3203 - Exploratory analysis based on tumor location and early metabolic tumor response of REACHIN, a randomized double-blinded placebo-controlled phase II trial of regorafenib after failure of gemcitabine/platinum-based chemotherapy for advanced and metastatic biliary tract tumors.

Background

REACHIN trial met its primary endpoint of improving PFS in patients with biliary tract cancer (BTC) treated with regorafenib (R) as compared to placebo (P): Median (m) PFS for R was 3.0 months (mo) and 1.5 mo for P (HR = 0.49; 95% CI: 0.29-0.81, p = 0.005)1. We exploratory analyzed the benefit of R according to tumor location (TL): intra-hepatic (ICC), extra-hepatic (ECC), gallbladder (GB) or peri-hilar (PH). The early metabolic response (EMR) by treatment was evaluated by FDG PET/CT.

Methods

TL were recorded at randomization of 66 patients in REACHIN (NCT02162914). We analyzed mPFS (95%CI) according to TL. EMR was evaluated comparing FDG tumor metabolism at baseline and after 2 weeks of treatment in the 2 arms if there was at least one FDG positive lesion at baseline, with a Tumour/Liver-uptake-ratio ≥1.4 (MIMvista software).

Results

Table:

743P

mPFS data are summarized in the table For PET study (29 patients, P n = 15/R n = 14), both ΔSUVmax, ΔMTV and ΔTLG (%change from baseline) show a significant decrease between baseline and FU scan in R (P = 0.0151*;0.037**; 0.0056**) as compared to P. Main tumour SUVmax was significantly decreased in R but not in P(P = 0.0353* vs 0.5614). Both TLG and MTV significantly increased in the P (p = 0.0004***; 0.0002***; ) while a stabilization was observed for R (P = 0.3910; 0.2412). There was no relevant treatment’s related impact on the reference-liver-uptake (P = 0.381(P); 0.6149(R)).

Conclusions

ICC location was predominant (64%). mPFS in P arm is similar according to TL. Despite exploratory value and small number of patients per tumor location, R seems superior to P in ICC, GB and PH. Despite limited number of patients, FDG PET/CT seems performant to discriminate EMR in patients treated with R as compared to P. Additional analysis are ongoing to estimate if these EMR also correlate with PFS and OS.

Clinical trial identification

NCT02162914.

Editorial acknowledgement

CUB Hôpital Erasme (Brussels, Belgium). Support from Bayer HealthCare. REACHIN is a BDGO cooperative study.

Legal entity responsible for the study

CUB Hôpital ERASME, Brussels, Belgium.

Funding

Bayer Healthcare.

Disclosure

A. Demols: Honoraria (self), Honoraria (institution), Advisory / Consultancy, Research grant / Funding (self): Bayer Healthcare. All other authors have declared no conflicts of interest. All other authors have declared no conflicts of interest.