Plenary Session: Lung Cancer in Never Smokers

The day started if with Dr. Thun from the American Cancer Society. He reviewed environmental factors contributing to Lung Cancer in never smokers. He started by reminding us that although only 10% of lung cancer deaths in men and 15-20% of lung cancer deaths in women are due to nonsmoking cancer, the burden of suffering caused by non-smoking lung cancer is actually rather high. If non-smoking lung cancer were treated as its own disease, separate from smoking-driven lung cancer, it would rank eighth among the most common fatal cancers in America! He reviewed environmental factors known to cause lung cancer: secondhand smoke, radon, asbestos, certain metals, some organic chemicals, radiation, air pollution, tubercoulosis, and other chronic inflammatory conditions. Others exposures likely also play a role, but have yet to be proven: human papilloma virus and chronic inhalation of cooking fumes and incense. Indoor air pollution from cooking, coal burning, and smoking men may explain the extraordinarily high rate of nonsmoking lung cancer among women in some areas of Northern China.

Dr. Pierre Massion of Vanderbilt took the stage second to talk about the molecular pathogenesis of never smokers. He reminded us of the different histologic tendencies of never smokers—less SqCC, more adenocarcinoma including the multiple subtypes once called BAC. He reviewed genes associated with susceptibility including cyp1a1, gstm1, xrcc1, gpc5, and fam38b. He pointed out the role of genetic differences in key molecules in inflammatory pathways: IL-1b, IL6, and IL1RN. Finally, insults from the environment may be expressed differently based on variations in genetic susceptibility.

Dr. Massion then used the figure, reproduced below, from Pao et al, Lancet Oncology 2011 to remind us how far we have come in understanding the molecular drivers in nonsmoking cancer:

Studies have shown particular genomic signatures in never smokers. But not only is the DNA changed, but DNA modifiers (epigenetics) are also changed and we have defined specific genes whose expression is modified.

Dr. Caicun Zhou from Shanghai, China discussed lung cancer in never-smoking Asian women. He noted that never smoking cancer is particularly common in Asian women. In Japan 76% of female patients are never smokers and 89% of Korean women with lung cancer are never smokers. Lung cancer incidence is increasing in China. The 5p15.33 locus is associated with the risk of lung adenocarcinoma in Asian women. Exposure to second-hand smoke may account for 20% of lung cancers in non-smokers. Indoor air pollution is also important in China—coal consumption for cooking, coal dust, indoor cooking oil and second-hand smoke all seem to influence risk. EML4/ALK rearrangement seems les common in Asian patients than EGFR mutations, with most studies showing rates of around 5%. Many studies have shown the benefits of EGFR TKIs in patients with the EGFR mutation. The better survival of women than men with lung cancer may be driven by higher frequency of EGFR mutation. I’m struck by two thoughts. First, much suffering from lung cancer in China could be prevented by efforts to improve indoor air quality. Second, we are making real advances in understanding how never-smoking works and the EGFR and EML4/ALK stories have already given us reason to believe that these advances in understanding can lead to better therapies.

Dr. Yang of the Mayo Clinic next spoke about prognostic implications of lung cancer in never smokers. In a Mayo series of 300 patients, ALK rearrangements occurred in about 10% of patients. Interestingly, prevalence of ALK depended on the definition used. When IHC 2-3+ was used as the definition, the prevalence was 10.7%, but when fish positivity was used, positivity was at 8.2%. The group at Mayo next looked at prognosis in a group of patients treated before crizotinib existed (meaning that none of them got crizotinib, allowing a look at the prognostic effects of alk alone, outside of drug therapy aimed at it). No matter how ALK positivity was defined and no matter how adjustment for other factors was done, patients with ALK had worse progression-free survival. In particular, they had more spread to brain (HR 4.75) and liver (HR 6.9). At Mayo, they have seen a few patients with both EGFR mutation and EML4/ALK rearrangement. Another Mayo study looked at SNPs and found several markers influencing prognosis. Multiple genes seem to modify risk in the EML4/ALK patient.

Morning Oral Session for Medical Oncology

Dr. Gitlitz presented phase II data from a randomized phase II study of erlotinib plus either apricoxib or erlotinib in the 2^nd or 3^rd line. Patient selection was done in an interesting way. Patients were treated for five days with apricoxib, and then were only eligible for the study if PGE-M in urine decreased. It is not clear to me what urinary decrease in PGE-M really means, although Dr. Gitlitz speculated that it reflects activity of the drug in the tumor. OS was not improved in the overall cohort (6.4 vs. 7.4 months, not significant) although there did seem to be an improvement in patients < 65 years of age, where median survival improved from 3.8 to 8.5 months, with HR 0.4, p.025. Of note, the negative overall results are similar to a study of celocoxib presented at ASCO 2010, where there was no improvement in treatment outcomes. On the other hand, elderly patients may have dragged down the results from poor tolerance leading to lack of significant exposure to the drug, leaving open the possibility that younger patients could benefit. Although a phase III study is planned, I think that there may be more promising studies being planned that I’d rather open for my patients.

