Page 160 - Read Online
P. 160

Battaglin et al. J Cancer Metastasis Treat 2018;4:12  I  http://dx.doi.org/10.20517/2394-4722.2018.04                        Page 5 of 25

               disease is radically resected, BRAF-mutated tumors tends to relapse early with extra-hepatic lesions [42,43] .
               A specific carcinogenesis pathway  and a distinct gene signature  have also been associated with BRAF
                                            [44]
                                                                        [45]
               V600E mutation. More recently, gene expression analyses allowed to identify two different BRAF V600E
               subtypes in a large cohort of BRAF V600E mutated patients unselected for tumor stage: the BM1 subtype
               characterized by KRAS/AKT activation, mTOR/4EBP deregulation and EMT, and the BM2 subtype
               characterized by cell cycle and checkpoint pathway deregulation . In contrast with BRAF V600E mutation,
                                                                     [46]
               metastatic tumors harboring rare mutations of BRAF codons 594 and 596 (less than 1% of CRCs) have been
               shown to have different prognosis and clinical outcome. These rare mutations are associated with a non-
               mucinous histology, a rectal primary tumor location, microsatellite stability, and lack of peritoneal disease.
               Moreover, no negative prognostic impact was observed although in a small series of patients (median OS
               62.0 vs. 12.6 months; HR, 0.36; 95% CI, 0.20-0.64; P = 0.002 for BRAF 594 or 596 mutant vs. BRAF V600E) .
                                                                                                        [47]
               Similar results on the impact and characteristics of BRAF nonV600E mutations were confirmed in a recent
               retrospective evaluation of a large cohort of patients .
                                                           [48]

               Although still debated, growing evidence is accumulating on the role of BRAF mutations as a negative
               predictive marker for anti-EGFR agents activity. Retrospective series showed that the response rate to
               anti-EGFR treatment with or without chemotherapy was significantly lower in  BRAF-mutated  vs. WT
               patients [22,23,49] . On the other hand, BRAF V600E mutation failed to demonstrate its predictive value in several
               sub-group analyses of phase III trials, possibly because of the small number of BRAF-mutated patients and
               lack of statistical power [24,50] . More recently, two meta-analyses showed a lack of improvement in PFS and OS
               in patients with BRAF-mutated mCRCs when treated with either cetuximab- or panitumumab-containing
               regimens compared to chemotherapy alone [51,52] . Additionally, a retrospective evaluation of the randomized
               phase III FIRE-3 trial, comparing FOLFIRI plus cetuximab or bevacizumab as first-line treatment in KRAS
               exon 2 WT mCRC patients, confirmed poorer survival outcomes for BRAF-mutated tumors irrespective
               of cetuximab and bevacizumab administration . Based on these data, it appears that anti-EGFRs do not
                                                        [53]
               demonstrate a clear outcome benefit in BRAF-mutated tumors, and their use should be restricted to patients
               with no alternative therapeutic options. Notably, however, in FIRE-3 cetuximab arm a small subgroup of
               BRAF-mutated tumors achieving an early tumor shrinkage ≥ 20% (9/17) showed significantly longer median
               PFS (9.0 vs. 1.9 months, log-rank test P = 0.002; HR = 0.14) and OS (29.8 vs. 5.9 months, log-rank test P =
               0.047; HR = 0.3) than those not achieving it . Despite the limitations due to the retrospective nature of this
                                                    [53]
               evaluation and the small patients numbers, these results highlight a significant heterogeneity among BRAF-
               mutated mCRCs warranting further investigation.


               While  FOLFOXIRI  plus  bevacizumab  represents the most  promising  treatment  option  in  the  first-line
               setting for clinically selected BRAF-mutated patients [2,54] , outcomes are still unsatisfactory. An extensive
               effort has been made in the last few years aiming to develop possible effective anti-BRAF strategies for
               mCRC patients. In contrast to melanoma, the use of BRAF inhibitors, such as vemurafenib and dabrafenib,
               as single-agents did not show significant activity in BRAF-mutated mCRC . Dual blockade of BRAF and
                                                                               [55]
               alternative survival pathways, such as MEK and EGFR, have been tested as well in clinical trials without
               convincing results [56-58] . Promising results are coming instead from a triple inhibition strategy combining
               BRAF-inhibitors, MEK-inhibitors and EGFR-inhibitors [59,60] . An additional strategy under study to increase
               the activity of dual targeted BRAF inhibition is its association with standard cytotoxic chemotherapy, such
               as the combination of vemurafenib with cetuximab plus irinotecan which have been explored in the SWOG
               1406 trial with encouraging results . Moreover, several other promising strategies designed to overcome
                                             [61]
               resistance pathways to BRAF-inhibitors are currently under investigation [62,63] . Final results from ongoing
               trials are warranted to improve targeted treatment options for BRAF-mutated patients.


               Microsatellite Instability
               MMR is a highly conserved DNA repair mechanism that ensures genomic integrity by correcting mispaired
               or unpaired bases which have escaped the proofreading activity of DNA polymerases during DNA replication
   155   156   157   158   159   160   161   162   163   164   165