|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Can Patients With Relapsed, Previously Untreated, Stage I Epithelial Ovarian Cancer Be Successfully Treated With Salvage Therapy?From the the Gynaecology Unit, Royal Marsden Hospital, London, UK. Address reprint requests to Martin Gore, MD, the Royal Marsden Hospital, Fulham Rd, London SW3 6JJ, UK; email: martin.gore{at}rmh.nthames.nhs.uk.
Purpose: The role of adjuvant chemotherapy in early-stage epithelial ovarian cancer (EOC) has been controversial. We have previously reported the cases of patients managed with a policy of observation only. We now present the salvage rate for the patients in that study who experienced relapse. Patients and Methods: One hundred ninety-four patients with stage I EOC presenting between 1980 and 1994 received no adjuvant chemotherapy, but were treated with platinum-based chemotherapy at relapse. We calculated the progression-free survival (PFS) and overall survival (OS) for the whole cohort and the salvage rate for those who experienced relapse. We defined salvage as freedom from relapse for 5 years after platinum treatment. Results: Sixty-one (31%) of 194 patients experienced relapse, and 55 received platinum-based chemotherapy. Twenty-four percent were progression-free at 5 years after this treatment. Clear-cell histology and cyst rupture before the patients’ original surgery were independent prognostic factors for PFS after salvage chemotherapy. The OS for all 194 patients is 72% at 10 years (median follow-up, 8.7 years), with an 80% disease-specific survival (DSS). Conclusion: We have shown that some patients with stage I EOC can be successfully treated with a salvage chemotherapy regimen after a policy of observation only. Interestingly, approximately 30% of stage I patients who die within 10 years do so from causes other than EOC (OS, 72%; DSS, 80%). Our findings need to be taken into consideration when the results from recent randomized trials of adjuvant chemotherapy in this patient population (International Collaborative Ovarian Neoplasm Trial 1/European Organization for Research and Treatment of Cancer Adjuvant Chemotherapy in Ovarian Neoplasm Trial) are being discussed with patients.
EPITHELIAL OVARIAN cancer (EOC) is the fifth most common cancer in women, with an estimated 23,100 new cases seen annually in the United States, and it accounts for more than 14,000 deaths per year.1 Stage I EOC accounts for approximately 25% of all cases, and in general, these women have a good prognosis.2 More precise prediction of outcome can be determined from a variety of other factors — for example, substage: patients whose disease is confined to one ovary have a 5-year survival rate of 72% to 92% (substage A); patients whose tumor is bilateral (substage B) have a 5-year survival rate of 62% to 85%; and if a tumor is on the surface of the ovary, its capsule is ruptured, or there are malignant cells in the peritoneal fluid (substage C), the 5-year survival rate is 57% to 82%.3 Other factors that may possibly predict for relapse include the presence of adhesions, tumor grade, histologic subtype, performance status, age, and serum albumin.3 Our group identified histologic grade, ascites, and surface tumor as significant independent prognostic factors predicting for relapse, but not International Federation of Gynecology and Obstetrics (FIGO) substage.4 In the largest published series on this subject, Vergote et al5 analyzed data from 1,545 patients who were entered onto five large studies, including our own. They found that the degree of differentiation, cyst rupture (preoperative and intraoperative), FIGO stage (1973), and age were all prognostic factors for disease-free survival.5 There has been controversy concerning the role of adjuvant chemotherapy in stage I patients. Most authors have felt that patients with stage IA grade 1 tumors do not require adjuvant chemotherapy, but there has been considerable uncertainty concerning patients who have features that are associated with relapse, such as high tumor grade, cyst rupture, or the presence of tumor on the surface of the ovary. In addition, there are few data on the fate of those patients who have not received any adjuvant chemotherapy but who subsequently experience relapse. We present here a group of such patients from our published prospective series of 194 patients with stage I EOC who underwent observation only after surgical staging. We attempted to treat these patients with a platinum-based salvage chemotherapy regimen at relapse, defining successful treatment with salvage therapy as freedom from disease at 5 years after this treatment. We also identified the prognostic factors associated with failure to successfully treat these patients with salvage therapy.
Patients Between 1980 and 1994, all patients with histologically proven, invasive stage I EOC referred to the Royal Marsden Hospital were entered onto a study of observation only after surgery. Patients with borderline ovarian tumors or those with a history of previous or concurrent malignancy other than squamous or basal carcinoma of the skin or cured carcinoma of the cervix were excluded. All patients underwent reassessment investigations postoperatively, which included histologic review, ultrasonography, or computed tomography (CT) of the abdomen and pelvis, chest x-ray, and serum CA-125 measurement. Only patients who had no evidence of residual disease at the time of reassessment (ie, staging investigations were normal) were entered onto the study. No patients received adjuvant chemotherapy or radiotherapy.
