|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.10.028 on September 13 2004 © 2004 American Society of Clinical Oncology. Should CA-125 Response Criteria Be Preferred to Response Evaluation Criteria in Solid Tumors (RECIST) for Prognostication During Second-Line Chemotherapy of Ovarian Carcinoma?From the Departments of Oncology and Gynecology, Rigshospitalet, Copenhagen University Hospital; Department of Biostatistics, University of Copenhagen; the Department of Clinical Biochemistry, Statens Serum Institute; and Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark Address reprint requests to Bo Gronlund, MD, Department of Oncology 5073, Rigshospitalet, DK-2100 Copenhagen, Denmark; e-mail: bo.gronlund{at}dadlnet.dk
PURPOSE: The aim of the study was to compare the prognostic value of a response by the Gynecologic Cancer Intergroup (GCIG) Cancer Antigen (CA) -125 response criteria and the Response Evaluation Criteria in Solid Tumors (RECIST) on survival in patients with ovarian carcinoma receiving second-line chemotherapy. PATIENTS AND METHODS: From a single-institution registry of 527 consecutive patients with primary ovarian carcinoma, 131 records satisfied the inclusion criteria: ovarian carcinoma of International Federation of Gynecology and Obstetrics stage IC to IV, first-line chemotherapy with paclitaxel and a platinum compound, refractory or recurrent disease, and second-line chemotherapy consisting of topotecan or paclitaxel plus carboplatin. Univariate and multivariate analyses of survival were performed using the landmark method. RESULTS: In patients with measurable disease by RECIST and with assessable disease by the CA-125 criteria (n = 68), the CA-125 criteria were 2.6 times better than the RECIST at disclosing survival. In a multivariate Cox analysis with inclusion of nine potential prognostic parameters, CA-125 response (responders v nonresponders; hazard ratio, 0.21; P < .001) and number of relapse sites (solitary v multiple; hazard ratio, 0.47; P = .020) were identified as contributory prognostic factors for survival, whereas the parameters of RECIST (responders v nonresponders), as well as the remaining variables, had nonsignificant prognostic impact. CONCLUSION: The GCIG CA-125 response criteria are a better prognostic tool than RECIST in second-line treatment with topotecan or paclitaxel plus carboplatin in patients with ovarian carcinoma.
Whereas cure is possible in primary epithelial ovarian carcinoma, recurrent disease is generally considered as incurable.1,2 Patients with recurrent disease are candidates for second-line cytotoxic treatment, and palliation of symptoms and improved survival are the main therapeutic goals. These end points may be difficult to monitor, and the adjustment of the salvage treatment most often has been guided by intermediate end points, such as response evaluation by imaging-based tumor response criteria, assuming a correlation between a tumor response and longer survival.3,4 In several phase II studies, a correlation between a cancer antigen (CA) -125 response and a reduction in tumor size was found when monitoring the impact of second-line chemotherapy in patients with pretreated ovarian carcinoma.5-7 All these studies included patients treated with first-line regimens other than paclitaxel plus platinum, now internationally regarded as standard first-line chemotherapy,8 which may change the chemosensitivity of the tumor in the second-line clinical setting. However, it is not clear whether tumor marker–guided response criteria or imaging-based response criteria best reflect the outcome of second-line chemotherapy in terms of survival.9 The aim of the study was to compare the prognostic value of a response by the recently introduced Gynecologic Cancer Intergroup (GCIG) simplified CA-125 response criteria and the Response Evaluation Criteria in Solid Tumors (RECIST) on survival after start of second-line chemotherapy in a well-defined cohort of patients with ovarian carcinoma all pretreated with paclitaxel plus platinum as first-line treatment.
