Thanks for posting this great webinar.

Although it’s very tangential to the presentation, I wanted to come back to Dr. Weiss’ statement that squamous tumors should not be tested for EGFR mutations. I’ve seen similar statements elsewhere, and they always make me shudder. My mom’s tumor was read as squamous, poorly differentiated, at 2 institutions (including Mayo, which I have to believe I can trust). After much prodding, we managed to get mutation testing done since she met all the epidemiologic criteria. And it turns out that she has an exon 19 mutation and has had a major response to tarceva.

I know that tarceva is always an option in the 2nd + line of therapy, but it isn’t necessarily going to be everyone’s 2nd line, or even everyone’s third line, and so I really worry about people missing an opportunity to experience longer, higher quality life. Obviously no one is going to find EGFR mutations in squamous tumors if no one is looking for them, so my question is whether enough squamous tumors have been assessed to issue a broad statement about not testing them for EGFR mutations. In other words, is our case really one in a million, or did we just hit upon 2 sets of pathologists who came to incorrect conclusions? Is it possible that testing a subset of people with squamous tumors might be higher-yield (perhaps those that meet the clinical criteria? or those with moderately or poorly differentiated cells and a higher chance that they were mis-read? or something else?)? Tarceva has been such a miracle for us. Though I understand the cost-benefit issues with testing, I hate to think of others missing an opportunity for a similar miracle.

Thanks,
Anna

Anna,

I think it’s more likely that her cancer is mixed and has some squamous areas and some adenocarcinoma, or it really is a (very) rare squamous tumor with an EGFR mutation. There have been hundreds and maybe thousands of squamous tumors tested for EGFR and other mutations (increasingly as part of a comprehensive battery), so I really don’t think it’s that these are actually more common than we appreciate. It’s more common that the cancer may be adeno in some locations and squamous in another, and someone has to be that 1 in a few hundred that is read as squamous but has a mutation.

But I think the real point is that the message, in my opinion, shouldn’t be that everyone with a squamous tumor gets tested for an EGFR mutation, because you’re going to need to do several hundred tests to find one positive. Meanwhile, if oncologists just use Tarceva as a second or third line treatment, a drug that is approved for squamous NSCLC and is on a short list of agents that we should consider after first line (alimta no longer on that list for squamous histology), it really obviates the value of testing for the mutation. Our health care costs are crippling in the US, and they’re rising at a completely untenable pace. It’s hard to decide whether and how to pay for very expensive treatments that add some additional value. I don’t see that it’s necessary to add considerable expense to do something that shouldn’t necessarily change management anyway if oncologists are following the data. And beside that, we do need to consider whether the extreme exception should change the policy for the vast majority of cases that are not, in fact, the exception.

There are many things that might be desirable to do if resources were infinite, but they’re not. There are people desperately struggling to cover their health care premiums, and various programs are being cut to pay for rapidly escalating costs of medical care. I just don’t think it’s feasible to do every test and every treatment out there that we can’t rule out being 100% value-less.

Dr. Weiss,

Correct me if I’m wrong, but I believe the RADIANT trial allows all patients to be tested but only the patients with EGFR mutations are then randomized 2:1 to Tarceva:Placebo.

No, radiant allows patients with or without mutations.

Dr. Weiss-

Excellent webinar- very relevant for me personally. Any idea when preliminary results from the RADIANT trial will be released? Will they do a sub-analysis on EGFR mutation positive patients who choose to have chemo prior to starting tarceva versus those who start tarcevca with no chemo?

Emily

RADIANT is still ongoing, so we certainly shouldn’t expect any real data at this year’s ASCO. Further, it takes time for adjuvant data to mature so we shouldn’t expect even preliminary data until at least a year after trial closure.

You ask a great question, that I really hope that we have the privilege of debating: If erlotinib adds a lot to survival as an adjuvant therapy for mutation + patients, will chemo still have benefit? I doubt that RADIANT will have the statistical power to answer this question if erlotinib is a success. I don’t know what portion of RADIANT patients will be mutation +, but it would take many such patients in each “group” to make a real comparison. If erlotinib is a success in this context, even a real benefit to chemo MAY be smaller. Plus, the patients who are given chemo in addition to erlotinib may not be the same as those who choose to not take chemo–high risk features may make a patient, together with her oncologist, choose in favor of the chemo.

