Platinum-based adjuvant chemotherapy for early-stage epithelial ovarian cancer: single or combination chemotherapy?

Transcription

1 DOI: /j x Gynaecological oncology Platinum-based adjuvant chemotherapy for early-stage epithelial ovarian cancer: single or combination chemotherapy? G Adams, a J Zekri, a H Wong, a J Walking, b JA Green a a Clatterbridge Centre for Oncology NHS Foundation Trust, Bebington, Wirral, UK b Wirral University Teaching Hospitals, Upton, Wirral, UK Correspondence: Dr G Adams, Clatterbridge Centre for Oncology NHS Foundation Trust, Clatterbridge Road, Bebington, Wirral, CH63 4JY, UK. Gerry.Adams@ccotrust.nhs.uk Accepted 9 May Published Online 18 June Objective To evaluate the clinical benefit and toxicity of two regimens; single agent carboplatin (C) and a carboplatin/paclitaxel (CP) combination in early epithelial ovarian cancer. Design A retrospective review. Setting Single cancer centre serving a population of 2.1 million in the northwest of England. Population All women treated with adjuvant chemotherapy for International Federation of Obstetrics and Gynecology stage Ia IIc ovarian cancer between 2002 and Methods Case and operation notes were reviewed. Details of the surgery performed, performance status (PS), tumour histology, stage, grade, intended chemotherapy, chemotherapy received, acute and late toxicity, relapse and death were all recorded. Main outcome measures Overall survival (OS), relapse-free survival (RFS), acute and late toxicity. Results Sixty women received CP and 35 received C. Younger women (P < ) and those with a better performance status (P = 0.045) were more likely to receive CP. Median follow up was 38 months (range 0 70). Five-year OS was 62% (95% CI 44 81%) for C and 73% (95% CI 61 85%) for CP P = For the subgroup with stage I disease and good PS (0/1) 5-year OS was 80% (59 100%) for C and 79% (63 95%) for CP; P = 1.0. For those with stage 2 disease, 5-year OS was 29% (95% CI 0 62%) for C and 63% (95% CI 44 82%) for CP; P = Subgroup analyses by grade or histology showed no difference in OS. P was discontinued prematurely in nine (15%) women on account of toxicity, whereas C was not stopped early. P-related neuropathy (G1/2) was reported in ten (17%) women at 6 months and in two (3%) at 1 year. Conclusions Combination therapy is administered more often than carboplatin; especially in those with younger age, better PS and nonmucinous histology. Recurrence and death rates were similar with both treatments. Well-designed trials are needed to identify the optimum chemotherapy regimen in this group. Keywords Adjuvant chemotherapy, carboplatin, early epithelial ovarian cancer, paclitaxel, platinum, taxane. Please cite this paper as: Adams G, Zekri J, Wong H, Walking J, Green J. Platinum-based adjuvant chemotherapy for early-stage epithelial ovarian cancer: single or combination chemotherapy? BJOG 2010;117: Introduction Epithelial ovarian cancer (EOC) is the fourth commonest cancer in women. Each year there are over 6000 new cases and around 4000 deaths in England and Wales. 1 Initial management involves complete staging by optimal surgical resection. Seventy-five percent of women will be found to have disease that has spread beyond the pelvis (stages III or IV). For these women the optimal chemotherapy regimen is carboplatin/paclitaxel (CP). 2 For the 25% of women with cancer confined to the pelvis (stages I and II) the optimal treatment is more controversial. Five-year survival rates of 90% have been reported for stage Ia, falling to 80% for Ic and 65% for IIc. 3 A retrospective review of results from six centres has shown that, for stage I disease, degree of differentiation but not histological subtype was a predictor of outcome. 4 Women whose tumours are moderately and poorly differentiated are three and nine times more likely to relapse compared with those with well-differentiated tumours. ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology 1459

