Validity of using immunohistochemistry to predict treatment outcome in patients with non-small cell lung cancer not otherwise specified.

Abstract

Histology samples are important for the appropriate administration of tumor type-specific cytotoxic and molecular-targeted therapies for the treatment of non-small cell lung cancer (NSCLC). When biopsy samples lack a definite morphology, a diagnosis can be selected from three subtypes based on immunohistochemistry (IHC) results, as follows: favor adenocarcinoma (ADC), favor squamous cell carcinoma (SQC), or not otherwise specified (NOS)-null. In terms of patient outcome, however, the validity of IHC-based classifications remains unknown. A large series of 152 patients with advanced NSCLC whose diagnoses had been made based on morphological findings and who had been homogeneously treated were enrolled. We used IHC staining (TTF-1, SP-A, p40, and CK5/6) to examine tumor samples and refined the diagnoses. We then analyzed the pathological subgroups according to the IHC staining results. IHC profiling resulted in 50% of the cases being classified as favor ADC, 31% being classified as favor SQC, and 19% being classified as NOS-null groups. Compared with the favor ADC and favor SQC groups, the NOS-null group had a significantly poorer outcome. Pemetrexed-containing platinum regimens produced a response rate similar to that of other platinum doublet regimens in the favor ADC group (44% vs. 46%), whereas it produced a poorer response in the favor SQC group (0% vs. 52%) and the NOS-null group (0% vs. 24%). The favor ADC group tended to have a higher percentage of EGFR positivity and ALK positivity than the favor SQC group (25% vs. 11% and 7% vs. 0%, respectively). These findings support the use of immunohistological subtyping of NSCLC biopsy specimens to select patient-appropriate treatments.