MET DNA Alterations in NSCLC—Reply

We thank Mengoli and colleagues for their interest in our article (1) and their comprehensive review of METΔ14-positive pulmonary sarcomatoid carcinomas (PSC) been reported in the literature. We agree with their observation that METΔ14 seems to be associated with the presence of adenocarcinoma component.

Of the 7 METΔ14-positive PSCs in our series, 5 contain adenocarcinoma components admixed with sarcomatous components (giant cell or spindle cell). These include four with adenocarcinoma (TTF-1 positive) and one with adenosquamous cell carcinoma (ADSQ, TTF-1, p40, and p63 positive). These 5 cases can be classified as pleomorphic carcinoma according to the 2015 WHO classification of lung tumors (2). For the other two METΔ14-positive tumors, one shows pure giant cell morphology but exhibits focal positivity for TTF-1. Although focal TTF-1 expression might suggest the differentiation toward adenocarcinoma, we diagnosed this tumor as giant cell carcinoma rather than pleomorphic carcinoma due to the absence of adenocarcinoma morphology. The remaining METΔ14-positive tumor is a pure spindle cell carcinoma that is positive for cytokeratin and negative for TTF-1.

An interesting observation of our cases is the presence of METΔ14 in both epithelial and sarcomatous components in four pleomorphic carcinomas that we analyzed the two components separately (unpublished data), suggesting the clonal origin of the two components. However, in two tumors with concomitant METΔ14 and high-level MET amplification, MET amplification is found in sarcomatous component only but not in epithelial elements (case #4 and #16 in Supplementary Table S2 of the original article; ref. 1). This is in keeping with the monoclonal theory of biphasic tumors (3) and the sarcomatous elements might derive from the carcinoma during the evolution of the tumor.

In keeping with other studies (4) that METΔ14 associates with adenocarcinoma component, 5 of 7 PSCs in our series contain adenocarcinoma component. METΔ14 can also be found in pure giant cell (n = 1) and spindle cell carcinoma (n = 1). Although there is no statistically significant association between the presence of adenocarcinoma component and METΔ14 (Table 1, P = 0.074) in PSC, a trend toward higher frequency of METΔ14 in PSC with adenocarcinoma component is noticed. However, we do demonstrate that the presence of TTF-1 expression is significantly associated with METΔ14 in PSC (P = 0.020). In clinical practice, only small biopsies are available for pathologic evaluation in most of the advanced cancers. The presence of any sarcomatoid component should alert the clinician for the possibility of METΔ14 mutation.

See the original Letter to the Editor, p. 3697