The Registry Since 1994, when paclitaxel plus platinum was introduced as standard first-line chemotherapy for ovarian carcinoma at the Rigshospitalet, all patients with ovarian tumors have been consecutively registered. The clinical data from patients with recurrent epithelial ovarian carcinoma have been included in a clinical database (Copenhagen Database for Ovarian Carcinoma; The Danish Data Protection Agency No. 2000-41-0126). First-line chemotherapy consisted of paclitaxel (175 mg/m2 as a 3-hour infusion) followed by either carboplatin (area under the curve of 5) or cisplatin (75 mg/m2) repeated every 3 weeks.10 Second-line chemotherapy has been standardized as follows: patients classified as platinum-resistant were offered single-drug treatment with topotecan 1.0 mg/m2 administered intravenously on days 1 through 5 every 21 days.11 They comprised patients with progression of disease (refractory disease) during first-line treatment, patients with persistent disease after the end of first-line therapy, and patients who responded and subsequently experienced relapse within 6 months after discontinuation of first-line chemotherapy. Patients with a treatment-free interval of more than 6 months after the end of first-line treatment have been considered platinum-sensitive, and they were re-treated with paclitaxel plus carboplatin using a schedule similar to the first-line treatment.12 Some other patients with refractory, persistent, or recurrent disease were included in various phase II studies of investigational approaches. Other second-line regimens included oral topotecan, topotecan with oral etoposide, topotecan plus doxorubicin, topotecan (1.2 mg/m2), ifosfamide, paclitaxel, encapsulated liposomal doxorubicin, oral melphalan, or no further chemotherapy in patients after complete tumor resection in relation to secondary cytoreductive surgery or in patients who refused treatment. The efficacy of the second-line chemotherapy was routinely assessed by imaging techniques (computed tomography [CT] scan or abdominal and endovaginal ultrasonography) after every two courses of treatment. Senior consultants at the department of radiology performed the ultrasonographies. All patients underwent serum CA-125 measurement before each cycle of second-line chemotherapy. Duration of treatment depended on the evaluation of response and followed departmental guidelines for standard second-line therapy. In patients declared to have obtained a complete response, chemotherapy was continued for two cycles after a complete response was achieved. In patients with a partial response or stable disease, standard antineoplastic therapy was continued until tumor progression. Patients with progressive disease or unacceptable toxicity were offered several different options, including inclusion in phase II protocols involving various investigational approaches, endocrine therapy, or supportive care.
Inclusion Criteria
RECIST
GCIG CA-125 Response Criteria
The CA-125 alterations were evaluated by using the recently introduced GCIG CA-125 response criteria.14 Briefly, two pretreatment samples at least twice (
Survival
Statistical Methods
Study Population In the period from August 1994 to July 2001, 527 consecutive patients with primary ovarian carcinoma received first-line treatment with paclitaxel plus platinum after an initial staging operation. Of these, a total of 131 patients fulfilled the inclusion criteria. A further 127 patients had refractory/persistent/recurrent disease treated with second-line regimens other than topotecan or carboplatin plus paclitaxel. The remaining 269 patients in the cohort had primary ovarian cancer without relapse. Forty-nine of the 131 patients had nonmeasurable disease by RECIST (n = 17), nonassessable disease by GCIG CA-125 criteria (n = 20), or simultaneous nonmeasurable disease and nonassessable disease (n = 12). Another 13 patients were removed from the survival analyses because they died before the landmark evaluation, in addition to one patient who received fewer than four cycles of second-line chemotherapy, thus leaving 68 patients for the direct comparison of the two prognostic classifications. The characteristics of the patients (n = 68) are listed in Table 1. The patients received a median of seven cycles (range, three to 14 cycles) of paclitaxel plus platinum as first-line treatment. Median TFI was 9.4 months (range, 0.9 to 51.3 months). The median patient age at the start of second-line chemotherapy was 60.5 years (range, 34.8 to 77.2 years). Second-line chemotherapy included topotecan (n = 31) or paclitaxel plus carboplatin (n = 37), and a median of seven second-line cycles (range, four to 16 cycles) was provided for salvage treatment. The latest day of start of second-line chemotherapy was December 17, 2001.
The response rates by the RECIST and the GCIG CA-125 criteria after four cycles of second-line chemotherapy (landmark time) are listed in Table 2. The response criteria were concordant in 81% (95% CI, 70% to 89%; 55 of 68 patients) and discordant in 19% (95% CI, 11% to 31%; 13 of 68 patients) of patients. Of the 34 patients categorized as nonresponders by RECIST at landmark time, five patients (15%) with stable disease obtained a subsequent partial response (late response) after further second-line treatment.
