|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 4 (February 1), 2007: pp. 437-443 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.07.3494 Health-Related Quality of Life During and After Intraperitoneal Versus Intravenous Chemotherapy for Optimally Debulked Ovarian Cancer: A Gynecologic Oncology Group Study
From the College of Medicine, University of California at Irvine, Irvine, CA; Gynecologic Oncology Group Statistical and Data Center, Roswell Park Cancer Institute, Buffalo, NY; Johns Hopkins Kimmel Cancer Center, Baltimore, MD; Gynecologic Oncology, University of Oklahoma, Oklahoma City, OK; Psychiatry and Behavioral Science Research, Institute for Health Services Research and Policy Studies, Northwestern University, Chicago; and Center on Outcomes, Research and Education, Evanston Northwestern Healthcare, Evanston, IL Address reprint requests to Denise Mackey, Gynecologic Oncology Group Administrative Office, Four Penn Center, 1600 JFK Blvd, Ste 1020, Philadelphia, PA 19103; e-mail: dmackey{at}gog.org
Purpose A Gynecologic Oncology Group (GOG) randomized phase III trial (GOG 172) in optimal stage III epithelial ovarian cancer showed that intravenous (IV) paclitaxel plus intraperitoneal (IP) cisplatin and paclitaxel significantly lengthened progression-free survival and overall survival compared with IV paclitaxel and cisplatin. The purpose of this report is to comprehensively evaluate the patient-reported outcomes associated with IP versus IV therapy. Patients and Methods Four hundred fifteen eligible women were enrolled onto GOG 172 at member institutions. The Functional Assessment of Cancer Therapy–Trial Outcome Index (FACT-TOI; which includes physical, functional, and ovarian subscales) and neurotoxicity (Ntx) and abdominal discomfort (AD) subscales were used to assess patient-reported outcomes. Assessments were completed before random assignment, before cycle 4, and 3 to 6 weeks and 12 months after treatment. Results Physical and functional well-being and ovarian cancer symptoms were significantly worse in the IP arm before cycle 4 (P < .001) and 3 to 6 weeks after treatment (P = .001 for FACT-TOI). Patients in the IP arm also reported significantly worse AD before cycle 4 (P < .001) and significantly worse Ntx 3 to 6 weeks (P = .001) and 12 months (P = .003) after completing IP treatment. In general, however, the quality of life of both groups improved over time. Conclusion During active treatment, patients on the IP arm experienced more health-related quality-of-life disruption, AD, and Ntx compared with patients receiving conventional IV therapy. However, only Ntx remained significantly greater for IP patients 12 months after treatment. This trade-off should be included when discussing treatment options with patients. Future studies to mitigate the added burden associated with IP therapy are planned.
Results of several clinical trials suggest that standard management of advanced epithelial ovarian cancer (EOC) should include surgery for purposes of staging and debulking gross tumor, followed by platinum-taxane combination chemotherapy.1-3 Three randomized trials have shown that intraperitoneal (IP) cisplatin provides an additional survival advantage beyond standard intravenous (IV) administration.4-6 A recent report by the Gynecologic Oncology Group (GOG) demonstrated the clinical superiority of IP therapy in a phase III treatment study (GOG 172) of women with optimal stage III EOC. Patients were randomly assigned to receive either IV paclitaxel plus cisplatin or IV paclitaxel plus IP cisplatin and paclitaxel to compare progression-free survival (PFS), overall survival (OS), adverse effects, and health-related quality of life (HRQOL). The improvement in median OS time was 15.9 months favoring the IP study arm.6 This is a magnitude of improvement in median OS similar to that observed with the introduction of either cisplatin or paclitaxel.6,7 However, HRQOL was significantly worse in the IP arm at the second (before cycle 4) and third (3 to 6 weeks after treatment) time points but not 1 year after treatment.6 This is the first study, to our knowledge, to assess comprehensive patient-reported outcomes in an IP chemotherapy trial. Therefore, results presented in this report specifically detail the aspects of HRQOL that accounted for the significant differences between the IP and IV arms, including patient-reported outcomes of neurotoxicity (Ntx) and abdominal discomfort (AD). Given the current controversy associated with IP therapy,8,9 it has been suggested that physicians should ensure that patients have a clear understanding of the benefits and risks associated with IP therapy so that patients can make an educated decision,10 particularly because there are technical challenges and adverse effects associated with this therapeutic approach that are difficult to mitigate.11 The patient-reported outcomes of this trial can inform future trial design and treatment decision making.
