Abstract 1225: Aberrant SMAD3 signaling in tumor-associated fibroblasts modulates fibrosis and response to the anti-fibrotic drug nintedanib in non-small cell lung cancer

A hallmark of non-small cell lung cancer (NSCLC) is a fibrotic/desmoplastic stroma rich in activated fibroblasts, which are critical regulators of cancer progression, response to therapies and radiotherapy resistance. Paradoxically, we recently reported that the important pro-fibrotic TGF-β transcription factor SMAD3 was epigenetically down-regulated through promoter hypermethylation in tumor associated fibroblasts (TAFs) from NSCLC patients compared to patient-matched control fibroblasts. In addition, we reported that the clinically approved antifibrotic drug nintedanib elicited a stronger inhibition of both the fibrotic phenotype and its associated tumor-promoting effects in TAFs from adenocarcinoma (ADC) patients compared to squamous cell carcinoma (SCC) patients upon TGF-β1 stimulation in vitro, which was consistent with the selective therapeutic response to nintedanib observed in a clinical trial in ADC (but not SCC) patients. Altogether, these previous results suggest that TGF-β1 signaling may be altered in lung TAFs, and that such alteration may depend on their histologic subtype. In this study we addressed these questions by determining the expression and activity of SMAD3 and its closely related homologue SMAD2 in patient-derived TAFs and paired control fibroblasts, and by dissecting their potential contribution to the differential therapeutic responses to nintedanib observed in ADC and SCC. In vitro studies revealed a marked SMAD3 epigenetic repression through promoter hypermethylation, a low pSMAD3/pSMAD2 ratio and a limited fibrotic phenotype selectively in SCC-TAFs. In contrast, ADC-TAFs overexpressed a panel of fibrotic markers upon TGF-β1 stimulation concomitantly with a high pSMAD3/pSMAD2 ratio and a limited SMAD3 promoter methylation. Histologic analysis of a large patient cohort (112 ADC, 96 SCC) confirmed that the extent of fibrosis is larger in ADC than SCC patients. In addition, knocking-down SMAD3 in ADC-TAFs was sufficient to reduce the antifibrotic and antigrowth effects of nintedanib in vitro and in tumor xenografts in vivo. On the other hand, long-term exposure of pulmonary fibroblasts to cigarette smoke condensate was sufficient to hypermethylate the SMAD3 promoter. Since SCC and ADC tumors typically arise in the upper airways and distal pulmonary sites, respectively, it is conceivable that fibroblasts might be more exposed to the smoking epigenetic effects on SMAD3 in SCC. In summary, we report for the first time that tumor fibr