2 Adams et al. An early trial 5 suggested that after comprehensive surgical staging, women with well-differentiated and moderately well-differentiated tumours confined to the ovary (stage Ia or Ib) have an excellent prognosis and do not benefit from adjuvant chemotherapy with melphalan. This subgroup has been defined by some as low-risk early epithelial ovarian cancer. Some (but not all), subsequent adjuvant chemotherapy trials have excluded this low-risk group. Two small randomised trials 6,7 failed to show any benefit from adjuvant platinum-based chemotherapy in women with stage 1 disease after surgery. However, these studies were underpowered to detect clinically meaningful differences. Subsequently, two large, multinational trials of platinum-based adjuvant chemotherapy for high-risk EOC were set up. Combined analysis of 925 women in the Adjuvant ChemoTherapy In Ovarian Neoplasm (ACTION); and The International Collaborative Ovarian Neoplasm 1 (ICON 1) trials 8 10 reported an improvement of 8% in overall survival (OS) and 11% in recurrence-free survival (RFS) at 5 years of follow up for those receiving adjuvant chemotherapy. There were some differences in the entry criteria for these trials. The ACTION trial included women with stage Ia to IIa disease excluding the low-risk group defined as well-differentiated stage Ia or Ib tumours. The ICON 1 trial defined entry based on clinician uncertainty as to the benefit of adjuvant chemotherapy. As such there was likely to be a significant proportion of women in ICON 1 who would be classed as low risk, but also some with stage IIb and IIc disease; both groups being excluded from ACTION and other trials. Nevertheless, it was felt that overall the trials were similar enough to be analysed in combination. These five trials have recently been combined in a Cochrane meta-analysis 11 of the role of adjuvant chemotherapy after surgery for women with early EOC (defined here as stage I and II disease). It concluded that adjuvant platinum-based chemotherapy is effective in prolonging the survival of the majority of women who are assessed as having early stage epithelial ovarian cancer. It also stated that in women with well-differentiated tumours who are optimally staged it is safe to withhold chemotherapy. In a similar vein, the 3rd International Gynecologic Cancer Intergroup Ovarian Cancer Consensus Conference (GCIG OCCC 2004) 12 recommended at least carboplatin (C) area under curve as adjuvant therapy for those with early-stage EOC [defined here as International Federation of Obstetrics and Gynecology (FIGO) stage I IIa] where chemotherapy is indicated. However, the optimum regimen and indeed the precise definition of high-risk EOC both remain uncertain. The recent GOG trial used three cycles of CP as the experimental arm, with six cycles of CP classified as the standard regimen. Early-stage ovarian cancer that would benefit from adjuvant chemotherapy was defined in this study as completely resected stages I and II excluding stages Ia and Ib with well-differentiated AND moderately welldifferentiated tumours. There was no difference in survival between regimens. Current National Comprehensive Cancer Network (NCCN) guidelines 14 recommend three to six cycles of a carboplatin/taxane combination when adjuvant chemotherapy is recommended in high-risk early-stage EOC (defined as stage I only excluding stage Ia/b welldifferentiated only). To examine practice and outcomes in our network we undertook a retrospective analysis of all women receiving adjuvant chemotherapy for FIGO stage I/II EOC. Surgery was performed at one of three gynaecology centres in the region by gynaecologists experienced in cancer surgery. All cases were discussed in one of two specialist gynaecology oncology multidisciplinary team (MDT) meetings in the region. All chemotherapy was provided by one of two consultant oncologists. In most cases chemotherapy commenced within 2 months of the date of surgery. Local protocol at that time stated that if it was felt that adjuvant chemotherapy was indicated for early-stage EOC, then six cycles of carboplatin either alone or in combination with paclitaxel should be used. Methods All women from Merseyside and Cheshire registered with a new diagnosis of EOC stage Ia to IIc and who received their first dose of chemotherapy between 1 January 2002 and 31 December 2005 were identified. Stage was defined by the MDT, which took into account all factors including operation notes; the opinion of the surgeon; histopathology reports; cytology reports; and the appropriate preoperative or postoperative radiological investigations (usually computed tomography). Operation notes and MDT reports were obtained. Details of the surgery performed as well as documentation of residual disease were recorded. Case notes were reviewed and details of performance status (PS), tumour histology, stage, grade and chemotherapy regimen (C or CP) were recorded. The dose of carboplatin was area under the curve (AUC) 5 or 6 according to the Calvert formula. 