Follow-Up In years 6 to 10, patients were monitored with annual physical examination and serum CA-125 measurements, with CT scanning or ultrasound imaging of abdomen and pelvis being performed only if clinically indicated. Post-1985, measurement of serum CA-125 was introduced 3-monthly in years 1 to 2, and 6-monthly years 3 to 5, then annually after year 6. CT scanning was substituted for ultrasound imaging and annual laparoscopy after 1985.
Patient Characteristics In those patients who had experienced relapse, the interval between diagnosis and relapse was recorded together with details of the timing of the relapse (ie, at the patient’s routine follow-up visit or between appointments [interval relapse]). Site(s) of relapse, the dates and dosage of any therapy such as chemotherapy or radiotherapy given, tumor responses as assessed clinically or radiologically according to the standard World Health Organization criteria6, and follow-up information such as date and cause of death were also recorded. Patients who had initially undergone laparotomy, total abdominal hysterectomy, bilateral salpingo-oophorectomy, or partial or total omentectomy, with peritoneal and nodal examination with biopsy if macroscopically abnormal, were classified as having had complete surgery and adequate staging. Thirty-four patients (56%) had complete surgery by this definition. Only four patients had node sampling or peritoneal biopsies, as this study predates the time when there was a consensus as to the precise definition of adequate staging. Data on ascitic cytology were incomplete, because the study was started before release of the revised 1985 FIGO criteria. Patients in whom ascites were present but for whom the cytologic result was not documented were analyzed with those patients who had positive cytology. Subsequent surgical interventions were documented as well as details of any past history of cancer or family history of malignancy. The criteria for operating on patients at first relapse after observation was not predefined. However, the following parameters were taken into account, timing of relapse (late, > 2 years), site of disease, and whether this was solitary or not.
Statistics
Patient Characteristics Our original cohort of 194 patients with stage I invasive EOC underwent surgery followed by observation only. Sixty-one (31%) of these patients have experienced relapse after primary surgery at a median time of 17 months (range, 6 months to 15.7 years). The median age of this relapsed population is 60 years (range, 32 to 78 years). Table 1  lists the characteristics of these 61 patients who experienced relapse after observation only and those of the original cohort of 194 patients from which they derive. In this initial cohort, 30% of the patient population had grade 1 tumors, but only 10% of the population of patients who experienced relapse had well-differentiated tumors. By contrast, in the initial cohort, 20% of patients had grade 3 tumors, but on relapse, nearly half of the patients had poorly differentiated tumors.
Thirty-eight (62%) of our patients experienced relapse during the first 2 years of observation, but relapses still occurred up to 10 years after the institution of platinum-based chemotherapy. Forty-nine patients (80%) had their relapse diagnosed at a routine follow-up appointment. The method of detecting relapse was evenly distributed between physical signs, serum CA-125 measurement, imaging, or laparoscopy. Ten patients had interval relapses (ie, between scheduled visits), and one patient had a CA-125 measurement at the upper limit of normal at a scheduled appointment, which was then elevated when it was repeated 1 month later. In 12 patients (11%), there was either cytologic or histologic confirmation of relapse. Thirty patients (49%) experienced relapse with bulk disease (tumor > 2 cm). Five patients underwent further surgery at relapse before their platinum-based salvage regimen; they all had bulk disease and a solitary metastasis, and three of these patients experienced relapse with a disease-free interval of more than 2 years. Nine patients (15%) had evidence of either hematogenous spread (lung, pleura, liver) or distant lymph node disease (supraclavicular node relapse).
Treatment at Relapse
Survival From Salvage Chemotherapy The median follow-up duration of the 61 patients who experienced relapse is 6 years (range, 1 to 139 months). This follow-up period and PFS and OS were measured from the date of relapse after initial surgery and observation only. The PFS curve shows that at 5 years, 24% of patients were progression-free (Fig 1 ), and there is the suggestion that the majority of these patients were successfully treated with salvage therapy, because the shape of the Kaplan-Meier plot plateaus between years 5 and 10. However, it should be noted that there were some relapses after 5 years, and therefore it is clear that not everyone can be regarded as truly salvaged.
Univariate analysis using the log-rank test identified clear-cell histology and preoperative cyst rupture as significant predictive factors for PFS. Multivariate analysis for PFS using Cox regression analysis showed that both clear-cell histology (P = .001) and preoperative cyst rupture (P = .004) were independent prognostic factors for PFS in these 61 patients who experienced relapse (Table 3 ). Figure 2 is a Kaplan-Meier plot that demonstrates the poor prognosis for patients with clear-cell histology or preoperative cyst rupture at initial diagnosis. We have performed a multivariate analysis for overall survival using all these parameters, and the only parameter that emerged as statistically significant is clear-cell histology (Table 3).