Dr. Weiss,

According to clinicaltrials.gov, RADIANT only randomizes EGFR-mutation-positive patients:

RADIANT: A Study of Tarceva After Surgery With or Without Adjuvant Chemotherapy in NSCLC Patients Who Have EGFR-positive Tumors (Adjuvant)

So does this change what you think the impact of the trial would be?

I think you’re misreading/misinterpreting the eligibility, which is very easy to do.

I just checked clinicaltrials.gov, and it confirms that RADIANT only includes patients who are positive for EGFR protein by immunohistochemistry (IHC) or gene amplification by fluorescence in situ hybridization (FISH). but that is far, far less restrictive (about 80-85% of NSCLC patients) than EGFR mutation testing (with only about 10% of patients in North America positive for an EGFR mutation). The EGFR mutation seems to identify the group overwhelmingly likely to benefit from an EGFR tyrosine kinase inhibitor far better than EGFR IHC or FISH testing, so I think it’s likely that even if mutation testing is available for a reasonable sized subset of people on RADIANT, only perhaps 15-20% will have an actual EGFR mutation (the 15-20% eliminated from eligibility because they’re negative for EGFR by IHC and/or FISH would almost certainly all be EGFR mutation negative).

Ah yes, thanks Dr. West for clearing up the point of confusion with regard to RADIANT’s EGFR positivity requirement requirement.

BR.19 study just presented @ ASCO. This study gave adjuvant gefitinib after surgery. Overall, patients randomized to iressa did worse than the control group. These results held for all subgroups evaluated, including the EGFR mutants. Like the results with radiation, these results continue to make us scratch our heads and continue to argue against early adoption of TKIs as adjuvant therapy unless a convincing positive trial is presented or published.

I think these results were among the most interesting of the day in lung cancer, unfortunately in the context of a trial that was very negative and defied our expectations, similar to the SWOG 0023 trial with maintenance Iressa after chemo and radiation. For the SWOG trial, we’ve never seen any breakdown by molecular subtypes, and I think even after the surprising and humbling results from the SWOG trial that showed a significantly harmful effect of Iressa vs. placebo, we would have again presumed that the one group in which these results wouldn’t hold true are patients with an EGFR mutation.

The BR.19 trial of adjuvant Iressa vs. placebo was stopped early, after the very negative results of the SWOG 0023 trial in 2005, and the patients on the BR.19 trial had only received a median of less than 5 months of either Iressa or placebo. This only made me more surprised to see that there could be any real signal of a difference, and yet it was very convincingly non-beneficial, and even quite clearly trending toward harmful, for patients with EGFR mutations.

Overall, then, this underscores how we absolutely cannot presume to know how results will turn out before we do the trials. It is also a strong argument to absolutely not recommend post-operative EGFR inhibitor in the absence of data to support a benefit, even in the patients we might have presumed would be absolutely the most likely to benefit from it, at least outside of a clinical trial.

Doctors:

Is the BR.19 Study the one that was conducted in Canada by Dr. Francis Shephard? And also, is there a study being presented where the participants (including those with EGFR+ mutations) were administered Tarceva rather than Iressa (and would we expect to see different results with Tarceva)?

Laya

A poster presented today also informs this conversation. The poster comes out of MGH from 2 friends, Dr. Joel Neal and Dr. Nate Pennell. Dr. Neal is a friend from residency, and in my opinion, the most promising thoracic oncologist entering the field. You all know Dr. Pennell from his insightful posts on GRACE.

Their study is the phase II trial of adjuvant erlotinib (tarceva) in EGFR mutated patients that I mentioned in the podcast. 36 patients have been accrued so far. The poster was primarily about toxicity—there was no grade IV toxicity, 14% grade III rash, 3% grade III diarrhea, 6% grade III fatigue, 3% grade III pruritis and 3% grade III LFT elevations. After 8.8 months median followup, there have not yet been any recurrences.

These preliminary results stand in contrast to the results from gefitinib with chemorads described in the webinar (SWOG trial) and after cytotoxic adjuvant chemotherapy (BR.19) described in my and Dr. West’s comments above (including in EGFR mutants). Is it an aberrant result in a small phase II study without a control group? Or have Dr. Neal and Dr. Pennell raised the hypothesis that tarceva, in the adjuvant setting, may be a more different drug than iressa than we think? The study has been expanded to 100 patients which should power it well (for a phase II study), and give us more information about whether tarceva should be further studied in the adjuvant setting.