|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology
Developing Drugs to Decrease the Toxicity of ChemotherapyUnited States Food and Drug Administration, Rockville, MD To the Editor:Batist et al1 report a clinical trial of Myocet (liposome-encapsulated doxorubicin; The Liposome Company, Elan Corporation, Princeton, NJ) for treatment of metastatic breast cancer (MBC) and conclude that "Myocet improves the therapeutic index of doxorubicin by significantly reducing cardiotoxicity and grade 4 neutropenia and provides comparable antitumor efficacy, when used in combination with cyclophosphamide as fist-line therapy for MBC." This study was one of two randomized trials of similar design submitted by The Liposome Company, Inc, to the Food and Drug Administration (FDA) as part of a New Drug Application (NDA). This NDA was discussed before the FDA Oncologic Drugs Advisory Committee (ODAC) on September 16, 1999 (for further information, contact the FDA’s website at http:www.fda.gov/cder/foi). Despite the encouraging results of the Batist study, a second study, not mentioned in the article, was less encouraging. ODAC recommended not approving this drug for the first-line treatment of MBC. Two Myocet trials were designed in consultation with the FDA in 1994 to compare the relative cardiotoxicity and efficacy of Myocet with doxorubicin in the treatment of MBC. The trial described by Batist et al1 (study 1) compared doxorubicin to Myocet both in combination with cyclophosphamide. Study 2 compared single-agent doxorubicin to Myocet. The two studies had a similar design for cardiac toxicity evaluation, and comparisons of response rates were the primary efficacy end points of both trials.1 Response rate was considered an acceptable surrogate for efficacy because the anticancer agent in each arm was a doxorubicin moiety. Each trial was to demonstrate, using 95% confidence intervals (CIs), that at least 75% of the doxorubicin response rate was retained with Myocet. Time to progression and survival were secondary end points. Both studies demonstrated that Myocet was less cardiotoxic than doxorubicin. The results of study 1 were previously described.1 In the second study, 224 patients were randomized either to Myocet (108 patients) or doxorubicin (116 patients). Cardiac events were observed in 36% versus 17% and congestive heart failure in 8% versus 2% of the doxorubicin and Myocet patients, respectively. Study 2 did not provide sufficient confirmation that the anticancer efficacy associated with doxorubicin is preserved by using Myocet. Even though the response rates for the two treatment arms were the same (26%), this study was inadequately powered to exclude the possibility that Myocet was less effective than doxorubicin. The sample size was calculated based on the erroneous assumption that these doxorubicin moieties would produce higher response rates. The trial’s regulatory goal was to prove that the response rate with Myocet was at least 75% of that with doxorubicin (ie, that the lower bound of the 95% CI for the ratio of response rates [doxorubicin/Myocet] was at least 0.75). Because the lower bound of the CI was 0.62, this study only supported the conclusion that the Myocet response rate was at least 62% of the doxorubicin response rates. In addition, a trend toward inferior survival in the Myocet arm compared with the doxorubicin arm was noted (median survival, 14.6 months v 20.1 months, respectively; P = .07). The FDA multivariate analysis correcting for imbalances in prognostic factors demonstrated a significant survival difference (P = .03). ODAC voted (nine to two) to recommend against approval of this drug for the first line of treatment of MBC. In evaluating agents to reduce toxicities of oncology drugs, two critical objectives must be addressed. Demonstrating a reduction in toxic effects is the initial step in evaluating a new agent. A more difficult challenge is demonstrating with confidence that the efficacy of the standard drug has not been compromised by the drug formulation or by chemotherapy protection. Dexrazoxane (Zinecard; Pharmacia & Upjohn, Peapack, NJ) is another example of the difficulties in evaluating chemotherapy protection agents. Although Myocet is a liposomal formulation of doxorubicin and dexrazoxane is a chelating agent administered before a doxorubicin infusion, both drugs are intended to retain the benefits of doxorubicin while decreasing the associated cardiotoxicity. The studies submitted in the NDA for dexrazoxane in 1992 were similar in design to those submitted in the Myocet NDA in December 1998. In the largest of the breast cancer studies submitted to the FDA for dexrazoxane NDA, comparing chemotherapy with fluorouracil, doxorubicin, and cyclophosphamide (FAC) to FAC plus dexrazoxane, the response rate was significantly lower in the FAC plus dexrazoxane arm. In 293 