Patients Eligible patients provided written informed consent for the therapeutic and HRQOL components of GOG 172 consistent with all federal, state, and local requirements before receiving protocol therapy at participation institutions with prior institutional review board approval of the protocol. Race or ethnicity was determined by the investigator or was self-reported at each site; these data are routinely collected as part of the GOG's effort to monitor and increase participation of minority (nonwhite) patients in clinical trials. Women with histologically confirmed stage III EOC or primary peritoneal carcinoma who had no residual disease more than 1.0 cm in diameter after surgical debulking and who met all other eligibility criteria were randomly assigned to receive paclitaxel 135 mg/m2 IV over 24 hours followed by either cisplatin 75 mg/m2 IV on day 2 (IV arm) or cisplatin 100 mg/m2 IP on day 2 plus paclitaxel 60 mg/m2 IP on day 8 (IP arm). At registration, patients decided whether they would or would not undergo a second-look laparotomy at completion of protocol therapy. Treatment cycles were every 3 weeks for six cycles. Details of treatment administration and outcome are reported elsewhere.6
QOL Assessment Twelve items added to the FACT-G reflect issues specific to ovarian cancer. The FACT-O has excellent psychometric properties including internal consistency and test-retest reliability13 and sensitivity to change over time in ovarian cancer patients responding to platinum-taxane therapy.14 The 11-item FACT/GOG-Ntx subscale was used to assess short- and long-term symptoms of Ntx.15 An internal consistency of 0.80 was obtained based on study data before chemotherapy cycle 4. An additional two items specific to AD were developed for this study and were added to two items from the FACT-O to address treatment-specific concerns potentially relevant to IP therapy. These four items formed a cohesive AD subscale that measured abdominal symptoms, with average interitem correlations of 0.64, an internal consistency of 0.875, and responsiveness to change over time computed on study data before cycle 4.16 The FACT-G, FACT-O subscale, Ntx subscale, and AD subscale were each scored using a 5-point scale (0 = not at all; 1 = a little bit; 2 = somewhat; 3 = quite a bit; and 4 = very much). A subscale score was computed when 50% or more of the items had been answered. The FACT-G score is the sum of the four subscale scores (FWB, emotional well-being, social well-being, and PWB) and is considered valid only when more than 80% of the items are complete. The FACT-O score is the sum of the FACT-G score and FACT-O subscale score. Higher FACT-O scores are associated with better HRQOL. The scoring of Ntx and AD subscales follows the same approach as described earlier, although higher scores imply greater symptoms. The ranges of scores are 0 to 156 (FACT-O), 0 to 44 (Ntx subscale), and 0 to 16 (AD subscale).