15 The dose for paclitaxel was 175 mg/m 2 over 3 hours. Both drugs were given on day one of a 3-week schedule. Also recorded were details of chemotherapy received, haematological and biochemical profiles, acute and late toxicity, relapse and death. Women were followed up at regular intervals typically 3-monthly for the first 2 years followed by 6-monthly in year three and annually thereafter until discharge (usually at the end of year five if no signs of recurrence). The RFS was calculated from the date of first cycle of first-line chemotherapy to the first radiological evidence of recurrence and OS was from the 1460 ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology

3 Chemotherapy for early EOC, single or combination? date of first chemotherapy to death or last follow-up. Data were inspected on 1 June 2008 for those who had neither relapsed nor died. Tests of statistical significance were log rank for survival, chi-square test for categorical and Mann Whitney U test for continuous data. Results From 1 January 2002 to 31 December women were identified, 60 receiving CP and 35 receiving C. For 83 women the MDT recorded complete surgical resection. In the remaining 12 the surgeon could not guarantee complete removal (typically because of stage II disease that could not all be removed safely). Of these 12 women, seven received CP and five received C. The five women receiving monotherapy were all more than 70 years old. Table 1 shows the breakdown of chemotherapy administered according to woman and tumour characteristics. Median age was 63 years, interquartile range (IQR) years. The group receiving CP was significantly younger than that receiving monotherapy. The median age for the CP group was 59 years (IQR 52 65) compared with 73 years (IQR 62 78) for the C group (P < 0.001). Women with better performance status (PS 0 or 1) were also more likely to receive CP than those with PS 2 or 3 (P = 0.045). Of the 63 women with stage I disease, 28 (44%) received C and 35 (56%) received CP. Of the thirty-two with stage II disease, seven (22%) received C and 25 (78%) received CP. The distribution of histological groups was serous 39%, endometrioid 30%, clear cell 23% and mucinous 8%. Women with serous, endometrioid and clear cell were more likely to receive CP, whereas those with mucinous tumours were more likely to receive C monotherapy (P = 0.019). Tumour grade influenced regimen choice less. Eighteen of 28 (64%) women with grade 3 tumours received CP and 22 out of 42 (52%) with grade 1 and 2 also received CP (P = 0.16). During the follow-up period 11 (31%) and 20 (33%) women relapsed in the C and CP groups, respectively. Five of the 11 (45%) who relapsed after C were able to receive second-line chemotherapy. Of the 20 who relapsed after Table 1. Breakdown of chemotherapy delivered according to woman and tumour characteristics Carboplatin (n = 35) Carboplatin Paclitaxel (n = 60) Total (n = 95) Age: Median (IQR) 73 (62 78) 59 (53 65) 63 (54 72) < Histology Serous % % % Endometrioid % % % Clear cell % % % Mucinous % 2 3.3% 8 8.4% Grade % 5 8.3% % % % % % % % Clear cell % % % Missing 0 0.0% 3 5.0% 3 3.2% Stage Ia % % % Ib 1 2.9% 1 1.7% 2 2.1% 1c % % % II (substage missing) 0 0.0% 1 1.7% 1 1.1% IIa 1 2.9% 3 5.0% 4 4.2% IIb 3 8.6% % 9 9.5% IIc 3 8.6% % % Performance status % % % % % % % 3 5.0% 7 7.4% % 0.0% 2 2.1% Missing 3 8.6% 3 5.0% 6 6.3% ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology 1461

4 Adams et al. CP, 12 (60%) were able to have second-line chemotherapy. Eleven (31%) and 15 (25%) women died in the C and CP groups. After a median follow up of 38 (0 70) months, 5-year OS and RFS for all women were and 70% and 56%, respectively. There was no difference in 5-year OS and RFS between the two regimens (Figure 1). The OS for C was 62% (95% CI 44 81%). That for CP was 73% (95% CI 61 85%); P = The RFS for C was 54% (95% CI 27 81%) and for CP was 57% (95% CI 38 76%); P = (Figure 1B). Median survival was not reached for either OS or RFS. Figure 2A shows that for the 35 women with stage 1 disease who received CP, 5-year OS was 80% (95% CI 65 95%) compared with 70% (95% CI 49 91%) for the 28 receiving C; P = For RFS the rates were 61% (95% CI 0 100%) for CP and 78% (95% CI 60 P = 95%) for C; P = 0.97 (curve not shown). For stage 2 disease, OS was 63% (95% CI 44 82%) for those who received CP (Figure 2B); compared