Long-Term Follow-Up of 194 Patients With Stage I EOC The median follow-up of our cohort of 194 patients with stage I invasive EOC who underwent surgery followed by observation only is now 8.7 years. The relapse-free survival from initial diagnosis at 3 and 5 years was 75% and 70%, respectively, and the OS at 3 and 5 years was 90% and 84%, respectively. The 10-year relapse-free survival rate and disease-specific survival rate are 65% and 80%, respectively (Fig 3 ). The length of follow-up in this patient population has allowed us to compare OS and disease-specific survival. The OS for our stage I patients is 72% at 10 years in contrast with an 80% disease-specific survival (Fig 4). These data suggest that 30% of all deaths in patients with an initial diagnosis of stage I disease are due to causes other than ovarian cancer. These were as follows: other cancers (pancreatic, one patient; colorectal, one patient; breast, one patient), ischemic heart disease (four patients), mitral valve disease (one patient), bronchopneumonia (one patient), and "old age" (one patient, aged 90 years).
Comparative Analysis of Stage I and Stage III Patients Who Experienced Relapse We have compared the 61 patients who experienced relapse from observation only for their stage I disease with patients who present with de novo stage III disease. We have superimposed the disease-specific survival curve for these 61 patients who experienced relapse from observation only onto the disease-specific survival curves for stage III patients who were treated in two prospective randomized trials of platinum-based chemotherapy that were running during the same period as our observational study.7,8 We have drawn separate curves for patients who had optimal debulking and those with residual disease greater than 2 cm (Fig 5 ). In the first 2 years, survival of the relapsed stage I patients migrates with the poor prognosis stage III patients, but then the longer-term survival (after approximately 4 years) is similar to good prognosis stage III patients.
Stage I patients have a good prognosis, but it is well recognized that some will experience relapse and die. For many years, controversy has surrounded the role of adjuvant chemotherapy. The problem with giving all stage I patients adjuvant therapy is that the majority will be exposed to chemotherapy unnecessarily. Therefore, an alternative strategy is to observe patients with stage I disease and only to treat those who experience relapse. However, there are no data on whether such a strategy is safe, in other words, that patients who experience relapse after observation only can be successfully treated with a salvage chemotherapy regimen. Our observation study, which was performed on 194 patients, has enabled us to produce data on 61 stage I patients who have experienced relapse after a policy of observation only. We have demonstrated that most patients relapse within 2 to 3 years from initial diagnosis of their disease, though sometimes extremely late relapses occur. We defined successful treatment with salvage therapy as freedom from disease at 5 years after the institution of platinum-based chemotherapy for relapse. Our data suggest that we were able to successfully treat approximately 20% of our patients with salvage therapy, but occasionally relapses continued to occur beyond this time point. Clear-cell histology and preoperative cyst rupture were poor prognostic factors in our 61 relapsed patients. It is of note that degree of differentiation was not predictive for PFS or OS in these patients, although it has been identified as the most powerful prognostic indicator for DFS in stage I patients.5 However, it needs to be emphasized that with a study of our size, we would not expect to detect any prognostic factor of significance with a hazard ratio of less than 3. Our salvage rate of 24% is not in our view reassuring, more so that none of the 13 patients with clear-cell histology or preoperative cyst rupture were salvageable. In our original cohort of 194 patients, cyst rupture was significant for OS on univariate analysis but not for relapse-free survival. Histology was not significant for either of these end points.4 We did, however, demonstrate on multivariate analysis that patients with grade 3 cancers, ascites, and surface tumor were at the greatest risk of relapse.4 Recently, data from the International Collaborative Ovarian Neoplasm Trial 1 (ICON 1) and European Organization for Research and Treatment of Cancer Adjuvant Chemotherapy in Ovarian Neoplasm (EORTC-ACTION) trials have been reported.9,10 Adjuvant chemotherapy in early ovarian cancer was associated with a statistically significant OS benefit in ICON 19 but not in EORTC-ACTION.10 However, the combined data from these studies suggest a benefit for the use of adjuvant chemotherapy, with improvements in relapse-free survival and overall survival at 5 years of 11% and 8%, respectively.11 Detailed analysis of the EORTC-ACTION trial suggests that any benefit in OS might be confined to those patients who have been inadequately staged. The authors of this study also make the important point that there may never be a trial large enough even to answer the central question raised by their analysis, namely, whether adjuvant chemotherapy is of benefit to optimally staged patients, because of the paucity of events in this patient group. It should be emphasized that our original cohort of 194 patients with stage I disease was not adequately staged by today’s standards. The important clinical question of how one tailors individual advice to a patient about adjuvant chemotherapy according to their particular prognostic features is not resolved by these recent publications. This is because each subset analysis contains too few events. The clinical course of patients with relapsed stage I disease seems to be similar to those with stage III ovarian cancer. The survival of our patients in the first 2 years after relapse from observation mimics that seen in patients with suboptimally debulked stage III disease. However, after approximately 4 years, survival seems to be similar to patients with optimally debulked, good prognosis stage III tumors. The initial dramatic decline in survival may possibly be accounted for by those patients with clear-cell histology and preoperative cyst rupture. However, approximately 20% of our patients are disease-free after platinum-based chemotherapy for relapse, and this suggests that there may be a subpopulation of patients who are truly treated with success by delayed salvage chemotherapy. We have not been able to precisely identify these patients, although we note that no patient with clear-cell histology or preoperative cyst rupture was successfully treated with salvage therapy according to our definition. Furthermore, even in our good prognostic group, the salvage rate was still only 30%, and we would consider this to be worrisomely low. It is important to develop molecular markers that can identify three groups of stage I patients: those who will not experience relapse, those who are likely to experience relapse without adjuvant chemotherapy, and patients who can be successfully treated with delayed salvage chemotherapy.