Statistical Analysis All statistical analyses were accomplished using SAS statistical software (SAS Institute, Cary, NC). Baseline HRQOL measures were examined for their comparability between arms using a two-sample t test. The independence of the follow-up HRQOL scores (before cycle 4, 3 to 6 weeks after treatment, and 12 months after treatment) and treatment was analyzed using a linear mixed model with unstructured covariance matrix adjusted for time effect, patients' age and performance status at random assignment, existence of gross residual disease, and baseline assessment scores. In the model fitting, assessment points were treated as values of a categoric variable. The restricted maximum likelihood was used to estimate the covariance parameters (assumed unstructured). The denominator degree of freedom was computed using a general Satterthwaite approximation. The interaction of treatment and assessment time was examined for constant treatment differences in HRQOL measures across time. When the interaction effect was significant, treatment differences in HRQOL measures were tested at each time point using appropriate contrast statements to obtain adjusted mean differences. When the interaction was not significant, the treatment was tested for its effect on HRQOL measures averaged over all follow-up assessments under the fitted mixed model. All patients were expected to complete the HRQOL assessments regardless of treatment delay or termination. Compliance was defined as the proportion of valid HRQOL assessments received to the number expected at each time point. Reasons for missed assessments were documented. To examine whether compliance was different between treatment arms over the assessment duration, the generalized linear mixed model (GLMM) with the generalized estimating equations (GEE) method and unstructured working correlation matrix was applied, with the assessment points treated as categoric variables. Compliance rates were assumed to follow binomial probability distribution, and logistic link function was chosen to relate to the linear predictor. The interaction of treatment and assessment time was examined for constant treatment effect on compliance across time. If the interaction was significant, compliance differences by treatment were tested at each time point using appropriate contrast statements to obtain adjusted mean differences. If the interaction was not significant, the treatment was tested for its effect on compliance averaged over all follow-up assessments under the fitted generalized linear mixed model.
Between March 1998 and January 2001, 429 patients were randomly assigned to either the IV or IP treatment arm. Of those, 415 patients were deemed eligible. The majority of eligible patients were non-Hispanic white (> 89%), were between the age of 41 to 70 years (80%), and had a performance status of 0 or 1 (> 92%; Table 1).
Compliance At random assignment, 96% of patients (399 of 415 patients) completed the HRQOL assessment (Table 2). At least three assessments were completed by 80.5% of patients (334 of 415 patients), and 54% of patients (224 of 415 patients) completed all four assessments. Forty-four patients discontinued HRQOL assessments after random assignment (IV arm, n = 14; IP arm, n = 30). Although fewer patients in the IP arm (P = .035) completed HRQOL assessments compared with patients in the IV arm at subsequent time points, the fitted GEE estimate on the interaction between treatment and time effect indicated that the differential compliance between the two arms did not change over time (P = .26). Reasons for missed assessments were similar with the exception of the second time point, when more assessments were missed in the IP arm (n = 17) than in the IV arm (n = 9) as a result of delinquency in data collection (institutional error), discontinuation of treatment (IP arm, n = 8; IV arm, n = 2), or unknown cause (IP arm, n = 10; IV arm, n = 4). It is important to note that patient refusal rates over the course of the study were overall less than 3%.
HRQOL Outcomes FACT-O total scores were reported by Armstrong et al.6 Before random assignment, FACT-G scores were 3.6 units lower (95% CI, 0.62 to 6.65 units; P = .018) and FACT-O subscale scores were 1.8 units lower (95% CI, 0.49 to 3.0 units; P = .007) in the IP arm (Table 3). There were no statistically significant Ntx and AD differences before random assignment.
During the treatment period, there were significantly different trends in the IP and IV arms regarding changes in FACT-G scores (P < .002), FACT-O subscale scores (P < .001), and AD scores (P = .006). These effects remained after adjusting for baseline scores, patient age and performance status at study entry, and presence of gross residual disease. To explore how the treatment affected HRQOL at each time point, the adjusted mean treatment differences in HRQOL scores were calculated under the fitted mixed model (Table 3). Before cycle 4, FACT-G (P < .001) and FACT-O subscale (P < .001) scores were reported as significantly lower in the IP arm. The IP scores were still lower for the FACT-G (P = .002) but not for the FACT-O subscale 3 to 6 weeks after completion of treatment. The differential HRQOL scores (FACT-G and FACT-O subscale) between the two arms were resolved 12 months after treatment. Both arms reported improved AD over the treatment duration (Fig 1). However, compared with the IV arm, patients on the IP arm reported significantly (P < .001) less improvement in AD before cycle 4, although there was no between-group AD difference soon after treatment ended. Chemotherapy-induced Ntx (Ntx subscale scores) worsened on both arms after baseline, but the degree to which it worsened was greater in the IP arm. Higher Ntx subscale scores in the IP arm were statistically significant during the follow-up assessment period (P < .001; Fig 2), and Ntx subscale score was the only patient-reported outcome that remained different between groups 1 year after treatment.
