• © 2005 by American Society of Clinical Oncology

In Reply:

There are many potential explanations as to why the INTACT (Iressa NSCLC Trial Assessing Combination Treatment) trials of the epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) gefitinib (Iressa) in combination with platinum-based doublet chemotherapy did not improve efficacy in patients with advanced non-small-cell lung cancer (NSCLC).^1,2 Dr Normanno correctly points out in his letter that preclinical studies have often shown synergy between platinum drugs and gefitinib, but conflicting data are also available in the literature. He proposes scheduling as a method to prevent a negative interaction between gefitinib and chemotherapy, as suggested by some preclinical studies.

The rationale for combining an EGFR-TKI with chemotherapy led to similar studies with erlotinib,^3,4 which again were negative, though activity was demonstrated in a subgroup of “never smoking” patients. Negative studies in NSCLC investigating other novel agents in combination with doublet chemotherapy, such as the matrix metalloprotease inhibitor prinomastat,⁵ have also been reported. However, in colon cancer, combination of the monoclonal antibody cetuximab with irinotecan does improve response and progression-free survival.⁶ Whether this is something specific either to colon cancer or the use of an antibody awaits further study.

Although the studies on concurrent gefitinib and chemotherapy have been the first to be completed, other studies employing sequential designs have been ongoing for some time. In the time-to-progression curves of the four EGFR-TKI combination studies, there seems to be a negative interaction (or no additive effect) in the initial part of the curves. After approximately 4 to 6 months from random assignment the curves tend to diverge, possibly suggesting that EGFR-TKIs may have a role when given after chemotherapy as a maintenance approach. This type of sequential approach is being investigated in two large phase III studies comparing gefitinib with placebo as maintenance treatment following chemotherapy (Southwest Oncology Group trial 0023 and European Organisation for Research and Treatment of Cancer trial 08,021), as well as in postadjuvant therapy completion in a lower disease setting (National Cancer Institute of Cancer-led trial BR19). In addition, there are ongoing phase I/II trials evaluating sequencing of gefitinib with chemotherapy cycles.

In general, the predictions of preclinical models are rather unsatisfactory when it comes to combination studies, and even where “intermittent” scheduling was used (eg, 5 days of treatment, 2 days without), data were not reported on pharmacokinetic profiles to show if there was a decline of gefitinib to baseline levels. Furthermore, it is difficult to test combinations of more than two drugs in preclinical models. The suggestion that Dr Normanno makes of running randomized phase II studies before embarking on large randomized phase III studies is well taken. Although these studies do not provide sufficient power to detect likely differences between arms, they may at least indicate the most promising strategy. Alternatively, a robust single-arm phase II study with more narrow CIs could also serve this purpose.

A major problem in the four negative randomized studies performed in advanced NSCLC was the unselected patient population. The results now need to be placed in the context of new information on markers of sensitivity to these agents. Activating mutations in the TK domain of EGFR appear to confer sensitivity to gefitinib,^7,8 and their role when these agents are combined with chemotherapy needs to be investigated. It is hoped that mutational studies and analysis of specimens from these trials may identify groups of patients who would benefit from an EGFR-TKI and chemotherapy given concurrently or sequentially.