline] 8. Souhami RL, Gil RA, Allan SE, et al: A randomized trial of chemotherapy followed by pelvic radiation therapy in stage IIIB carcinoma of the cervix. J Clin Oncol 9: 970-977, 1991[Abstract] 9. Hornbach NB, Shupe RE, Shidnia H, et al: Advanced stage IIIB cancer of the cervix treatment by hyperthermia and radiation. Gynecol Oncol 23: 160-167, 1986[CrossRef][Medline] 10. Thigpen T, Shingleton H, Homesley H, et al: Cisplatinum in the treatment of advanced or recurrent squamous cell carcinoma of the cervix. Cancer 48: 899-903, 1981[CrossRef][Medline] 11. Szumiel I, Nias AHW: The effect of combined treatment with a platinum complex and ionizing radiation on Chinese hamster ovary cells in vitro. Br J Cancer 33: 450-458, 1976[Medline] 12. Douple EB, Richmond RC: Radiosensitization of hypoxic tumour cells by cis- and trans-dichlorodiamineplatinum (II). Int J Radiat Oncol Biol Phys 5: 1369-1372, 1979[Medline]
13.
Pearcey RG, Mohamed IG, Hanson J: Treatment of high risk cervical cancer. N Engl J Med 341: 695-696, 1999
14.
Whitney CW, Sause W, Brundy BN, et al: A randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjuvant to radiation therapy in stages 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
15.
Rose PG, Bundy BN, Watkins EB, et al: Concurrent cisplatin based chemoradiation improves progression free and overall survival in advanced cervical cancer: Results of a randomized Gynecologic Oncology Group study. N Engl J Med 340: 1137-1143, 1999
16.
Keys HM, Bundy BM, Stehman FB, et al: A comparison of weekly cisplatin during radiation therapy versus irradiation alone each followed by adjuvant hysterectomy in bulky stage 1B cervical carcinoma: A randomized trial of the Gynecologic Oncology Group. N Engl J Med 340: 1154-1161, 1999 17. Morris M, Eifel PJ, Lu J, et al: Pelvic radiation with concurrent chemotherapy versus pelvic and para-aortic radiation for high risk cervical cancer: A randomized Radiation Therapy Oncology Group study. N Engl J Med 340: 1137-1143, 1999
18.
Peters WA III, 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 19. National Cancer Institute: Clinical Announcement: Concurrent Chemoradiation for Cervical Cancer. Washington, DC, United States Department of Public Health, February 1999 20. Fyles AW, Pintilie M, Kirkbride P, et al: Prognostic factors in patients with cervix cancer treated by radiation therapy: Results of a multiple regression analysis. Radiother Oncol 35: 107-117, 1995[CrossRef][Medline] 21. Pearcey RG, Stuart GC, MacLean GD, et al: Phase II study to evaluate the toxicity and efficacy of concurrent cisplatin and radiation therapy in the treatment of patients with locally advanced squamous cell carcinoma of the cervix. Gynecol Oncol 58: 34-41, 1995[Medline] 22. Bush RS: The significance of anaemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 12: 2047-2050, 1986[Medline] 23. Grogan M, Thomas GM, Melamed I, et al: The importance of hemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 86: 1528-1536, 1999[CrossRef][Medline] Submitted February 21, 2001; accepted October 16, 2001.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