Additional analyses indicate that more than 80% of differences in FACT-O scores are attributable to the TOI. Table 3 illustrates the TOI differences between arms over time, including a 10-point difference before cycle 4 and a 7-point difference 3 to 6 weeks after treatment. Both are statistically significant (Fig 3) and also exceed what would be considered clinically meaningful,13,14,18 with patients in the IP arm reporting more physical, functional, and ovarian cancer–specific problems during and shortly after active treatment. Table 3 and Figure 3 also illustrate that, with the exception of the IP pre–cycle 4 assessment, patients in both study arms reported improved HRQOL over time.
Patients' FACT-O, Ntx subscale, and AD subscale scores were examined for associations with missing (immediately subsequent) assessments using the GEE method with unstructured covariance matrix adjusting for treatment effect and assessment point. No statistically significant relationship was detected for any of these measures.
Patients with good performance status and optimally debulked ovarian cancer experience better clinical outcomes when higher dose IP administration is added to more standard-dose IV chemotherapy.6 This more dose-intense therapy carries increased toxicities and adverse impact on HRQOL, which must be reconciled with the superior PFS and OS. Controversy over the relative worth of IP therapy persists based on concerns related to toxicities and adverse events. In this context, patient-reported outcomes have significant value when interpreting clinical trial results. To our knowledge, this is the first report to identify and longitudinally compare multiple patient-reported outcomes associated with IP/IV therapy versus IV therapy. This study is distinguished from others because the patients' personal experiences and perceptions related to specific disease- and treatment-related symptoms, including AD, Ntx, and other concerns, are documented. A statistically significant difference was observed with scores that clearly exceeded what would be considered a minimally important clinical difference between the IV (control) regimen and the IP (experimental) regimen. This poses a dilemma because the clearly superior therapy is associated with significantly more discomfort and HRQOL disruption in the short term and with more Ntx in the longer term. However, patients on the IP arm had increased toxicities attributed to the higher dose of cisplatin and paclitaxel, which, at minimum, contributed to the significant Ntx differences. It is less clear to what extent these higher doses contributed to AD and HRQOL differences because, in addition to the dose differences, there were toxicities related to the peritoneal catheter required to deliver IP therapy.11 More IP than IV patients (30 v 14 patients, respectively) did not complete on-treatment and follow-up questionnaires. This relative decrease likely corresponds to the dropout observed in the IP therapy arm before cycle 4 (98 patients in IP arm v 30 patients in IV arm).6 Usually, noncompletion of questionnaires reflects greater impairment and, in this case, is likely to further emphasize the disadvantage of IP therapy. We cannot prove that this difference in compliance had a measurable effect, but if it did, previous studies would suggest the bias is in favor of IV chemotherapy. For more than 50 years, oncology clinical trials have been designed within a paradigm that pushes dose and various combinations of chemotherapeutic agents found to be effective under the theory that more is better, and indeed, this is often true for studies having traditional clinical end points. Increased treatment-related toxicity is often viewed as a temporary but tolerable consequence of increased efficacy. In this trial, the higher chemotherapy dose delivered via IP administration was associated with significantly better survival but also increased toxicity.6 However, QOL and AD improved in both groups by 6 weeks after treatment, and the overall significant HRQOL differences disappeared within 1 year. Therefore, important questions and considerations remain. How should short- and long-term disruptions be evaluated with respect to benefits gained from treatment? Are there subgroups of patients less likely to tolerate IP therapy? How should treatment decision making unfold for optimally debulked ovarian cancer patients? Under what circumstances would patients prefer IP over IV therapy? These analyses illustrate the complex relationship between treatment efficacy and toxicity, where HRQOL data can be particularly useful in interpreting treatment implications. Overall HRQOL disruption was largely captured by the FACT-TOI, which is a summation of physical and functional well-being subscales, in addition to the FACT-O subscale. This HRQOL cluster, as well as patient-reported outcomes of Ntx and AD, should be targeted in future trials not only to evaluate treatment arm differences, but also to measure a patient's responsiveness to new supportive care strategies or surgical techniques aimed at making IP therapy more tolerable. The GOG plans additional studies to identify less complex regimens with the potential for reduced toxicity; however, these regimens may be less effective. In that setting, HRQOL and other patient-reported outcomes will provide important insight to further evaluate IP treatment benefits and costs. Although controversy over IP therapy persists, these findings can assist patient-physician discussions of treatment options.10
The authors indicated no potential conflicts of interest.
