The prognostic and predictive role of class III β-Tubulin and hENT1 expression in patients with advanced pancreatic ductal adenocarcinoma

BACKGROUND

FOLFIRINOX and gemcitabine-nabpaclitaxel (GnP) are standard first-line treatment regimens for advanced pancreatic ductal adenocarcinoma (PDAC). However, currently, there is a lack of predictive biomarkers to aid in the treatment selection. We aimed to explore the prognostic and predictive value of class III β-Tubulin (TUBB3) and human equilibrative nucleoside transporter 1 (hENT1) expression, which have previously been shown to be associated with taxane and gemcitabine resistance in advanced PDAC.

METHODS

We conducted a retrospective analysis of 106 patients with advanced PDAC treated with GnP and/or FOLFIRINOX at our institution. TUBB3 and hENT1 immunohistochemical staining was performed on tumor specimens and subsequently evaluated based on the intensity and percentage of expression.

RESULTS

In patients who received the GnP regimen, a high combined score (TUBB3low/hENT1high) was associated with a higher DCR and longer PFS compared to those with intermediate (TUBB3high/hENT1high or TUBB3low/hENT1low) and low score (TUBB3high/hENT1low). In the multivariate analysis, a high combined score was an independent predictor of higher DCR (OR:11.96; 95 % CI:2.61-54.82; p = 0.001) and longer PFS (HR:0.33; 95%CI:0.18-0.60; p < 0.001). However, there was no difference in response rates or PFS based on TUBB3 and hENT1 expression among patients receiving the FOLFIRINOX regimen.

CONCLUSION

Our findings indicate that tumor TUBB3 and hENT1 expression may predict the efficacy of the GnP regimen, and low TUBB3 and high hENT1 expression (TUBB3low/hENT1high) are associated with a higher DCR and longer PFS in patients treated with GnP. Evaluating TUBB3 and hENT1 jointly can identify the patients most (as well as least) likely to benefit from GnP chemotherapy.

Prognostic and predictive value of human equilibrative nucleoside transporter 1 (hENT1) in extrahepatic cholangiocarcinoma: a translational study

Introduction: Effective (neo) adjuvant chemotherapy for cholangiocarcinoma is lacking due to chemoresistance and the absence of predictive biomarkers. Human equilibrative nucleoside transporter 1 (hENT1) has been described as a potential prognostic and predictive biomarker. In this study, the potential of rabbit-derived (SP120) and murine-derived (10D7G2) antibodies to detect hENT1 expression was compared in tissue samples of patients with extrahepatic cholangiocarcinoma (ECC), and the predictive value of hENT1 was investigated in three ECC cell lines. Methods: Tissues of 71 chemonaïve patients with histological confirmation of ECC were selected and stained with SP120 or 10D7G2 to assess the inter-observer variability for both antibodies and the correlation with overall survival. Concomitantly, gemcitabine sensitivity after hENT1 knockdown was assessed in the ECC cell lines EGI-1, TFK-1, and SK-ChA-1 using sulforhodamine B assays. Results: Scoring immunohistochemistry for hENT1 expression with the use of SP120 antibody resulted in the highest interobserver agreement but did not show a prognostic role of hENT1. However, 10D7G2 showed a prognostic role for hENT1, and a potential predictive role for gemcitabine sensitivity in hENT1 in SK-ChA-1 and TFK-1 cells was found. Discussion: These findings prompt further studies for both preclinical validation of the role of hENT1 and histochemical standardization in cholangiocarcinoma patients treated with gemcitabine-based chemotherapy.

hENT1 Expression Predicts Response to Gemcitabine and Nab-Paclitaxel in Advanced Pancreatic Ductal Adenocarcinoma

PURPOSE

Modified FOLFIRINOX (mFFX) and gemcitabine/nab-paclitaxel (GnP) remain standard first-line options for patients with advanced pancreatic ductal adenocarcinoma (PDAC). Human equilibrative nucleoside transporter 1 (hENT1) was hypothesized to be a biomarker of gemcitabine in the adjuvant setting, with conflicting results. In this study, we explore hENT1 mRNA expression as a predictive biomarker in advanced PDAC.

EXPERIMENTAL DESIGN

COMPASS was a prospective observational trial of patients with advanced PDAC. A biopsy was required prior to initiating chemotherapy, as determined by treating physician. Biopsies underwent laser capture microdissection prior to whole genome and RNA sequencing. The cut-off thresholds for hENT1 expression were determined using the maximal χ2 statistic.

RESULTS

253 patients were included in the analyses with a median follow-up of 32 months, with 138 patients receiving mFFX and 92 receiving GnP. In the intention to treat population, median overall survival (OS) was 10.0 months in hENT1high versus 7.9 months in hENT1low (P = 0.02). In patients receiving mFFX, there was no difference in overall response rate (ORR; 35% vs. 28%, P = 0.56) or median OS (10.6 vs. 10.5 months, P = 0.45). However, in patients treated with GnP, the ORR was significantly higher in hENT1high compared with hENT1low tumors (43% vs. 21%, P = 0.038). Median OS in this GnP-treated cohort was 10.6 months in hENT1high versus 6.7 months hENT1low (P < 0.001). In an interaction analysis, hENT1 was predictive of treatment response to GnP (interaction P = 0.002).

CONCLUSIONS

In advanced PDAC, hENT1 mRNA expression predicts ORR and OS in patients receiving GnP.

Evaluation of NUC-1031: a first-in-class ProTide in biliary tract cancer

PURPOSE

NUC1031 is a first-in-class ProTide, that is a gemcitabine pro-drug designed to overcome putative mechanisms of resistance, including decreased expression of hENT/hCNT transporters, absence of activating enzymes such as deoxycytidine kinase (dCK) and presence of degrading enzymes such as cytidine deaminase (CDA). We undertook comprehensive pre-clinical evaluation of NUC1031 in biliary tract cancer (BTC) models, given that gemcitabine/cisplatin is a standard first-line therapy in advanced BTC.

METHODS

Here, we compared the in vitro activity of NUC1031 in comparison to gemcitabine, validate putative mechanism(s) of action, assessed potential biomarkers of sensitivity or resistance, and performed combination studies with cisplatin. We also evaluated the in vivo efficacy of NUC1031 and gemcitabine using a CDA-high cholangiocarcinoma patient-derived xenograft (PDX) model.

RESULTS

In a panel of BTC cell lines (N = 10), NUC1031 had less potency than gemcitabine in multiple cellular assays. NUC1031 did not demonstrate evidence of greater synergy over gemcitabine in combination with cisplatin. Surprisingly, efficacy of both gemcitabine and NUC1031 was not found to be correlated with hENT/hCTN, dCK or CDA transcript levels. Gemcitabine and NUC1031 showed equivalent efficacy in a CDA-high PDX model in vivo contradicting the primary rationale of NUC1031 design.

CONCLUSION

NUC1031 did not exhibit evidence of superior activity over gemcitabine, as a single-agent, or in combination with cisplatin, in either our in vivo or in vitro BTC models. Given that the largest Phase 3 study (ClinicalTrials.gov: NCT0314666) to date in BTC is underway (N = 828) comparing NUC1031/cisplatin to gemcitabine/cisplatin, our results suggest that a more conservative clinical evaluation path would be more appropriate.