Survival At the time of analysis, four patients (6%) are alive without demonstrable disease, and six patients (9%) are alive with disease, whereas 58 patients (85%) have died with disease. All patients died from cancer-related disease. No patients were lost to follow-up. In a univariate analysis, survival after the landmark time was significantly longer in patients with a response according to RECIST compared with nonresponders (P = .035). Median survival (50% survival fraction) of responders and nonresponders was 20.2 months (range, 0.8 to 39.3 months) and 9.2 months (range, 1.7 to 37.0 months), respectively (Fig 1). With the GCIG CA-125 criteria, the results were analogs. The survival of responders according to the CA-125 response criteria was significantly longer compared with that of nonresponders (P < .0001). Median survival of responders and nonresponders was 20.4 months (range, 1.2 to 39.3 months) and 5.3 months (range, 0.8 to 23.3 months), respectively (Fig 2).
Survival from the landmark time was also significantly different for the parameters of TFI  6 versus more than 6 months (P = .0002) and TFI 12 versus more than 12 months (P = .012). No significant univariate differences in survival were found for the following factors: FIGO stage (I + II v II I+ IV; P = .19), histology (serous v nonserous; P = .47), residual disease after initial operation ( 1 cm v > 1 cm; P = .15), initial performance status (0 v 1 to 2; P = .19), response to first-line treatment (response v no response; P = .78); age at time of second-line chemotherapy ( 65 years v > 65 years; P = .63), number of relapse sites (solitary v multiple; P = .07), and performance status at time of second-line chemotherapy (0 v 1 to 2; P = .07).
The results from the Cox analyses are listed in Table 3. The GCIG CA-125 response criteria are thus two to three times ([1/0.23]/[1/0.57]) better than the RECIST at disclosing survival (Cox analyses No. 1 and 2). A more refined estimate takes the responder-to-nonresponder ratio into account (Table 2) and forms the ratio of the two prognostic efficacies, as follows: [ln 0.23
The survival of the different patient subgroups assessable by none, one, or both of the two response classifications are illustrated in Figure 3. The median survival of patient groups A, B, C, and D were 11.7 months (range, 0.8 to 39.3 months), 9.3 months (range, 0.5 to 58.8 months), 12.1 months (range, 1.1 to 50.7 months), and 17.5 months (range, 5.5 to 45.2 months), respectively. There was no difference in survival between the patient categories A, B, C, and D (log-rank tests: P > .05).
In second-line treatment of epithelial ovarian cancer, CA-125–based tumor response criteria and imaging-based response criteria have been compared in several studies, finding a concordance between an imaging-based tumor response and a response by CA-125 criteria ranging from 30% to 85%.5-7,17,18 This is the first study that compares the prognostic impact of a response by different tumor response criteria on survival in patients with recurrent or refractory ovarian tumors, demonstrating that response assessment by GCIG CA-125 criteria is superior to RECIST in predicting survival in this group of patients. The parameter of CA-125 response after four cycles of second-line chemotherapy was found to possess an independent impact as a prognostic factor for survival, whereas the parameter of RECIST response had no demonstrable independent influence on survival, neither in a bivariate nor in a multivariate analysis of prognostic factors. Hence the inclusion of nine potential prognostic parameters to adjust for the influence of confounding factors did not change the result of GCIG CA-125 response as holding independent prognostic information on survival. In a round-shaped tumor, a 30% reduction in the diameter (to obtain a response by RECIST) equals a 50% decrease in the area of the tumor.13,19 Halving of baseline CA-125 levels is thus a better predictor of survival than a halving of the tumor surface area determined by conventional imaging methods. These findings are in agreement with the biology of the disease. Recurrent ovarian carcinoma often spreads to peritoneal surfaces, forming multiple peritoneal implants that cannot readily be detected by conventional imaging techniques (CT scans or ultrasonography).20 It can be questioned how