Conception and design: Lari B. Wenzel, Deborah K. Armstrong, Joan L. Walker Collection and assembly of data: Helen Q. Huang, Deborah K. Armstrong, Joan L. Walker Data analysis and interpretation: Lari B. Wenzel, Helen Q. Huang, David Cella Manuscript writing: Lari B. Wenzel, Helen Q. Huang, Deborah K. Armstrong, Joan L. Walker, David Cella Final approval of manuscript: Lari B. Wenzel, Helen Q. Huang, Deborah K. Armstrong, Joan L. Walker
The following Gynecologic Oncology Group member institutions participated in this study: University of Alabama at Birmingham, Duke University Medical Center, Abington Memorial Hospital, Walter Reed Army Medical Center, Wayne State University, University of Minnesota Medical School, Emory University Clinic, University of Southern California at Los Angeles, University of Mississippi Medical Center, Colorado Gynecologic Oncology Group PC, University of California at Los Angeles, University of Washington/Puget Sound Oncology Consortium, University of Pennsylvania Cancer Center, Milton S. Hershey Medical Center, University of Cincinnati, University of North Carolina School of Medicine, University of Iowa Hospitals and Clinics, University of Texas Southwestern Medical Center at Dallas, Indiana University Cancer Center, Wake Forest University School of Medicine, University of California-Irvine, Tufts-New England Medical Center, Rush-Presbyterian-St Luke's Medical Center, University of Kentucky, Community Clinical Oncology Program, Cleveland Clinic Foundation, Johns Hopkins Oncology Center, State University of New York–Stony Brook, Eastern Pennsylvania Gynecology/Oncology Center Inc, Washington University School of Medicine, Columbus Cancer Council, University of Massachusetts Medical School, Women's Cancer Center, University of Oklahoma, University of Virginia, University of Chicago, Tacoma General Hospital, Thomas Jefferson University Hospital, Mayo Clinic, Case Western Reserve University, Tampa Bay/H. Lee Moffitt Cancer Center, North Shore University Hospital, Gynecologic Oncology Network, Ellis Fischel Cancer Center, and Fletcher Allen Health Care.
The authors gratefully acknowledge the editorial assistance provided by Caron Modeas and Anne Reardon.
Supported by National Cancer Institute Grant No. CA 27469 to the Gynecologic Oncology Group (GOG) Administrative Office and Grant No. CA 37517 to the GOG Statistical and Data Center. In addition to financial support, the sponsor reviewed and approved the study design and monitored its progress and results. Presented in part at the 40th Annual Meeting of the American Society of Clinical Oncology, June 5-8, 2004, New Orleans, LA; and the 41st Annual Meeting of the American Society of Clinical Oncology, May 13-17, 2005, Orlando, FL. H.Q.H. had full access to all the data in the study and takes responsibility for accuracy of the data analysis. L.B.W. and H.Q.H., on behalf of the GOG and its Statistical and Data Center, take responsibility for the integrity of the data. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. McGuire WP, Hoskins WJ, Brady MF, et al: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 334:1-6, 1996 2. Piccart MJ, Bertelsen K, James K, et al: Randomized intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: Three-year results. J Natl Cancer Inst 92:699-708, 2000 3. Muggia FM, Braly PS, Brady MF, 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 18:106-115, 2000 4. Markman M, Bundy BN, Alberts DS, et al: Phase III trial of standard-dose intravenous cisplatin plus paclitaxel versus moderately high-dose carboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin in small-volume stage III ovarian carcinoma: An intergroup study of the Gynecologic Oncology