Histopathologic response and growth patterns of colorectal cancer liver metastases (CRCLM) in patients treated with triplets plus bevacizumab (bev) or anti-EGFRs.

Abstract

636 Background: The histopathological response to pre-operative chemotherapy is associated with clinical outcome in patients (pts) undergoing secondary resection of CRCLM. Three different histopathological growth patterns (HGPs) of CRCLM have been described: desmoplastic (i.e. with a capsule of stroma separating tumor and normal cells), pushing (i.e. with limited infiltration of normal hepatic plates by tumor cells) and replacement (i.e. with abundant infiltration of normal hepatic plates by tumor cells and vessel cooption). Methods: Histopathological parameters of response were evaluated in specimens from 159 pts who underwent secondary resection of CRCLM, after receiving triplet (FOLFOXIRI or COI) plus bev (N = 103) or anti-EGFR (N = 56) in 5 first-line clinical studies (TRIBE, MOMA, MACBETH, COI-B and COI-E). The aims of this analysis were to evaluate the prognostic role of histopathologic response and to investigate the prognostic role of HGPs and their potential different sensitivity to targeted agents. Results: When compared with partial (TRG 3) and no (TRG 4-5) pathologic response (N = 118), major response (TRG 1-2, N = 41) was associated with longer RFS (mRFS: 28.0 versus 11.0 mos, HR = 0.57, 95%CI = 0.39–0.86; p = 0.007) and OS (mOS: unreached versus 42.1 mos, HR = 0.54, 95%CI = 0.31-0.93; p = 0.027). No association of baseline clinical characteristics and RAS and BRAF status with major response was found. Major response was more frequent among pts treated with bev than with anti-EGFRs (OR = 2.83, 95%CI = 1.20–6.65; p = 0.015) and was associated with deepness of response as a continuous variable (HR = 1.02, 95% CI = 1.00–1.04; p = 0.011). In the desmoplastic HGP (N = 28) a higher percentage of major response was reported (57% vs 17% in pushing and 22% in replacement HGP, p < 0.001) and a numerical advantage from anti-EGFR vs bev was evident in terms of both major response and RFS. Conversely, in the pushing HGP (N = 66) a significant benefit from bev vs anti-EGFR in major response and RFS was observed. No difference was described in the replacement HGP (N = 65). Conclusions: The assessment of HGPs may be useful to predict benefit from available targeted agents.