with 29% (95% CI 0 62%) for those receiving C; P = For RFS the rates were 53% (95% CI 32 73%) for CP and 14% (95% CI 0 40%) for C; P = (curve not shown). For the subgroup of 53 women with stage 1 disease and good performance status (PS 0 or 1) there was no difference in survival with either regimen (Figure 3). For the 33 who received CP OS was 79% (95% CI 63 95%) compared with 80% (95% CI %) for the 20 receiving C; P = 1. For RFS the rates were 60% (95% CI 34 85%) for CP and 76% (95% CI 55 97%) for C; P = 0.9 (curve not shown). Further analysis did not reveal any differences in either OS or RFS for any of the subgroups of grade or histology (data not shown). Both regimens were well tolerated; Table 2 summarises the common toxicities recorded. Eighty-nine (94%) out of A Overall survival B RFS Figure 1. (A) Overall survival; (B) relapse-free survival. A Overall survival Stage = 1 B Overall survival Stage = Figure 2. (A) Overall survival stage 1; (B) overall survival stage ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology

5 Chemotherapy for early EOC, single or combination? OS - PS 0-1 & Stage I - Figure 3. Overall survival in women with good performance status and stage I disease. 95 women completed the full number of cycles. Two women in each group stopped because of disease progression. One woman receiving C died after cycle 1 for an unknown reason and one woman receiving CP was found to have a second primary lung cancer after two cycles. None of the C group received fewer than six cycles because of toxicity. P was omitted prematurely in nine (15%) women in the CP group (four because of a drug reaction, four because of neuropathy and one not stated). Six (17%) of the C group had the dose of carboplatin reduced. In the CP group four (6%) had their carboplatin dose reduced and three (5%) had their paclitaxel dose reduced. One case of grade 3 neutropenia and none of grade 4 were recorded. This woman was receiving CP. She was asymptomatic and did not require any additional treatment. Two hospital admissions related to chemotherapy were recorded; one was the result of an infection (not in a neutropenic woman) and the other was because of vomiting and dehydration. Both women were being treated with CP. Persistent P-related neuropathy was reported in ten (17%) women at 6 months and in two (3%) women at 12 months. Discussion Adjuvant platinum-based chemotherapy has now become the standard of care for women with high-risk early EOC. However, there are controversies about how this high-risk group is defined as well as the most appropriate chemotherapy regimen. The 2004 consensus statement from the GCIG 12 and the recent Cochrane meta-analysis 11 both make statements advocating the use of platinum-based chemotherapy. However, the current NCCN 14 guidelines make a specific recommendation for three to six cycles of a carboplatin/taxane combination. This is despite the fact that only 6% of women in the combined ACTION/ICON 1 trial 8 10 and none of the women in the other two trials 6,7 making up the meta-analysis received taxanes. The majority of women received carboplatin monotherapy (57% in ACTION/ICON 1 and all of the women in the study by Trope et al. 7 ). The others received either cisplatin or cisplatin combinations that have been shown to be equivalent to carboplatin monotherapy. In our series, combination therapy with CP was used twice as often as carboplatin monotherapy. Five-year OS was 10% greater in the CP group [73% (95% CI 61 85%) versus 62% (95% CI 44 81%)]. In this small sample the difference did not reach statistical significance (P = 0.316). The difference for 5-year RFS was even less with only 3% in favour of the combination [57% (95% CI 38 76%) versus 54% (95% CI 27 81%), P = 0.661]. Table 2. Toxicity summary Toxicity Carboplatin Paclitaxel Carboplatin Paclitaxel Anaemia Sample Grade 1/2 Grade 3/4 Grade 1/2 Grade 3/4 15 (40%) 2 (8%) 19 (47.5%) 1 (2.5%) Neutropenia Sample Grade 1/2 Grade 3/4 Grade 1/2 Grade 3/4 7 (50%) 0 (0%) 18 (53%) 1 (3%) Thrombocytopenia Sample Grade 1/2 Grade 3/4 Grade 1/2 Grade 3/4 15 (60%) 2 (8%) 22 (55%) 0 (0%) Neuropathy (6 months) 0 (0%) 0 (0%) 10 (17%) 0 (0%) Neuropathy (12 months) 0 (0%) 0 (0%) 2 (3%) 0 (0%) ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology 1463