Group, Southwestern Oncology Group, and Eastern Cooperative Oncology Group. J Clin Oncol 19:1001-1007, 2001 5. Alberts D, Liu PY, Hannigan EV, et al: Intraperitoneal cisplatin plus intravenous cyclophosphamide versus intravenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med 335:1950-1955, 1996 6. Armstrong D, Bundy B, Wenzel L, et al: Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med 354:34-43, 2006 7. National Cancer Institute: NCI clinical announcement on intraperitoneal chemotherapy in ovarian cancer (January 5, 2006). http://ctep.cancer.gov/highlights/ovarian.html. 8. Gore M: Intraperitoneal chemotherapy in ovarian cancer remains experimental. J Clin Oncol 24:4528-4530, 2006 9. Armstrong CK, Brady MF: Intraperitoneal therapy for ovarian cancer: A treatment ready for prime time. J Clin Oncol 24:4531-4533, 2006 10. Cannistra S: Intraperitoneal chemotherapy comes of age. N Engl J Med 354:77-79, 2006 11. Walker J, Armstrong DK, Huang HQ, et al: Intraperitoneal catheter outcomes in a phase III trial of intravenous versus intraperitoneal chemotherapy in optimal stage III ovarian and primary peritoneal cancer: A Gynecologic Oncology Group study. Gynecol Oncol 100:27-32, 2006[CrossRef][Medline] 12. Cella DF, Tulsky DS, Gray G, et al: The Functional Assessment of Cancer Therapy Scale: Development and validation of the general measure. J Clin Oncol 11:570-579, 1993 13. Basen-Engquist K, Bodurka-Bevers D, Fitzgerald MA, et al: Reliability and validity of the Functional Assessment of Cancer Therapy-Ovarian (FACT–O). J Clin Oncol 19:1809-1817, 2001 14. Wenzel L, Huang H, Monk B, et al: Quality of life comparisons in a randomized trial of interval secondary cytoreduction in advanced ovarian carcinoma: A Gynecologic Oncology Group study. J Clin Oncol 23:5605-5612, 2005 15. Huang H, Brady M, Cella D: Validation and reduction of the FACT/GOG-NTX subscale for platinum/paclitaxel-induced neurologic symptoms. Int J Gynecol Cancer 17:1-7, 2007[CrossRef][Medline] 16. Wenzel L, Huang H, Armstrong D, et al: Validation of a FACT/GOG-Abdominal Discomfort (AD) subscale: A Gynecologic Oncology Group trial. J Clin Oncol 23:754, 2005 (suppl, abstr 8101) 17. Sloan J, Cella D, Hays R: Clinical significance of patient-reported questionnaire data: Another step toward consensus. J Clin Epidemiol 58:1217-1219, 2005[CrossRef][Medline] 18. Yost KJ, Eton DT: Combining distribution- and anchor-based approaches to determine minimally important differences: The FACIT experience. Eval Health Prof 28:172-191, 2005 19. Calhoun EA, Welshman EE, Chang CH, et al: Psychometric evaluation of the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group–Neurotoxicity (FACT/GOG-Ntx) questionnaire for patients receiving systemic chemotherapy. Int J Gynecol Cancer 13:741-748, 2003[CrossRef][Medline] 20. Cella D, Peterman A, Hudgens S, et al: Measuring the side effects of taxane therapy in oncology: The Functional Assessment of Cancer Therapy–Taxane (FACT-Taxane). Cancer 98:822-831, 2003[CrossRef][Medline] 21. Moore DH, Donnelly J, McGuire WP, et al: Limited access trial using amifostine for protection against cisplatin- and three-hour paclitaxel-induced neurotoxicity: A Phase II study of the Gynecologic Oncology Group. J Clin Oncol 21:4207-4213, 2003 Submitted May 13, 2006; accepted November 14, 2006.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
0.36 between follow-up scores and baseline scores for all measures. Previous literature was considered to examine whether the magnitude of differences was clinically meaningful. Specifically, on the FACT-O total score, a 6-point difference corresponds to the average improvement in women with ovarian cancer who show improvement in performance status over time.