well the indicator lesions for tumor measurement reflect the overall tumor load. A chemotherapy-induced stabilization of the indicator lesions may be registered simultaneously with a considerable regression of diffuse carcinosis that is not measurable by RECIST and is therefore not included in the overall evaluation of the RECIST tumor response. CA-125, a glycoprotein expressed on the surface of ovarian cancer cells, may thus better reflect the total tumor load. However, it is uncertain whether a CA-125 response predicts the efficacy of chemotherapy or merely reflects the course of the disease, regardless of treatment given. In fact, it is unknown how the CA-125 levels would evolve if the chemotherapy had been discontinued. Ovarian cancer relapses are heterogeneous tumors with different expression of CA-125 in the different tumor clones. Agents that induce a reduction in CA-125 levels do not necessarily reduce the overall tumor growth. The alterations in CA-125 levels caused by the administration of an agent depends on several factors, such as the tumor clone mix, the agent’s affinity for tumor cells expressing CA-125, and the agent’s interference with CA-125 metabolism. The changes in CA-125 levels after the administration of a specific agent may be different using another agent. However, the advantage of the CA-125 criteria to RECIST in prognosticating survival was present both in patients receiving topotecan and in patients receiving paclitaxel-carboplatin (Table 3). The difference in the impact on survival between the RECIST and the CA-125 criteria may be further explored by an examination of the discordant cases. At the landmark time, 19% of patients had discordant response status (Table 2). It may be that a RECIST response occurred later than the landmark time, thus including a landmark time–related selection bias in the survival analysis. However, late RECIST response was observed in only five (15%) of 34 patients. Interestingly, all the five patients had stable disease (RECIST, nonresponders) at time of landmark, and antineoplastic therapy was continued in these patients until tumor progression according to departmental guidelines. Therefore, there was not any potential danger of inappropriately having changed therapy in these patients. A central question is whether the findings have any implication for how to monitor second-line chemotherapy in daily clinical practice. Prolongation of survival is a major goal among other goals, which also include the palliation of symptoms and increasing the general quality of life. As the CA-125 criteria better prognosticate survival than the RECIST, the study may indicate that tumor marker–guided response criteria should be preferred to imaging-based response criteria in the monitoring of salvage chemotherapy. However, a number of caveats should be voiced in this regard. The advantage of the CA-125 criteria compared with the RECIST is manifest only in patients assessable by both criteria. A total of 32 patients had nonassessable disease by CA-125 criteria, but 20 of these had measurable disease by RECIST, and in such patients, the treatment should still be monitored by imaging-based methods. Twelve patients had nonassessable and nonmeasurable disease, and in such cases, the monitoring of second-line therapy is quite difficult. This emphasizes the necessity of evaluation of other potentially usable tumor markers such as tetranectin21 or cancer-associated serum antigen22 in patients with nonmeasurable and CA-125 nonassessable disease. The findings in the present study are valid only for patients treated with either topotecan or paclitaxel plus carboplatin (Table 3). Other agents or combinations of agents may act differently, and other findings of the impact of CA-125 criteria and RECIST may be observed. The impact of the CA-125 criteria and the RECIST in prognosticating survival should thus be examined for every single agent or combinations. Noteworthy, 127 patients receiving second-line regimens other than topotecan or paclitaxel plus carboplatin in the cohort were not included in this analysis, which may present a potential selection bias. However, the inclusion of these groups would add to the heterogeneity of the study population and potentially obscure and confound the analyses. Imaging-based tumor response criteria, such as the RECIST, are