We have one more negative study to discuss before we move on to some exciting data. At ASCO 2010, we saw the negative results of the trial of the anti-IGFR drug figitumumab, when combined with carboplatin and taxol—the trial was stopped early for both excess early deaths and futility. Dr. Rosell presented data from a randomized phase II study of erlotinib combined with either placebo or dalotuzumab for 2^nd or 3^rd line patients. Overall survival was higher with erlotinib alone at 12.4 months, than with dalotuzuman, at 6.9 months. In fairness, I doubt that dalotuzumab really hurt patients; the 12.4-month average survival in 2^nd to 3^rd line is unusually high; 6.9 months is more typical for pretreated patients. Although IGFR expression was evaluated as a biomarker, patient numbers were extremely small and there was signal suggesting benefit in a subpopulation.

Dr. David Spigel described the final results of a phase II study of erlotinib plus either MetMAb or placebo. As a reminder, MetMAb blocks HGF signaling. IHC (staining) played a key role in this trial. “Met diagnostic positive,” was defined as the majority of tumor cells with moderate or strong staining intensity on IHC. Co-primary endpoints were PFS in the overall study and PFS in Met diagnostic positive patients. The most common side effect that was different between placebo and MetMAb was peripheral edema, which was mostly low-grade and reversible. PFS and OS were improved, but only in the met diagnostic group. In this subgroup, the addition of MetMAb increased PFS from 1.5 months to 2.9 months, with a HR of.53, and p.04. OS improved from 3.8 months to 12.6 months, with HR .37 and p.002. These results held for both patients with and without EGFR mutations. In the diagnostic negative group, there was slight harm in both PFS and OS. Data was shown supporting that 50% is the right cutoff, and that IHC (simple staining) is the right test to determine who will benefit. A phase III study testing MetMAb and erlotinib in Met diagnostic patients is planned to start later this year and I think that there’s real reason to be excited.

Disease flare is defined as rapid acceleration of disease progression after discontinuation of gefitinib or erlotinib. In this study, it was more specifically defined as flare significant enough to lead to hospitalization or death before initiation of another treatment. On the GRACE forums, we frequently discuss this phenomenon, as well as the potential benefits of continuing a TKI after progression. MSKCC has many trials of new agents for acquired resistance. Because these trials have a mandated washout period of 7-21 days after stopping initial TKI treatment, they have a large population of patients with EGFR mutation who were taken off TKI for a time before entering their next trial (61 patients in this series). 14 of the 61, or 23% had disease flare, including 3 deaths. Disease flare was associated with a shorter TTP on the original TKI (9 months vs. 15 months). Other clinical factors included escalating symptoms, CNS disease, and pleural disease. Dr. Chaft concluded that clinical trials for patients with acquired resistance should eliminate the required washout period and that for patients not on trial should continue their TKI while adding chemotherapy. I agree entirely with the first point, and, in fact, plan to modify a protocol in development based on this data. Regarding the second point, I think that the question remains unanswered and that both cessation of TKI and continuation with pharmacodynamics separation are both reasonable. As a final point, this data makes coming off tarceva (when you have mutation) much scarier than my experience and that of other oncologists here suggests; I don’t believe that sever flair is really this common.

Dr. Shinji Atagi presented the results of Lux-Lung 4, a phase II trial of afatinib in advanced NSCLC previously treated with erlotinib or gefitinib. Afatinib is in oral irreversible inhibitor of HER1 (EGFR), HER2, and HER4 with preclinical (lab) activity against the T790M resistance mutation. The trial enriched for patients with acquire resistance by requiring that patients have been previously treated with at least 12 weeks of TKI. In this population, there was clear activity, with ORR of 13%, DCR 72%, and a waterfall plot showing most patients below the 0% line. PFS was 4.4 months. Median overall survival was 10.6 months. 80% of patients received some kind of chemotherapy or radiation therapy after the study.

The next study, presented by Dr. Horn of Vanderbilt was also about afatinib. This phase Ib study treated patients with EGFR mutation and T790 leading to resistance. 28 patients were treated, 24 of these at the final maximal-tolerate dose. The final dose was 500mg/m² every two weeks of cetuximab with afatinib 40mg daily. 2/3 of the patients are still on therapy. The response rate was 35%, with DCR of 95%. Responses are seen both with and without T790M mutation. AS 2/3 of patients are still on study, PFS and OS data are immature. A phase II study is planned and I am excited about this combination.

In the afternoon, most of the conference participants are headed to social excursions. I’ve chosen to instead spend the time writing this blog, which is now done. I have an hour or two left before the Fun, Young, Lung Club dinner (founded by one Dr. Jack West). I may be its youngest member, but I need a bit of sleep before I can be any fun!