6 Adams et al. This is a nonrandomised study and is influenced by bias. We have shown that both age and PS strongly influence choice of regimen with younger and fitter women more likely to receive CP. As these factors are independent positive predictors for survival, selection bias alone could explain the better outcome with CP. It must also be recognised that stage is likely to influence regimen choice although this was less so within our study. More advanced stage is a negative predictor of survival and if it biases towards selecting combination chemotherapy this would have the effect of reducing any real benefit of CP over C in our study. To try and minimise the effect of these selection biases we looked at outcomes in small subgroups. The subgroup with good PS and stage I disease (Figure 3) shows absolutely no difference in 5-year OS between the two regimens. For those with stage II disease (Figure 2B) there was a statistically significant improvement in 5-year OS for the combination [63% (95% CI 44 82%) versus 29% (95% CI 0 62%), P = 0.025]. However one must be cautious about interpreting this as evidence for the benefit of CP in stage II disease as only seven women with stage II disease received C and this is likely to be strongly influenced by the poor prognostic features of advanced age and poor performance status. We have not commented on the subgroup with poor PS who received chemotherapy because this group was small and CP would be an inappropriate choice. A potential criticism of our study is the low numbers and relatively short follow up. Early-stage EOC is relatively uncommon and taxanes were not used routinely by us before Further time to allow the data to mature will be beneficial. Our data, along with the recent GOG-157 trial 13 and other case series, 16,17 show that CP is frequently used as the standard of care for early EOC albeit sometimes with an abbreviated course of three or four cycles. Yet given the lack of published category 1 evidence supporting the use of CP in early EOC why is it that CP is recommended 14 and used so frequently? The answer may be concerns about suboptimal surgical staging. The GOG definition of comprehensive surgical staging 13 includes total hysterectomy, bilateral salpingooophorectomy, resection of all gross disease, aspiration of free peritoneal fluid, peritoneal washings for cytology, infracolic omentectomy, selective bilateral pelvic and aortic node dissections, peritoneal biopsies from four pelvic locations and bilateral paracolic areas, and right diaphragm cytology or biopsy. The ACTION trial 9 also tried to prospectively ascertain the extent of surgery in a clinical trial setting. While the trial protocol strongly recommended optimal staging using a similar definition to the GOG; it found that only 151 (34%) women were optimally staged. It is known that if surgery is less than comprehensive there is a chance of missing more advanced disease. In 1983, a series of 100 women with apparent early-stage EOC underwent a second, comprehensive surgical staging procedure at which 23 were found to have stage III disease. 18 Although there have been no randomised trials showing that a more aggressive surgical approach improves survival; it is argued by some 19 that comprehensive surgical staging is critical in all women with early-stage ovarian cancer. It may negate the need for chemotherapy for some because subset analysis of the 151 (34%) optimally staged women in the ACTION trial 9 showed no benefit of adjuvant chemotherapy. It may also alter the prognosis for others (up to 23%) who have stage III disease. Worldwide the majority of women labelled as having early-stage EOC will have undergone less than comprehensive surgery. Some of these will have occult stage III disease. It may be a desire not to undertreat women with occult stage III disease that explains the use of CP in early EOC. However this would subject the majority to overtreatment. A balance must be struck between providing the best chance of long-term survival and minimising long-term toxicity. The addition of paclitaxel to carboplatin increases the toxicity for all women but with no gain for the many who are already cured. For those who do relapse, quality of life in the time to first relapse will be worse in those treated with CP. It is known that the risk of paclitaxel-related neuropathy increases with cumulative dose 20 and if paclitaxel is used in the initial regimen, our ability to administer sufficient doses to women who relapse will be reduced. This additional toxicity could be justified if there was an overall survival benefit. However, in modern clinical practice (with routine computed tomography scanning) the number of women with occult stage III disease is likely to be less than the 23% found on surgical restaging before Moreover, there is some controversy about the benefit of carboplatin plus paclitaxel over carboplatin alone even in advanced disease. Some guidelines 2 recommend CP as the optimum first-line chemotherapy in advanced disease. This is based on analysis of four studies comparing the addition of paclitaxel to platinum-based chemotherapy However, the largest individual trial (ICON 3), 23 was the only one to directly compare carboplatin with carboplatin plus paclitaxel. 