When to test for BRAF and what are the consequences?
by Dr. Chiara Cremolini, Dr. Armin Gerger and Dr. Guillem Argilés
WHEN TO TEST FOR BRAF AND WHAT ARE THE CONSEQUENCES?
B Y
Dr. Chiara Cremolini MD, U.O. Oncologia Medica II, Azienda Ospedaliero-Universitaria Pisana Istituto Toscano Tumori, Pisa, Italy Dr. Armin Gerger MD, Division of Clinical Oncology, Department of Internal Medicine, Medical University of Graz, Graz, Austria Dr. Guillem Argilés MD, Medical Oncology Department, Vall d’Hebron University Hospital and Vall d’Hebrón Institute of Oncology, Barcelona, Spain
BRAF MUTATION, GENERAL CONSIDERATIONS
• BRAF V600E is found in approx. 5-10% of mCRC • 4% of non-hypermutat CRC • 46% of hypermutant CRC
• Driver mutation
• Considered mutually exclusive with RAS mutations
(though ultra-sensitive NGS platform unveiled concomitancy with minor RAS mut. allele fractions in certain tumors)
ROLE OF BRAF V600E IN LOCALIZED CRC PETACC8 MOSAIC
BRAF testing on limited disease not recommended, prognostic and predictive data are pending to be clarified
(disparate results in different trials)
PROGNOSTIC ROLE OF BRAF MUTATION IN MCRC
Schirripa et al, Int J Canc 2005, Cremolini et al, Ann Onc 2015
BRAF V600E mutation confers bad prognosis in the metastatic setting
BRAF mutations confer bad prognosis
Different biology of BRAF mutations
PROGNOSTIC ROLE OF BRAF MUT. IN mCRC (2)
The trend towards bad prognosis still persist in BRAF V600E metastatic patients undergoing resection
Schirripa et al, BJC 2015 Yaeger et al, Cancer 2014
BRAF wt (n=178) mRFS=11.0 m BRAF mut (n=23) mRFS=7.0 m p=0.084
PREDICTIVE ROLE OF BRAF MUTATION IN MCRC
BRAF V600E diminish benefit derived from anti-EGFR MoAbs However intensive strategies using extended cytostatic combinations seem to improve patient outcomes
Rowland et al, Br J Canc 2015, Pietrantonio et al, Eur J Canc 2015, Cremolini et al, Lancet Oncol 2015
TRIBE: subgroup analysis according to RAS/BRAF status
Role of BRAF Mut anti-EGFR predictive factor
* P for interaction
Subgroupstudy
RAS WT / BRAF WT22834910163 52183186292 289
190188
97381218
0.58 [ 0.29 , 1.15 ]
0.62 [ 0.50 , 0.77 ]0.93 [ 0.78 , 1.10 ]
0.68 [ 0.54 , 0.87 ]0.64 [ 0.52 , 0.79 ]0.41 [ 0.30 , 0.55 ]0.37 [ 0.24 , 0.55 ]0.71 [ 0.57 , 0.88 ]0.68 [ 0.51 , 0.90 ]
0.86 [ 0.61 , 1.21 ]
0.69 [ 0.32 , 1.49 ]1.25 [ 0.81 , 1.94 ]
1.40 [ 0.82 , 2.39 ]0.34 [ 0.09 , 1.24 ]0.76 [ 0.19 , 3.08 ]0.67 [ 0.34 , 1.29 ]
1402
24324 6
69372240168 183
0.2 0.5 1 2 5
HR (log scale)
Favours Cmab/Pmab Favours comparator
2350
31
3829
1415
RAS WT / BRAF WT
Test for effect: P < 0.001Heterogeneity: l 2 = 82%, P < 0.001
Test for effect: P = 0.38Heterogeneity: l 2 = 39%, P = 0.13
PRIMECRYSTAL and OPUS
CRYSTAL and OPUS
CO.17
PICCOLO20050181
20020408
20020408
COINSummary:
PRIME
CO.17
PICCOLO20050181COINSummary:
Sample size of Tx groups PFS hazard ratio [95% Cl]
Cmab/pmab Comparator
Figure 3. Forest plot of the progression-free survival benefit with anti-EGFR mAb therapy for subgroups defined by tumour RAS and BRAFmutations. Cmab¼ cetuximab; MT¼mutant; Pmab¼panitumumab; WT¼wild type.
Subgroupstudy
RAS WT / BRAF WT228349101183186317 310
19018897381218 0.74 [ 0.57 , 0.96 ]
0.84 [ 0.71 , 1.00 ]0.52 [ 0.37 , 0.71 ]
0.81 [ 0.70 , 0.95 ]
0.90 [ 0.46 , 1.76 ]
0.97 [ 0.67 , 1.41 ]1.18 [ 0.76 , 1.81 ]0.64 [ 0.32 , 1.28 ]1.84 [ 1.10 , 3.08 ]0.84 [ 0.20 , 3.58 ]0.62 [ 0.36 , 1.06 ]
0.01 [ 0.84 , 1.22 ]0.83 [ 0.64 , 1.07 ]0.91 [ 0.74 , 1.13 ]
1364
24324
372245164 184
0.2 0.5 1 2 5HR (log scale)
Favours Cmab/Pmab Favours comparator
57233163829
1384
RAS WT / BRAF WT
Test for effect: P = 0.009Heterogeneity: l 2= 64%, P = 0.02
Test for effect: P = 0.88
Heterogeneity: l 2 = 53%, P = 0.06
PRIMECRYSTAL and OPUSCO.17PICCOLO20050181COINSummary:
PRIMECRYSTAL and OPUSCO.17PICCOLO20050181COINSummary:
Sample size of Tx groups OS hazard ratio [95% Cl]Cmab/Pmab Comparator
Figure 2. Forest plot of the overall survival benefit with anti-EGFR mAb therapy for subgroups defined by tumour RAS and BRAF mutations.Cmab¼ cetuximab; MT¼mutant; Pmab¼panitumumab; WT¼wild type.
