Electronic Letters to:
|
|
E-letters: Electronic letters published:
-
![[Read eLetter]](/icons/misc/sectionDOWN.gif)
TKIs before chemotherapy in advanced NSCLC
- Francesco Perrone, [Ciro Gallo], [Fortunato Ciardiello], and [Cesare Gridelli] (15 January 2008)
|
|
|||
|
Francesco Perrone, Director Clinical Trials Unit NCI, Via Mariano Semmola, 80131 Naples, Italy, [Ciro Gallo], [Fortunato Ciardiello], and [Cesare Gridelli]
Send e-letter to journal:
|
We read with great interest the paper by D’Addario et al. reporting a phase 2 trial in unselected NSCLC patients of first-line gefitinib followed at progression by standard chemotherapy with gemcitabine and cisplatin (1). The Authors conclude that such treatment strategy should be further explored only in selected patients, according to the translational analyses, because the trial did not reach the primary outcome envisaged in the study design. We disagree with this message and consider their results promising also for the unselected population and, thus, worthy of further studies also in this setting. Hereafter are the reasons of our interpretation. First, overall survival in the whole population of 63 patients is excellent with a median approaching 1 year and close to 30% of patients alive at 18 months. As a reference, in a trial recently published by our group (2), in a similar population of patients (performance status 0-1) all treated with a similar chemotherapy, the corresponding results were a median survival of 44 weeks and a similar rate of patients alive at 18 months. It does matter, in our opinion, that the result of D’Addario et al. comes from a study where only two-thirds of patients did actually receive chemotherapy. So, survival results per se should be considered as interesting and worthy of further studies in the unselected population. Second, the outcome of chemotherapy, among 41 patients who received it, was also considerable. They report a 34.1% objective response rate and a 70.7% disease stabilization rate, that compare well with 31.0% and 59.2% observed in our study cited above, respectively. Such data suggest that activity of chemotherapy was not impaired by the delay induced by first- line gefitinib. This is a very important finding, because we still have very few data on chemotherapy after TKIs and on how much delay of treatment could impair efficacy of chemotherapy. Of course, we acknowledge that chemotherapy was given in this case to a selected population: all, but one, of 22 patients who did not receive chemotherapy had apparently a very bad prognosis. However, given the good results among those who received chemotherapy, maybe that those who never underwent chemotherapy would not have had a significant benefit if they had received it up-front. If this were true, first line gefitinib would have worked as a way to exclude from chemotherapy a group of patients who would have had no benefit from it. Not bad, in our view. Third, the Authors report that a disease-stabilization rate at 12 weeks of 38% as claimed by the Investigators was down-sized to 24% by independent radiological review. We understand that, given the choice of the primary end-point, the Authors made their best to give a reliable result. However, it’s not clear how such review affected conduct of the study. Was it done before assessment of the first stage of the study? Did it produce suspension of gefitinib or the patients were actually treated according to Investigators’ interpretation? If the latter were true, some patients would have remained on gefitinib although having progressed; but given final survival, this did not produce too many damages. Anyway, we argue that independent radiological review does not belong to real life and that changing the label of a response because of an eventually small discrepancy in tumor measurement does not probably reflect true prognosis. Our benevolent interpretation of the “mistakes” made by the Investigators in response assessment is supported by the good result in terms of survival. It has been recently suggested that, particularly with target- based agents, changes in tumor size might be better considered as a continuous variable rather than within rigid categories (3). Of course, we understand that even without independent radiological review the result would have been negative (24 successes out of 26 required by study design). Our fourth comment, indeed, is that the Authors chose a wrong primary end-point. As they state in the title, the aim was to study gefitinib followed at progression by chemotherapy (whenever possible), but they actually sacrificed trial design to test the activity of the first half of the strategy. It would have been much better, in our view, to define overall survival or a survival rate at a given time as the primary end-point, testing the efficacy of the strategy; given the survival outcome, the result would have probably been positive. At most, some constraints could be applied to stop the trial early in case time to progression with first line gefitinib were disastrous (clearly lower than the 30% considered as undesired p0 in their trial design), to warrant patients’ protection. Of course, we agree that among selected patients TKIs might produce their maximum benefit. But, we believe that proposed methods for selection are not easily applicable in clinical practice, their cost is still high, and the consensus of the scientific community on the best way to select patients is still low. The paper by D’Addario witnesses difficulties in selecting patients. Should we study only patients with EGFR mutation (4 out of 57)? Or only never smokers (9 out of 63), for whom the correlation seems to be non statistically significant? Our group is running a phase 3 trial comparing first line erlotinib followed at progression by chemotherapy with the standard reversal strategy, chemotherapy first and erlotinib after progression (4). The study is planned as a non-inferiority survival comparison and requires 900 unselected patients. Although we agree with general criticisms on non- inferiority trials (5), this represents one of the few clinical and experimental conditions where non-inferiority might be of great interest for scientific community and patient well-being, because it could be obtained sparing chemotherapy to approximately one-third of patients, according to the study of D’Addario et al. In case non-inferiority were demonstrated, an important piece of information will be how the patients cope with the different patterns of toxicity of TKIs and chemotherapy and which is their effect on quality of life. To warrant protection of patients, we planned an interim analysis of erlotinib activity in the experimental arm; such assessment will be performed early, after 9 weeks of treatment with erlotinib; the threshold will be overcome if 33 patients out of the first 103 assigned to the experimental arm (presumably after 206 patients will have been randomised) will be free from progression 9 weeks after random date. As of December 31st, 2007, 107 patients have been randomised. References 1. D'Addario G, Rauch D, Stupp R, Pless M, Stahel R, Mach N, Jost L, Widmer L, Tapia C, Bihl M, Mayer M, Ribi K, Lerch S, Bubendorf L, Betticher DC. Multicenter phase II trial of gefitinib first-line therapy followed by chemotherapy in advanced non-small-cell lung cancer (NSCLC): SAKK protocol 19/03. Ann Oncol. 2007 Dec 19; [Epub ahead of print] 2. Gridelli C, Gallo C, Ceribelli A, Gebbia V, Gamucci T, Ciardiello F, Carozza F, Favaretto A, Daniele B, Galetta D, Barbera S, Rosetti F, Rossi A, Maione P, Cognetti F, Testa A, Di Maio M, Morabito A, Perrone F; GECO investigators. Factorial phase III randomised trial of rofecoxib and prolonged constant infusion of gemcitabine in advanced non-small-cell lung cancer: the GEmcitabine-COxib in NSCLC (GECO) study. Lancet Oncol. 2007 Jun;8(6):500-12. 3. Karrison TG, Maitland ML, Stadler WM, Ratain MJ. Design of phase II cancer trials using a continuous endpoint of change in tumor size: application to a study of sorafenib and erlotinib in non small-cell lung cancer. J Natl Cancer Inst. 2007 Oct 3;99(19):1455-61. Epub 2007 Sep 25. 4. http://www.clinicaltrial.gov/ct2/show/NCT00349219?term=TORCH&rank=1, accessed January 8th, 2008 5. Garattini S, Bertele' V. Non-inferiority trials are unethical because they disregard patients' interests. Lancet 2007 Dec 1;370(9602):1875-7. Epub 2007 Oct 23. Conflict of Interest:None declared |
|||