based on visualization of the shrinkage and disappearance of the tumor. This gives rise to some methodologic problems because imaging-based response criteria can be confounded by inaccuracies in the use of the radiographic techniques.23 In the RECIST guidelines, the ultrasound technique is generally discouraged for tumor response evaluation because of the interference from bowel gas in the measurement of para-aortic lesions and subjectivity in the interpretation of tumor size.13 In ovarian carcinoma, most recurrences are located in the pelvic region, where the vaginal ultrasound technique is comparable to CT scans.24 Furthermore, many pelvic relapses are located around the top of the vagina, which may act as an anatomic landmark for reproducible measurements. The relationship between response and survival contains a number of potential biases and has been intensively discussed in the oncologic literature.15,16,25-27A bias that may contribute to the long survival in patients with a response is the guarantee-time effect. Patients whose records refer them to a definite response category must have lived long enough for the change in their condition to occur and be formally recorded. To minimize the effect of this bias, we used the landmark method in which the patients were followed forward in time from the postfourth cycle evaluation, and survival of the response categories from this arbitrarily defined landmark could be compared without bias, according to Buyse and Piedbois.16 In fact, by using this method, the time origin is shifted to the right, so that response can be handled as any other prognostic factor known at the origin of the survival assessment. The major disadvantage of the landmark method is that the results and conclusions depend on the selection of the landmark time. If the landmark interval chosen is too short, many responses are ignored; if it is too long, many early deaths are ignored. In this study, the selected landmark was the time of the clinical evaluation after four cycles of second-line chemotherapy, and by that, 14 patients (17%) were discarded from the survival analyses because of having received fewer than four cycles or because of early death. If, alternatively, the evaluation after six cycles of chemotherapy had been chosen for landmark, even more cases would have been discarded. In conclusion, a response by GCIG CA-125 criteria to second-line chemotherapy with topotecan or paclitaxel plus carboplatin is associated with a survival gain also after correction for prognostic factors and elimination of the guarantee-time effect. This retrospective analysis indicates that CA-125 response criteria are a better prognostic tool than the RECIST in the second-line treatment of ovarian carcinoma pretreated with paclitaxel plus platinum and are presumably also preferable as a therapeutic monitoring tool in salvage treatment. The question of monitoring, however, should be elucidated in a randomized trial comparing the CA-125 criteria and the RECIST in the monitoring of salvage chemotherapy using survival and quality of life as outcome parameters, and preferably, it should include agents other than topotecan or paclitaxel plus carboplatin.
The authors indicated no potential conflicts of interest.
Supported by grants from the M. Kristjan and Margrethe Kjær Foundation, and the H:S Copenhagen Hospital Corporation (grant No. 24-03R), Copenhagen, Denmark. Authors’ disclosures of potential conflicts of interest are found at the end of this article.
1. Cannistra SA: Is there a "best" choice of second-line agent in the treatment of recurrent, potentially platinum-sensitive ovarian cancer? J Clin Oncol 20: 1158-1160, 2002
2. Ozols RF: Recurrent ovarian cancer: Evidence-based treatment. J Clin Oncol 20: 1161-1163, 2002 3. Berek JS, Bertelsen K, du BA, et al: Advanced epithelial ovarian cancer: 1998 consensus statements. Ann Oncol 10: S87-S92, 1999 (suppl 1) 4. Markman M: Clinical response versus clinical benefit in oncology: Not necessarily equivalent terms. J Cancer Res Clin Oncol 123: 363-364, 1997[Medline]
5. Bridgewater JA, Nelstrop AE, Rustin GJ, et al: Comparison of standard and CA-125 response criteria in patients with epithelial ovarian cancer treated with platinum or paclitaxel. J Clin Oncol 17: 501-508, 1999
