Article Text

Download PDFPDF

329 Prolonged overall survival (OS) in patients with metastatic colorectal cancer (mCRC) in SPICE, a phase I study of enadenotucirev in combination with nivolumab
  1. Tom Lillie1,
  2. Andrew Stone2,
  3. Sarah Lockwood3,
  4. Richard Brown1,
  5. Andrew Fox1,
  6. Eirini Bournazou4 and
  7. John Beadle1
  1. 1PsiOxus Therapeutics Ltd, Abingdon, UK
  2. 2Stone Biostatistics Ltd, Crewe, UK
  3. 3Pivotal Statistics Ltd, Macclesfield, UK
  4. 4Bristol-Myers Squibb, Lawrenceville, NJ, USA


Background SPICE is a phase 1 study of the oncolytic adenovirus enadenotucirev in combination with nivolumab in patients with advanced epithelial tumors (NCT02636036). Preliminary data has indicated a survival advantage in patients with mCRC resulting in a median OS of 14 months.1 To further understand this OS signal, a comparison to historical patient-level data from the placebo arm of the CORRECT study (NCT01103323) was performed using data obtained from Project Data Sphere.

Methods Individual patients from SPICE were matched with patients in the placebo arm of the CORRECT study in terms of covariates known to be associated with OS (ECOG, presence of liver mets, haemoglobin, albumin, LDH and platelet count). The OS outcomes were then compared between the matched SPICE and CORRECT patients to minimise any bias due to patient selection. The distribution of the covariates was broadly similar between studies with minor differences favouring the SPICE study.

Results The mOS in confirmed microsatellite stable mCRC patients (n=25) in the SPICE study was 15.4 months (95% CI; 11.8 m, 21.0 m) as compared to 5.0 months for patients in the placebo arm of the CORRECT study (n=251).Two different statistical analyses were performed to compare the outcomes between studies: (1) A comparison of OS matching each SPICE patient to a maximum of 10 (average of 5.5) placebo patients from CORRECT using M:1 variable nearest neighbour propensity score matching; (2) Multivariate analysis of SPICE vs CORRECT adjusting for all covariates in a Cox regression model.The Hazard Ratio (SPICE:CORRECT) from the regression model was 0.28 with an upper 2-sided 95% confidence limit of 0.48, which was consistent with results using propensity score matching. The upper 95% CL for the HR for method (2) was 0.61.

Conclusions The results appear promising, particularly in a population that has historically shown little response to PD-1 intervention and warrant further exploration in a randomised study. However, these analyses cannot be regarded as definitive, due to the possible presence of unmeasured confounders between a small phase 1 cohort and a large phase 3 control group.

Ethics Approval The study was approved by the Western Institutional Review Board, study approval number 1160755.


  1. Fakih M, Wang D, Harb W, Rosen L, Mahadeven D, Berlin J, Basciano P, Arogundade O, Cox C, Krige D, McElwaine-Johnn H, Brown R, SPICE: A phase I multicenter study of enadenotucirev in combination with nivolumab in tumors of epithelial origin: an analysis of the metastatic colorectal cancer patients in the dose escalation phase. Annals of Oncology 2019 30 (suppl_5)::198-v252. 10.1093/annonc/mdz246

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.