|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 25 (September 1), 2005: pp. 6268-6269 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.01.6162
In Reply:Beth Israel Deaconess Medical Center, Boston, MA
Dana-Farber Cancer Institute, Boston, MA Alikhan et al have raised concerns about the design of our phase III trial and the implications of the trial's conclusions. The primary objective of our phase III study1 was to determine whether high-dose (HD) interleukin (IL) -2 was superior to outpatient subcutaneous IL-2 and interferon (IFN) in terms of progression-free survival (PFS). Based on previous studies, it was assumed that the percentage of patients who would remain alive and progression free at 3 years was 10% for those receiving HD IL-2, and 2% in the IL-2 and IFN arm. The sample size calculations used these 3-year PFS estimates, assumed PFS to be exponentially distributed and analyzed using a two-sided log-rank test, and desired .05 alpha level and 90% power. The protocol documentation considered the durations of accrual and follow-up and allowance for 5% ineligibility, and the trial planned accrual of 174 patients with final analysis after all patients had been followed for at least 3 years. For simplification, if we ignored the durations of accrual and follow-up and assumed no dropouts, then these assumptions would provide the desired power with a sample size of 168 patients (84 per arm) and 157 events (using PASS 2002 software, Number Cruncher Statistical Systems, Kaysville, UT). Our omission of the information that sample size was planned using a log-rank test is unfortunate.
Given limited information, Alikhan et al incorrectly assumed that we designed the trial using Fisher's exact test, with 3-year PFS as a binary end point; they calculated power for a one-sided The end point, and hence analysis approach, was changed from time-to-event to binary at trial completion. Scientifically, the expected benefit of high-dose IL-2 was the ability to increase response durability, without altering median PFS; 3-year PFS was felt to represent a clinically meaningful timepoint at which to assess durability of response. The analysis as a binary end point was possible, as there were no censored observations, with all patients having progressed, died, or observed to be free from progression at 3 years. On further data analysis, the assumptions for a time-to-event analysis were determined to be questionable. For these reasons, we changed the primary analysis plan. Alikhan et al further expressed concern about the implications of an inadequately powered comparison of PFS for the subset analyses that we presented on patients' overall survival. Although we state that the results of certain analyses are statistically significant, we agree with Alikhan et al, as we discussed in our article, that "...the results of the subset analyses should be seen as hypothesis generating and will need to be confirmed in future trials."1 We are currently planning a prospective trial that will attempt to confirm these findings. The observed difference in 3-year PFS (10.5% for HD IL-2 v 3.3% for IL-2/IFN) was consistent with the Cytokine Working Group experience in previous phase II studies of these two IL-2 regimens,2-5 but did not meet the accepted .05 level of statistical significance. While we believe that this randomized phase III trial provides additional evidence that HD IL-2 should remain the preferred therapy for "selected" patients with metastatic renal cell carcinoma, in consideration of the toxicity and limited efficacy of this treatment, we need to improve our criteria for selecting patients for IL-2 therapy and continue efforts to develop new treatment options for those patients with unfavorable selection features or those who experience disease progression after IL-2 therapy. Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Dollar Amount Codes (A) < $10,000 (B) $10,000-99,999 (C)
REFERENCES
1. McDermott DF, Regan MM, Clark JI, et al: Randomized phase III trial of high-dose interleukin-2 versus subcutaneous interleukin-2 and interferon in patients with metastatic renal cell carcinoma. J Clin Oncol 23:133-141, 2005 2. Atkins MB, Sparano J, Fisher RI, et al: A randomized phase II trial of high dose IL-2 either alone or in combination with interferon alpha 2B in advanced renal cell carcinoma. J Clin Oncol 11:661-670, 1993[Abstract] 3. Dutcher JP, Fisher RI, Weiss G, et al: Outpatient subcutaneous interleukin-2 plus alpha interferon in metastatic renal cell cancer three year follow-up of the Cytokine Working Group Study. Cancer J Sci Am 3:157-162, 1997[Medline] 4. Dutcher JP, Atkins MB, Fisher R, et al: Interleukin-2 based therapy for metastatic renal cell cancer: The Cytokine Working Group Experience, 1989–1997. Can J Sci Am 3:S73-S78, 1997
5. Dutcher JP, Logan T, Gordon M, et al: Phase II trial of interleukin-2, interferon-alpha, and 5-fluorouracil in metastatic renal cell cancer: A Cytokine Working Group Study. Clin Cancer Res 6:3442-3450, 2000
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
= .05 level test with our planned sample size of 174 patients and concluded that we had planned a severely underpowered trial. 
$100,000 (N/R) Not Required