6. Dieras V, Bougnoux P, Petit T, et al: Multicentre phase II study of oxaliplatin as a single-agent in cisplatin/carboplatin +/- taxane-pretreated ovarian cancer patients. Ann Oncol 13: 258-266, 2002
7. Rustin GJ, Nelstrop AE, Crawford M, et al: Phase II trial of oral altretamine for relapsed ovarian carcinoma: Evaluation of defining response by serum CA125. J Clin Oncol 15: 172-176, 1997 8. Cannistra SA: Progress in the management of gynecologic cancer: Concensus summary statement. J Clin Oncol 21: S129-S132, 2003 (suppl 10)[CrossRef]
9. Markman M: The myth of measurable disease in ovarian cancer. J Clin Oncol 21: 3013-3015, 2003
10. Neijt JP, Engelholm SA, Tuxen MK, et al: Exploratory phase III study of paclitaxel and cisplatin versus paclitaxel and carboplatin in advanced ovarian cancer. J Clin Oncol 18: 3084-3092, 2000 11. Gronlund B, Hansen HH, Hogdall C, et al: Efficacy of low-dose topotecan in second-line treatment for patients with epithelial ovarian carcinoma. Cancer 95: 1656-1662, 2002[CrossRef][Medline] 12. Gronlund B, Hogdall C, Hansen HH, et al: Results of reinduction therapy with paclitaxel and carboplatin in recurrent epithelial ovarian cancer. Gynecol Oncol 83: 128-134, 2001[CrossRef][Medline]
13. Therasse P, Arbuck SG, Eisenhauer EA, et al: New guidelines to evaluate the response to treatment in solid tumors: European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92: 205-216, 2000 14. Rustin GJ: Use of CA-125 to assess response to new agents in ovarian cancer trials. J Clin Oncol 21: S187-S193, 2003 (suppl 10) 15. Anderson JR, Cain KC, Gelber RD: Analysis of survival by tumor response. J Clin Oncol 1: 710-719, 1983[Abstract] 16. Buyse M, Piedbois P: On the relationship between response to treatment and survival time. Stat Med 15: 2797-2812, 1996[CrossRef][Medline] 17. Davelaar EM, Bonfrer JM, Verstraeten RA, et al: CA 125: A valid marker in ovarian carcinoma patients treated with paclitaxel? Cancer 78: 118-127, 1996[CrossRef][Medline] 18. Morgan RJ Jr, Speyer J, Doroshow JH, et al: Modulation of 5-fluorouracil with high-dose leucovorin calcium: Activity in ovarian cancer and correlation with CA-125 levels. Gynecol Oncol 58: 79-85, 1995[CrossRef][Medline]
19. Gehan EA, Tefft MC: Will there be resistance to the RECIST (Response Evaluation Criteria in Solid Tumors)? J Natl Cancer Inst 92: 179-181, 2000
20. Tempany CM, Zou KH, Silverman SG, et al: Staging of advanced ovarian cancer: Comparison of imaging modalities—Report from the Radiological Diagnostic Oncology Group. Radiology 215: 761-767, 2000 21. Deng X, Hogdall EV, Hogdall CK, et al: The prognostic value of pretherapeutic tetranectin and CA-125 in patients with relapse of ovarian cancer. Gynecol Oncol 79: 416-419, 2000[CrossRef][Medline] 22. McGuckin MA, Layton GT, Bailey MJ, et al: Evaluation of two new assays for tumor-associated antigens, CASA and OSA, found in the serum of patients with epithelial ovarian carcinoma: Comparison with CA125. Gynecol Oncol 37: 165-171, 1990[CrossRef][Medline]
23. Thiesse P, Ollivier L, Stefano-Louineau D, et al: Response rate accuracy in oncology trials: Reasons for interobserver variability—Groupe Francais d’Immunotherapie of the Federation Nationale des Centres de Lutte Contre le Cancer. J Clin Oncol 15: 3507-3514, 1997
24. Kurtz AB, Tsimikas JVTCMC, et al: The comparative values of Doppler and conventional US, CT, and MR imaging correlated with surgery and histopathologic analysis: Report of the Radiology Diagnostic Oncology Group. Radiology 212: 19-27, 1999
25. Pazdur R: Response rates, survival, and chemotherapy trials. J Natl Cancer Inst 92: 1552-1553, 2000
26. Piantadosi S, McGuire WP: Assessing the effect of response on survival in ovarian cancer. J Natl Cancer Inst 84: 376-378, 1992
27. Torri V, Simon R, Russek-Cohen E, et al: Statistical model to determine the relationship of response and survival in patients with advanced ovarian cancer treated with chemotherapy. J Natl Cancer Inst 84: 407-414, 1992 Submitted October 3, 2003; accepted April 29, 2004.
Related Editorial
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
70 units/mL) the upper cutoff of normal (> 35 units/mL) and at least two additional samples after the start of treatment are required to have assessable disease. A response has occurred if there is at least a 50% decrease in CA-125 levels that is confirmed by the fourth sample. 
(41/68)(27/68)/ln 0.57