20% of the population actually had stage I or II disease. There was no benefit in survival either in the trial as a whole or in the 413 women with early-stage disease. It may be that in ovarian cancer carboplatin monotherapy is superior to other platinum-based regimens and the addition of paclitaxel is not necessary even in advanced disease. Therefore the only benefit from comprehensive surgical staging may be in selecting the small group of well-differentiated stage 1a/b tumours where there may be little benefit from adjuvant chemotherapy ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology

7 Chemotherapy for early EOC, single or combination? In our network, comprehensive staging to GOG or ACTION standards has not been adopted as routine practice because this has not been shown to improve outcome and could potentially result in excessive morbidity. The number of women undergoing optimal surgical staging in our sample is likely to be very small (if any). We identified only women who received adjuvant chemotherapy and so cannot say how many women in our network were not offered chemotherapy because they had optimally staged well-differentiated stage 1a/1b tumours. However, all operations were performed by experienced surgeons in centres with a high level of experience in gynaecological oncology surgery. Moreover all cases were thoroughly discussed in an MDT and so we are confident that all surgery could be defined as at least adequate according to ACTION study standards 9 that is, at least total hysterectomy, bilateral salpingo-oophorectomy followed by inspection and palpation of all peritoneal surfaces and the retroperitoneal area; biopsies of any suspect lesions for metastases; peritoneal washing and infracolic omentectomy. This is likely to be standard surgical practice worldwide. Our series shows that combination therapy is safe. Although 15% of women had paclitaxel stopped early because of neuropathy and acute allergic reactions, all women completed their course of carboplatin. Persistent low-grade neuropathy was recorded in 17% and 3% of women treated with CP at 6 and 12 months, respectively. With its retrospective nature, it is likely that our series under-reports adverse effects. However, it is known that acute neuropathy is a significant source of toxicity with CP. In a large trial of CP versus cisplatin/paclitaxel chemotherapy for women with advanced ovarian cancer 25 75% of those receiving CP reported sensory peripheral neuropathy during chemotherapy, with 28% reporting grade 2 and 7% reporting grades 3 or 4. We have reported rates of persistent neuropathy (17% at 6 months) that are not well addressed by larger trials. It should also be noted that the two hospital admissions related to chemotherapy and the case of grade 3 neutropenia were all in women receiving combination therapy, which indicates a higher toxicity burden in the combination arm. Our data are compared with those from published studies in Table 3. It should be noted that we had no exclusion criteria. Our series also contains more women with higher stage disease than some other studies. In our series the women who had stage I disease and good performance status (PS 0/1) are likely to be more comparable to each other and to other series. Figure 3 shows that their survival was the same whichever regimen was used and Table 3 shows that this group s survival compares well to other published series. In our study there were 12 women classed as having incomplete resection. These were typically women with stage II disease with tumour stuck down in the pelvis either from rupture or associated infection. It cannot Table 3. Published studies of adjuvant chemotherapy in early-stage ovarian cancer adapted and updated Study Study type Stage Systemic therapy Women 5-year RFS 5-year OS