Authors
Hans-Joachim Schmoll
University Clinic Halle (Saale), Halle, Germany
Hans-Joachim Schmoll , Josep Tabernero , Jean Alfred Maroun , Filippo G. De Braud , Timothy Jay Price , Eric Van Cutsem , Mark Hill , Silke Hoersch , Karen Rittweger , David Chen , Daniel G. Haller
Organizations
University Clinic Halle (Saale), Halle, Germany, Vall d'Hebron University Hospital, Barcelona, Spain, Ottawa Regional Cancer Centre, Ottawa, ON, Canada, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy, The Queen Elizabeth Hospital, Adelaide, Australia, University Hospital Gasthuisberg/Leuven, Leuven, Belgium, Kent Oncology Centre, Maidstone, United Kingdom, Dr. Manfred Köhler GmbH/Roche, Freiburg, Germany, Hoffmann-La Roche, Nutley, NJ, University of Pennsylvania, Department of Hematology/Oncology, Philadelphia, PA
Background: In NO16968, XELOX was superior in terms of disease-free survival (DFS) and overall survival (OS) to bolus 5-FU/LV as adjuvant therapy for stage III colon cancer (Schmoll et al. ASCO GI 2012). Three key enzymes appear to have the potential to predict efficacy and/or safety of fluoropyrimidine-based treatment: thymidine phosphorylase (TP), thymidylate synthase (TS), and dihydropyrimidine dehydrogenase (DPD). We evaluated the association between baseline TP, TS and DPD and outcome (DFS and OS).
Methods: Pts with stage III colon cancer received either XELOX (8 cycles, 24w) or bolus 5-FU/LV (Mayo Clinic, 6 cycles, 24w; Roswell Park, 4 cycles, 32w). The primary study endpoint was DFS; secondary endpoints included OS. TP, TS and DPD expression levels were determined in formalin-fixed, paraffin-embedded tissues by RT-PCR, and the median used as a cut-off point: high (above median) vs. low (below median).
Results: The biomarker population included 498 (26%) of 1886 pts entered (XELOX, n=242; 5-FU/LV, n=256). Baseline demographics, tumor characteristics, cancer history and efficacy (DFS and OS) were similar to those in the main study population. Cox regression analysis for DFS (Table). In the XELOX group pts with low DPD and TP levels and a high TP/DPD ratio appeared to have significantly better DFS; this effect was not observed with 5-FU/LV. Subgroup analysis shows that the difference between XELOX and 5-FU/LV was also higher in pts with low DPD levels.
Conclusions: These exploratory findings suggest that tumor DPD and TP RNA levels could be used to predict outcomes of adjuvant treatment with fluoropyrimidine/oxaliplatin combinations, and should be validated prospectively. Analysis of the current dataset is ongoing and further details on potential biomarkers will be available.
Covariate (high vs. low) |
XELOX
|
5-FU/LV
|
HR |
95% CI |
P value |
HR |
95% CI |
P value |
DPD |
2.46 |
1.54–3.94 |
0.0002 |
0.67 |
0.45–1.00 |
0.0522 |
TP |
1.73 |
1.08–2.76 |
0.0229 |
0.80 |
0.53–1.19 |
0.2650 |
TS |
0.90 |
0.57–1.41 |
0.6369 |
0.91 |
0.61–1.36 |
0.6408 |
TP/DPD ratio |
0.60 |
0.37–0.96 |
0.0325 |
0.81 |
0.53–1.24 |
0.3320 |