A larger number of our patients might have survived if they had received chemotherapy at the time of initial diagnosis. The ICON 1/EORTC-ACTION data, which show an 8% survival benefit for adjuvant chemotherapy, supports such a suggestion. It is possible that in our series, the patients who were not successfully treated with salvage therapy are the women who would have benefited from immediate adjuvant therapy. However, our mature survival data clearly demonstrate that there is an 80% 10-year disease-specific survival rate for patients with stage I disease who are managed by observation only (Fig 3 In conclusion, we now have long-term follow-up on 194 patients with stage I EOC who were managed by observation only. Our 10 year relapse-free survival and disease-specific survival rates are 65% and 80%, respectively. Sixty-one patients (31%) have experienced relapse after an initial diagnosis of stage I disease and no adjuvant chemotherapy. We were able to demonstrate that patients with clear-cell histology or preoperative cyst rupture cannot be successfully treated with salvage therapy. However, salvage seems to be possible in approximately 20% of our patients, but we need to interpret the data with caution, because patients are continuing to experience relapse up to 10 years after salvage chemotherapy. The prognosis of stage I patients who experience relapse after a policy of observation only is similar to that of stage III patients. It is important to identify those patients who are not going to experience relapse after an initial diagnosis of stage I EOC and in whom no adjuvant chemotherapy is necessary. It is interesting to note that 30% of deaths in stage I patients are due to causes other than ovarian cancer.
1. Greenlee RT, Murray T, Bolden S, et al: Cancer statistics, 2000. CA Cancer J Clin 50:7–33, 2000[Abstract] 2. Annual report on the results of treatment in gynecological cancer, vol 23: Statements of results obtained in patients treated in 1990–1992. J Epidemiol Biostat 3:1–135, 1998[Medline] 3. Fiedlander M: Prognostic factors, in Sharp F, Blackett T, Berek J, et al (eds): Ovarian Cancer (ed 5, part III). Oxford, UK, Isis Medical Media Ltd, 2000, pp 199–215 4. Ahmed FY, Wiltshaw E, A’Hern RP, et al: Natural history and prognosis of untreated stage I epithelial ovarian cancer. J Clin Oncol 14:2968–2975, 1996[Abstract] 5. Vergote I, De Brabanter J, Fyles A, et al: Prognostic importance of degree of differentiation and cyst rupture in stage I invasive epithelial ovarian carcinoma. Lancet 357:176–182, 2001[CrossRef][Medline] 6. World Health Organization: WHO Handbook for Reporting Results of Cancer Treatment. Offset publication. Geneva, Switzerland, World Health Organization, 1979
7. Taylor AE, Wiltshaw E, Gore ME, et al: Long-term follow-up of the first randomized study of cisplatin versus carboplatin for advanced epithelial ovarian cancer. J Clin Oncol 12:2066–2070, 1994 8. Gore M, Mainwaring P, A’Hern R, et al: Randomized trial of dose-intensity with single-agent carboplatin in patients with epithelial ovarian cancer. J Clin Oncol 16:2426–2434, 1998[Abstract]
9. International Collaborative Ovarian Neoplasm Trial 1: A randomized trial of chemotherapy in women with early-stage ovarian cancer. J Natl Cancer Inst 95:125–132, 2003
10. Trimbos JB, Vertgote I, Bolis G, et al: Impact of adjuvant chemotherapy and surgical staging in early-stage ovarian carcinoma: European Organization for Research and Treatment of Cancer Adjuvant Chemotherapy in Ovarian Neoplasm trial. J Natl Cancer Inst 95:113–124, 2003
11. 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 95:105–112, 2003
12. Young RC: Early-stage ovarian cancer: To treat or not to treat. J Natl Cancer Inst 95:94–95, 2003 Submitted June 19, 2002; accepted May 15, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