Young, Prospective, Randomised StageIG3, Stage II Melphalan, 32P 68, 73 80% 80% 81% 78% Bolis, Prospective, Randomised Stage Ia/b G2/3 Cisplatin, Observation 41, 44 83% 65% 88% 82% Bolis, Prospective, Randomised Stage Ic Cisplatin, 32P 82, 79 85% 65% 81% 79% Trope, Prospective, Randomised Stage I Carboplatin, 81, 81 70% 71% 86% 85% Observation Trimbos, Prospective, Randomised Ia-IIa except Ia/b G1 Platinum, Observation 224, %** 68% 85% 78% ICON 1, Prospective, Randomised Uncertain benefit Platinum, Observation 241, %*** 62% 79%* 70% immediate chemotherapy Bell, Prospective, Randomised I and II except Ia/b CP 6CP 3 225, % 75% 83% 81% G1/2 Bamias, Retrospective, I IIb except Ia/b G1 CP % 87% Nonrandomised Shimada, Retrospective, I II CAP 3orCP 3 69, 31 92% (total group) Nonrandomised Present study Retrospective, Nonrandomised Ia IIc CP 6C 6 60, 35 57% 54% 73% 62% Present study (subgroup PS0/1 stage 1) Retrospective, Nonrandomised CAP, cisplatin 50 mg/m 2, doxorubicin 40 mg/m 2 and cyclophosphamide 400 mg/m 2. *P = 0.03, **P = 0.02, ***P = Ia Ic CP 6C 6 33, 20 60% 80% 80% 80% ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology 1465

8 Adams et al. always be possible to safely remove all of this tissue. It is impossible to say with any certainty if this is true residual disease or scar tissue. This group do not reflect the discussion above because they are a group who are excluded from clinical trials but who are likely to be at very high risk. All of these women were treated with combination chemotherapy apart from the five aged over 70 years. Nine of the 12 went on to relapse. The three who did not were all under 70 and received CP. Conclusion This retrospective study shows that CP is administered more often than P to women with high-risk early EOC especially in those with younger age, better PS and nonmucinous histology. However, it does not necessarily add extra benefit even to women with very good PS in stage I disease. There does appear to be a survival advantage for the combination in women with stage II disease. Alternatively this may reflect selection bias in those who received monotherapy in stage II disease. Although the combination is well tolerated it does add a significant burden of toxicity in both the short and long term which cannot always be justified. Carboplatin monotherapy remains an attractive alternative for many women in this group especially those with an earlier stage or poorer performance status. For those with stage II disease and good performance status clinician may prefer combination therapy but it must be recognised that this is largely based on extrapolation from data from advanced disease or nonrandomised data in early disease. Without a randomised controlled trial it is difficult to justify the routine use of combination chemotherapy for women with early ovarian cancer. Disclosure of interests There are no conflicts of interest for this paper. Contribution to authorship All authors were involved in the initial design of the study. GA was responsible for acquiring the data for many of the women, interpreting the data drafting and revising the manuscript. JZ and JG both provided several critical reviews of the manuscript. HW was responsible for acquiring and analysing data. All authors approved the final version of the manuscript. Details of ethics approval Approval for the study was sought and obtained from the Clatterbridge Centre for Oncology Audit Committee. Funding There was no funding for the study. j References 1 Quinn M, Babb P, Brock A, Kirby E, Jones J. Cancer Trends in England and Wales Studies on Medical and Population Subjects No. 66. London: Office for National Statistics, Covens A, Carey M, Bryson P, Verma S, Fung K, Fung M, et al. Systematic review of first-line chemotherapy for newly diagnosed postoperative patients with stage II, III, or IV epithelial ovarian cancer. Gynecol Oncol 2002;85: Heintz AP, Odicino F, Maisonneuve P, Beller U, Benedet JL, Creasman WT, et al. Carcinoma of the ovary. J Epidemiol Biostat 2001;6: Vergote I, De Brabanter J, Fyles A, Bertelsen K, Einhorn N, Sevelda P, et al. Prognostic importance of degree of differentiation and cyst rupture in stage I invasive epithelial ovarian carcinoma. Lancet 2001;357: Young RC, Walton LA, Ellenberg SS, Homesley HD, Wilbanks GD, Decker DG, et al. Adjuvant therapy in stage I and stage II epithelial ovarian cancer. Results of two prospective randomized trials. N Engl J Med 1990;322: Bolis