BACKGROUND
Patients with stage II-III colorectal cancer (CRC) treated with adjuvant chemotherapy, gain a 25% survival benefit. In the context of personalized medicine, there is a need to identify patients with CRC who may benefit from adjuvant chemotherapy. Molecular profiling could guide treatment decisions in these patients. Thymidylate synthase (
TYMS
) gene polymorphisms,
KRAS
and
BRAF
could be included in the molecular profile under consideration.
AIM
To investigate the association of
TYMS
gene polymorphisms,
KRAS
and
BRAF
mutations with survival of CRC patients treated with chemotherapy.
METHODS
A retrospective study studied formalin-fixed paraffin-embedded tissues (FFPEs) of consecutive patients treated with adjuvant chemotherapy during January/2005-January/2007. FFPEs were analysed with PCR for the detection of
TYMS
polymorphisms, mutated
KRAS
(m
KRAS
) and BRAF (m
BRAF
). Patients were classified into three groups (high, medium and low risk) according to 5’UTR
TYMS
polymorphisms Similarly, based on 3’UTR polymorphism ins/loss of heterozygosity (LOH) patients were allocated into two groups (high and low risk of relapse, respectively). Cox regression models examined the associated 5-year survival outcomes.
RESULTS
One hundred and thirty patients with early stage CRC (stage I-II: 55 patients; stage III 75 patients; colon: 70 patients; rectal: 60 patients) were treated with surgery and chemotherapy. The 5-year disease free survival and overall survival rate was 61.6% and 73.9% respectively. 5’UTR polymorphisms of intermediate
TYMS
polymorphisms (2RG/3RG, 2RG/LOH, 3RC/LOH) were associated with lower risk for relapse [hazard ratio (HR) 0.320,
P
= 0.02 and HR 0.343,
P
= 0.013 respectively] and death (HR 0.368,
P
= 0.031 and HR 0.394,
P
= 0.029 respectively). The 3’UTR polymorphism ins/LOH was independently associated with increased risk for disease recurrence (
P
= 0.001) and death (
P
= 0.005). m
BRAF
(3.8% of patients) was associated with increased risk of death (HR 4.500,
P
= 0.022) whereas m
KRAS
(39% of patients) not.
CONCLUSION
Prospective validating studies are required to confirm whether 2RG/3RG, 2RG/LOH, 3RC/LOH, absence of ins/LOH and wild type
BRAF
may indicate patients at lower risk of relapse following adjuvant chemotherapy.