EGFR tyrosine kinase inhibitors are commonly used in the United States for the treatment of patients with advanced BAC.8,17
Erlotinib and gefitinib have both demonstrated response rates of 10% to 22% and a median survival of 13 to 17 months in patients with BAC.10,11,18
Our study was designed to evaluate the efficacy of targeting the external domain of the EGFR with cetuximab for the treatment of advanced BAC or adenocarcinoma with BAC. The overall response rate with cetuximab was low at 7%, but the median time to progression and overall survival were comparable to that of the EGFR tyrosine kinase inhibitors in advanced BAC. In general, responses are difficult to assess in BAC because this disease has unique radiographic features that make it hard to assess the anticancer effects of new agents in BAC.19
Even progression-free survival could be influenced by the varying course of disease based on individual subtypes of BAC and the frequency of imaging studies. We considered both of these factors in choosing the end points for this study. Another limitation of our study is that information on poststudy therapy was not collected. It is likely that many patients received an EGFR tyrosine kinase after progression that could have had an effect on the overall survival results.
An important strength of this study is that the pathologic diagnosis was confirmed by an independent pathologist in 45 of the 49 patients with available tumor tissue. We have reported the results separately for those with a confirmed diagnosis and for all treated patients. At the time of conception of this study, the presence of activating mutations in the EGFR
was not known.20
mutation analysis was done post hoc in nearly two thirds of the participants. Interestingly, only one of the six patients with an activating mutation in the EGFR
experienced an objective response with cetuximab. This is in stark contrast to erlotinib, which was associated with a response rate of more than 80% in patients with BAC who had an EGFR
The differential sensitivity of EGFR
-mutated NSCLC cells to tyrosine kinase inhibitors and monoclonal antibodies has been reported in a preclinical study.21
Compared with gefitinib, cetuximab had relatively minimal effect on inhibition of phosphorylation of downstream signals in vitro. The low response rate with cetuximab in patients with EGFR
mutations in this study provides clinical affirmation for these preclinical observations. One patient with a KRAS
mutation experienced a partial response in our study, which is also different from the tyrosine kinase inhibitors. These important observations highlight the differential mechanism(s) of anticancer effects between cetuximab and the tyrosine kinase inhibitors. Of note, KRAS
mutations are frequently present in advanced BAC.
This study also substantiates the observation from prior studies that objective responses with EGFR inhibitors are not seen in mucinous BAC.17
Nearly a third of the patients enrolled onto this study had the mucinous subtype, among which there were no partial responses. It is notable that none of the patients with mucinous BAC who were treated with EGFR tyrosine kinase inhibitors on the three studies reported to date had a complete or partial response. Mucinous BAC tends to present with a multicentric pattern and lobar consolidation, and it is often associated with debilitating symptoms.22 EGFR
mutations are not frequent in mucinous BAC. It will be necessary to study new classes of anticancer agents for this subgroup of patients.
Though cetuximab has enhanced efficacy when used with chemotherapy, no predictive markers have been described for patient selection in NSCLC. A recent analysis of tumor specimens from the FLEX study23
noted a correlation between efficacy with cetuximab and EGFR protein expression. A composite score of EGFR expression based on the number of positively stained cells and the intensity of staining was used and, accordingly, a score of ≥ 200 was predictive of a higher response rate with chemotherapy in combination with cetuximab. The correlation with progression-free survival and overall survival has not been reported. If this observation can be validated, it could potentially help with patient selection for cetuximab therapy. As was observed in the FLEX study,13
we also found a favorable correlation between the severity of skin rash and efficacy of cetuximab, though it was not statistically significant. The biologic reasons behind this observation are not entirely clear.
In conclusion, cetuximab has modest efficacy in the treatment of advanced BAC and is likely effective in a different molecular subset than the EGFR tyrosine kinase inhibitors.