G, Colombo N, Pecorelli S, Torri V, Marsoni S, Bonazzi C, et al. Adjuvant treatment for early epithelial ovarian cancer: results of two randomized trials comparing cisplatin to no further treatment or chromic phosphate (32P). Ann Oncol 1995;9: Trope CJ, Kaern J, Hogberg T, Abeler V, Hagen B, Kristensen G, et al. Randomized study on adjuvant chemotherapy in stage I highrisk ovarian cancer with evaluation of DNA-ploidy as prognostic instrument. Ann Oncol 2000;11: Trimbos JB, Parmar M, Vergote I, Guthrie D, Bolis G, Colombo N, et al. International collaborative ovarian neoplasm trial 1 and adjuvant chemotherapy in ovarian neoplasm trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 2003;95: Trimbos JB, Vergote I, Bolis G, Vermorken JB, Mangioni C, Madronal C, et al. Impact of adjuvant chemotherapy and surgical staging in early-stage ovarian carcinoma: European Organisation for research and treatment of cancer adjuvant chemotherapy in ovarian neoplasm trial. J Natl Cancer Inst 2003;95: Colombo N, Guthrie D, Chiari S, Parmar M, Qian W, Swart AM, et al. International collaborative ovarian neoplasm trial 1: a randomized trial of adjuvant chemotherapy in women with early-stage ovarian cancer. J Natl Cancer Inst 2003;95: Winter-Roach BA, Kitchener HC, Dickinson HO. Adjuvant (postsurgery) chemotherapy for early stage epithelial ovarian cancer (review). In: The Cochrane Library. 2009, issue 4. Published by John Wiley & Sons, Ltd [ clsysrev/articles/cd004706/frame.html] Last accessed 10 June Du Bois A, Quinn M, Thigpen T, Vermorken J, Avall-Lundqvist E, Bookman M, et al consensus statements on the management of ovarian cancer: final document of the 3rd International Gynecologic Cancer Intergroup Ovarian Cancer Consensus Conference (GCIG OCCC 2004). Ann Oncol 2005;16(Suppl 8):viii Bell J, Brady MF, Young RC, Lage J, Walker JL, Look KY, et al. Randomized phase III trial of three versus six cycles of adjuvant carboplatin and paclitaxel in early stage epithelial ovarian carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2006;102: National Comprehensive Cancer Network (NCCN). Clinical Practice Guidelines in Oncology. Ovarian Cancer Including Fallopian Tube Cancer and Primary Peritoneal Cancer. V [ professionals/physician_gls/pdf/ovarian.pdf] last accessed 10 June ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology

9 Chemotherapy for early EOC, single or combination? 15 Calvert AH, Newell DR, Gumbrell LA, O Reilly S, Burnell M, Boxall FE, et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989;7: Bamias A, Papadimitriou C, Efstathiou E, Rodolakis A, Vlahos G, Voulgaris Z, et al. Four cycles of paclitaxel and carboplatin as adjuvant treatment in early-stage ovarian cancer: a six year experience of Hellenic Cooperative Oncology Group. BMC Cancer 2006;6: Shimada M, Kigawa J, Kanamori Y, Itamochi H, Oishi T, Minagawa Y, et al. Outcomes of patients with early ovarian cancer undergoing three courses of adjuvant chemotherapy following complete surgical staging. Int J Gynecol Cancer 2005;15: Young RC, Decker DG, Wharton JT, Piver MS, Sindelar WF, Edwards BK. Staging laparotomy in early ovarian cancer. J Am Med Assoc 1983;250: Fader AM, Rose PG. Role of surgery in ovarian carcinoma. J Clin Oncol 2007;25: Argyriou A, Koltzenburg M, Polychronopoulos P, Papapetropoulos S, Kalofonos HP. Peripheral nerve damage associated with administration of taxanes in patients with cancer. Crit Rev Oncol Hematol 2008;66: McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334: Piccart MJ, Bertelsen K, James K, Cassidy J, Mangioni C, Simonsen E. Randomized intergroup trial of cisplatin paclitaxel versus cisplatin cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results. J Natl Cancer Inst 2000;92: The International Collaborative Ovarian Neoplasm (ICON) Group. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet 2002;360: Muggia FM, Braly PS, Brady MF, Sutton G, Niemann TH, Lentz SL, et al. Phase III randomized study of cisplatin versus paclitaxel versus cisplatin and paclitaxel in patients with suboptimal stage iii or iv ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol 2000;18: du Bois A, Lück HJ, Meier W, Adams HP, Möbus V, Costa S, et al. A randomized clinical trial of cisplatin/paclitaxel versus carboplatin/paclitaxel as first-line treatment of ovarian cancer. J Natl Cancer Inst 2003;95: ª 2010 The Authors Journal compilation ª RCOG 2010 BJOG An International Journal of Obstetrics and Gynaecology 1467