DISCUSSION

EFG, when bound to its receptor EFGR, activates two main signaling pathways, RAS-BRAF-MEK-ERK and PIK3CA-AKT, which are responsible for initiating processes of cell progression and differentiation, apoptosis and DNA repair (2,11,17). Mutations in KRAS, BRAF, NRAS genes (via MAPK/ERK) and PIK3CA (via PIK3CA/AKT) have been linked in numerous studies to monoclonal antibodies (Cetuximab and Panitumumab) response, progression and overall survival in CRC (3,8,10,13,14,18–27).
We analyzed the overall survival in 194 CRC patients according to the status of different mutations in components of the EGFR pathway. Our results showed association of the PIK3CA/AKT pathway, but not the MAPK/ERK cascade on CRC overall survival. In particular, we observed an association between risk of death and presence of H1047Y mutation in exon 20 of PIK3CA (HR = 8.93, CI95%, 1.20-66.57, p = 0.03268). We also observed a strong effect of ECOG 0 on greater overall survival (HR = 0.17, CI95%, 0.10 -0.31, p = 1656.10-9). ECOG PS is a scale of 0 to 5 published in 1982 by the Eastern Cooperative Oncology Group that measures how the disease affects the patient’s daily activities. It seems logical to think that patients with ECOG 0 (full functionality) have more chances to survive than those with ECOG 1-2. Given that the effect of ECOG could be masking the effect of other variables on overall survival, we performed a analysis stratifying the sample by ECOG status, which confirmed that patients with PIK3CA20 H1047Y mutation had a higher risk of death (HR = 8.93, CI95%, 1.20-66.57, p = 0.03268) and showed a new association between the E545K mutation of exon 9 ofPIK3CA (HR = 5.49, CI95%, 1.28-23.51, p = 0.021720) and overall survival in patients with ECOG 0.
Our findings are comparable to those obtained by Sartore-Bianchi (19) and De Roock (8). Sartore-Bianchi and colleagues investigated the influence of loss PTEN, PIK3CA exon 9 and 20 and KRAS exon 2 mutations on objective response and survival in 110 metastatic CRC patients treated with anti-EFGR monoclonal antibodies (Panitumumab or Cetuximab). Patients with at least one alteration in the PIK3CAand PTEN genes had a higher risk of progression (HR = 1.86, CI95%, 1.16-2.96, p = 0.009). As observed in our patients, KRASmutations could not predict progression-free survival (HR = 1.50, CI95%, 0.89-2.52, p = 0.128). They also failed to find association with overall survival for any gene (PIK3CA , p = 0.2518; KRAS , p = 0.1127), while in our patients, PIK3CA exon 20 (H1047Y) mutation was associated with lower survival. De Roock explored the possible effect of KRAS, NRAS, BRAF and PIK3CA mutations on objective response, overall survival and progression-free survival of 773 CRC chemotherapy-refractory patients treated with Cetuximab plus chemotherapy. As our results, they found that mutations of exon 20 were capable to predict survival (progression-free and global), but only in patients with KRAS wild-type tumors (HR = 2.27, CI95%, 1.10-4.66, p = 0.042 / HR = 3.30, CI95 %, 1.46-7.45, p = 0.012).
Wendy De Roock, in addition, suggests that PIK3CA exons 9 and 20 mutations should be studied separately, since the alterations they produce are different and can lead to biased results (8). Exon 9 (helical domain) mutation produces the activation of the RAS protein, while exon 20 (kinase domain) mutation induces a gain in function independent of RAS (28). In our study, as De Roock suggests, and unlike the Sartore-Bianchi study, we separately analyzed PIK3CA exons 9 and 20 mutations.
The limited size of our sample did not allow us to explore the effect of the mutations on response to the treatment, since the patients were in different stages of disease and lines of treatment at the time of our study, and the follow-up period (two years) was also insufficient for study progression-free survival in CRC. The statistical power proved to be insufficient to investigate the influence on CCR survival of some of the mutations in the EFGR pathway mutations KRAS (rs121913529, rs121913530, rs121913535, rs112445441, rs17851045, rs121913238, rs121913240), NRAS (rs121913237, rs121434596, rs11554290, rs121913254) and BRAF (rs113488022)] and no mutations were observed on those genes. Therefore, further studies with larger size will be necessary to properly explore and elucidate those genes as potential biomarkers in CRC survival. Despite the limited sample size, we clearly demonstrated the influence of PIK3CA gene mutations on CCR survival.