BRAF V600E and anti-EGFR mAb benefit in mCRC BRITISH JOURNAL OF CANCER
www.bjcancer.com | DOI:10.1038/bjc.2015.173 5
the clinical course of the disease. Indeed, since patientswith all-RAS-wt CRC constitute about half of all cases,it should be pointed out that BRAF-mut cancers are far
from being underrepresented in that population –increasing to up to 15–20% of RAS-wt CRC. Thus,nearly one in every five patients with RAS-wt statuscould be treated with first-line doublet chemotherapyplus anti-EGFR without a clear demonstration of bene-fit based on data in the literature data and this meta-analysis. We believe that the toxicity and socio-sanitarycosts of anti-EGFR treatments does not justify the useof C and P in BRAF-mut CRC in front of a clinicallynot meaningful, and statistically not significant increaseof outcome end-points, as observed in this meta-analysis.
From this point of view, intensification of first-linetherapy has shown encouraging results in terms of out-come in a phase II [32], non-randomised trial that inves-tigated the triplet chemotherapy combination(FOLFOXIRI regimen) with bevacizumab. Moreover,a subgroup analysis according to the BRAF mutationwas recently reported in patients enrolled in the phaseIII TRIBE study who were treated with FOLFOXIRIplus bevacizumab or FOLFIRI plus bevacizumab [33].Even if BRAF mutation was a poor prognostic factorin both arms and the number of patients was low, theFOLFOXIRI plus bevacizumab arm seemed to improvePFS and OS in BRAF-mut subgroup, as compared to
Fig. 2. Forest plots showing hazard ratio for overall survival for anti-epidermal growth factor receptor (EGFR) treatment in BRAF-mutatedcolorectal cancer patients.
Fig. 3. Forest plots showing hazard ratio for progression-free survival for anti-epidermal growth factor receptor (EGFR) treatment in BRAF-mutated colorectal cancer patients.
0.2 0.5 1 2 5
0
0.2
0.4
0.6
0.8
1Hazard Ratio
SE(log[Hazard Ratio])
Fig. 4. Funnel plot of hazard ratio for overall survival (horizontalaxis) and the standard error for the hazard ratio (vertical axis) for anti-EGFR treatment in BRAF-mutated colorectal cancer patients. Eachstudy is represented by one circle- the vertical line represents thepooled effect estimate.
592 F. Pietrantonio et al. / European Journal of Cancer 51 (2015) 587–594
FIRST GENERATION OF BRAF THERAPEUTICS
Contrary to melanoma, initial trials with BRAF V600E inhibitors s/a failed to demonstrate clinical activity in mCRC
BRAF INHIBITORS + EGFR INHIBITORS HAVE IN VIVO ACTIVITY IN BRAF V600E MUTATED CRC XENOGRAFTS
The genetic context is different in CRC. EGFR constitutive expression lead to a feedback crosstalk with BRAF downstream
effectors that functionally rescue BRAF inhibition
Corcoran RB et al, Cancer Discov 2012; 2:227-35, Prahallad A et al, Nature 2012; 483:100-3
SECOND GENERATION BRAF THERAPEUTICS
Regimen N PR/CR (%) SD (%) mPFS (m)
Dabrafenib + Panitumumab 20 10% 80 3.4 Van Cutsem WGIC 2015
Encorafenib + Cetuximab (ph II) 26 11% 54 (53) 3.7 Elez WGIC 2015
Vemurafenib + Cetuximab 26 4% 16(40) 3.7 Hyman NEJM 2015
THIRD GENERATION BRAF INHIBITORS COMBOS
ALPELISIB
Regimen N PR/CR (%) SD (%) mPFS (m) Dabrafenib + Trabetinib + Panitumumab 35 26% 50 4.1
Van Cutsem WGIC 2015
Encorafenib + Cetuximab + Alpelisib (ph II)
28 (49) 32% 44 4.3
Elez WGIC 2015
Vemurafenib + Cetuximab + CPT 11 ONGOING
DIFFERENTIAL DEGREE OF MODULATION OF PERK BY VARIOUS TREATMENTS IN BRAF V600MUT CRC AND MELANOMA
Even so… numbers are still far distant fom those seen in melanoma. We have a long and fascinating way to walk
CONCLUSIONS
• BRAF testing can not be recommended in localized setting
• BRAF V600E testing should be perform at the debut of metastatic disease, based on:
• Bad prognostic implications
• Need from intensive chemotherapy combos to overcome bad outcome
(FOLFOXIRI-bevacizumab)
• Less benefit from anti-EGFR monoclonal antibodies
• Refer patients to trials including BRAF inhibitor combos with anti-
EGFR monoclonal antibodies