FLEX: A randomized phase-3 of cisplatin/vinorelbine with or without cetuximab in advanced NSCLC

Dr. R. Pirker presented the results from their international, multicenter trials during the plenary session at ASCO this past weekend. Dr. Thomas Lynch from Harvard was the discussant. The abstract may be viewed here.

This study of 1,125 patients evaluated the efficacy of combining the EGFR monoclonal antibody, cetuximab (Erbitux, ImClone), with cisplatin/vinorelbine chemotherapy in the first-line treatment of patients with advanced NSCLC. An important inclusion criterion was that only patients with EGFR-detectable NSCLC were included (immunohistochemistry). The researchers found a statistically significant difference in their primary endpoint of overall survival after 868 events had occurred. The median overall survival between the two arms was 11.3 months vs. 10.1 months (HR=0.87, 95% CI, 0.76-0.996). Dr. Pirker noted that secondary endpoint analysis was ongoing and would be expected to continue for several more months as additional events occur. The FLEX results are shown here graphically:

FLEX Results:

	N=	RR	PFS	OS
Cis/Vin	568	29%	4.8	10.1
Cis/Vin/C225	557	36%*	4.8	11.3*

* P< 0.05

This study merited a plenary session at ASCO, I believe, because of its size and that a statistically significant survival advantage was found with the monoclonal antibody. Previously Rosell, Butts, and Herbst had combined C225 with platinum-based chemotherapy, in smaller studies, and found improvements in survival that did not reach statistical significance. In addition, previously reported chemotherapy combination studies with EGFR-TKI’s have not proved beneficial and in some cases detrimental.

In his discussion, Dr. Lynch noted that the researchers reported a 22% rate of febrile neutropenia. He questioned the unusually high rates as even the chemotherapy-alone arm had a febrile neutropenia rate of 15%. Both of these are unacceptably high in the setting of metastatic NSCLC. He confirmed that IHC testing for EGFR should be considered standard before using C225 in NSCLC. Dr. Lynch concluded that the average FLEX patients received 18 weeks of cetuximab at a cost of between $54,000 and $62,208. Based on a 1.2 month median overall survival benefit, the cost per year life gained is $540,000 to $622,080; far higher than the accepted standard of $50,000 to $100,000. Finally, he concluded that cetuximab could be used with chemotherapy in front-line treatment of squamous cell NSCLC, bevacizumab (Avastin, Genentech) ineligible patients and with chemotherapy other than cisplatin/vinorelbine.

Let me preface by saying that I have great respect for Dr. Lynch and his body of work. I do, however, question his conclusion about the broad use of this drug when, by his own arguments, it was expensive, toxic and conferred only a 1.2 month survival advantage. The survival curves separate at six months and then come back together again at 24 months. In my mind, prolonging life for one month with a drug requiring weekly administration that comes at an extraordinary cost is irresponsible. Perhaps if the curves remained separated and there were a persistent benefit, even for a small number of patients, I would have more enthusiasm for the combination. However, there is evidence to suggest an even greater survival advantage is conferred simply by educating patients about their disease and symptom management. I respectfully disagree with Dr. Lynch’s conclusions and would welcome more discussion on this topic from the readers.