patients with measurable disease, response rates were 63% and 48%, respectively (P = .0007). Based on this observation, the ODAC did not recommend approval of dexrazoxane in 1992. However, in 1994, the FDA reviewed additional data and analyses demonstrating that dexrazoxane provided cardioprotection even when its administration ws delayed. Dexrazoxane was approved for treatment of MBC in patients who had received a cumulative dose of 300 mg/m2 of doxorubicin. The benefits of cardioprotection in women who continue doxorubicin after receiving large cumulative doses were judged to outweigh the risks of tumor protection from dexrazoxane. These results were later published2,3 and formed the basis for dexrazoxane treatment recommendations in the American Society of Clinical Oncology Guidelines for the Use of Chemotherapy and Radiotherapy Protectants.4 The development of drugs aimed at reducing toxic effects and improving patients’ quality of life is laudable. Unless one convincingly demonstrates that the mechanism of chemotherapy protection will not interfere with the antitumor activity of anticancer drugs (eg, when evaluating antinausea agents), clinical trial designs should not only focus on toxicity reduction but also assure that the effectiveness of therapy is not compromised. Because the latter goal of retention of anticancer efficacy tends to be a more arduous task, clinical trials should be powered appropriately to ensure the reduction of toxicity is not at the expense of established clinical benefit. REFERENCES
1.
Batist G, Ramakrishnan G, Sekhar R, et al: Reduced cardiotoxicity and preserved antitumor efficacy of liposome-encapsulated doxorubicin and cyclophosphamide compared with conventional doxorubicin and cyclophosphamide in a randomized, multicenter trial of metastatic breast cancer. J Clin Oncol 19: 1444-1454, 2001
2.
Swain SM, Whaley FS, Gerber MC, et al: Cardioprotection with dexrazoxane for doxorubicin-containing therapy in advanced breast cancer. J Clin Oncol 15: 1318-1332, 1997
3.
Swain SM, Whaley FS, Gerber MC, et al: Delayed administration of dexrazoxane provides cardioprotection for patients with advanced breast cancer treated with doxorubicin-containing therapy. J Clin Oncol 15: 1333-1340, 1997
4.
Hensley M, Schuchter LM, Lindley C, et al: American Society of Clinical Oncology clinical practice guidelines for the use of chemotherapy and radiotherapy protectants. J Clin Oncol 17: 3333-3355, 1999
Response
Jewish General Hospital, McGill University, Montreal, Canada In Reply:We would like to respond to the letter by Drs Williams, Cortazar, and Pazdur with respect to the relative complexities of clinical trial design and analysis and the need for randomized clinical trials with sufficient power to detect true efficacy differences, should they exist. They note that our published trial1 met all safety and efficacy end points, but that a second, smaller, single-agent study, which has not yet been published, was inadequately powered and couldn’t exclude the possibility that the cardiac sparing qualities of the TLC D-99 liposomal formulation (Myocet) might reduce doxorubicin’s antitumor efficacy. First, as we are both among the authors of an upcoming report of that study, we would like to assure Williams et al that the report of that study has been submitted for publication. The manuscript provides detailed discussion of specific points and interpretations of the data referred to in the Williams et al letter. However, we do agree that the study had several design flaws among which was, as Williams et al correctly point out, the fact that it was underpowered due, in part, to erroneous assumptions regarding anticipated response rates in both the control and the experimental arms. With the benefit of hindsight, if we were to repeat the trial (which, unfortunately, is not possible, because single-agent doxorubicin is no longer standard of care for first-line therapy of metastatic breast cancer), we would alter the design and statistical approach. Additional clinical trials with TLC D-99 in combination with various other agents are underway. Preliminary results of one such trial will be presented at the 2001 Annual Meeting of the American Society of Clinical Oncology, and a larger, randomized phase III trial of TLC-99 in a combination regimen is anticipated.
NOTES The views expressed herein do not necessarily represent the views or findings of the United States Food and Drug Administration or the United States Government. REFERENCES 1. Batist G, Ramakrishnan G, Sekhar Rao C, et al: Reduced cardiotoxicity and preserved antitumor efficacy of liposome-encapsulated doxorubicin and cyclophosphamide compared with conventional doxorubicin and cyclophosphamide in a randomized multicenter trial of metastatic breast cancer. J Clin Oncol 19: 1444-1454, 2001
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
