Caratterizzazione molecolare dei tumori: impatto sulla pratica clinica Nicola Normanno ISTITUTO NAZIONALE PER LO STUDIO E LA CURA DEI TUMORI FONDAZIONE G. Pascale – NAPOLI SC Biologia Cellulare e Bioterapie CENTRO RICERCHE ONCOLOGICHE MERCOGLIANO (AV) Laboratorio di Farmacogenomica

Genomic alterations affecting actionable signaling pathways Garraway JCO 2013

Genomics-Driven Oncology Garraway JCO 2013

The major classes of genomic alterations that give rise to cancer Modified from McConaill JCO 2010 EGFR ErbB-2 BRAF KRAS NRAS PIK3CA AKT1 MAP2K1 EML4-ALK ROS-1 RET EGFR ErbB-2 MET Sequencing, Real Time PCR etc. FISH, Immunohistochemistry

Genotyping and genomic profiling in personalized medicine Li JCO 2013

Clinical cancer genomics Dienstmann JCO 2013

Colon and Lung Panel - a single workflow solution Tumor Sample Colon and Lung AmpliSeq ™ panelv2 DNA mutations Ion PGM ™ System semiconductor sequencing Ion Reporter ™ Software automated analysis & reporting Sample to report in less than 36 hours DNA analysis RNA lung fusion research panel ALK, ROS,RET, NTKR1 fusions & expression RNA Analysis KRAS, EGFR, BRAF, PIK3CA, AKT1, ERBB2, PTEN, NRAS, STK11, MEK1, ALK, DDR2, CTNNB1, MET, TP53, SMAD4, FBXW7, FGFR3, NOTCH1, ERBB4, FGFR1, FGFR2

Gene Number of cases (>2%) with mutations, n (%) (N=182 analyzed) TP5372* (39.5%) KRAS 45 ^ (24.7%) 30 at codon 12 or 13 (16.5%); 16 at other (8.8%) PIK3CA 24 § (13.2%) 16 at exon 9 (8.8%); 10 at exon 20 (5.5%) BRAF15 (8.2%)10 at codon 600 (5.5%); 5 at other (2.7%) NRAS13 (7.1%) FBXW79 (4.9%) MET7 (3.8%) 22 multiple gene mutation analysis in mCRC treated with FOLFIRI + cetuximab *7 cases with double TP53 mutation; ^1 case with double KRAS mutation; § 2 cases with double PIK3CA mutation Mutations in genes EGFR, CTNNB1, FGFR3, SMAD4 occurred in 2 cases each (1.1%); mutations in genes ERBB2, FGFR2, PTEN occurred in 1 case each (0.55%) Ciardiello, Normanno et al Ann Oncol 2014

Scarpa A & Normanno N - PlosONE /38 (95%) adequate libraries EGFR KRAS PIK3CA BRAF TP53 STK11 6/36 (16%) 10/36 (28%) 7/36 (18%) 3/36 (8%) 2/36 (5%) 1/36 (3%) 24/36 (67%) at least one 9/36 (25%) multiple Molecular typing of lung adenocarcinoma on cytological samples using a multigene next generation sequencing panel

A patient story in 2014 A 39-year-old woman, was referred to S. Orsola Malpighi oncology unit in November 2013 for stage IV lung cancer. Total body CT scan showed: massive expansive process that occupies much of the middle lobe and basal portion; enlarged lymphnodes both on mediastinum and mesentere; neoformations of both adrenals; presence of two nodular brain. 18 FDG-PET additionally showed intense metabolic activity of bone (D7, L4, 3 and 10 right rib), liver (segment 7) and peritoneum. She underwent to a needle biopsy of the lung with a diagnosis of undifferentiated carcinoma (immunohistochemistry negative for CD 117, synaptophysin, CD30, S100, calretinin, TTF1, smooth muscle actin; partially positive for vimentine ed EMA; positive per citokeratin 8). A second histological revision suggested for a high grade sarcomatoid carcinoma. EGFR and ALK negative

A patient story in 2014 Between November and December 2013 she underwent to a first line of chemotherapy with cisplatin (60 mg/mq), epirubicin (60 mg/mq) and ifosfamide (5000 mg/mq) but after only two cycles has been progressing on all sites of disease Subsequently she was subjected to a second line of treatment with carboplatin (AUC 5) and paclitaxel (175 mg/mq). After three doses, there was a rapid clinical deterioration for worsening of intestinal subocclusion due to carcinomatosis. Request of wide molecular screening at the Laboratory of Pharmacogenomic of CROM with the Ion AmpliSeq™ Colon and Lung Cancer Panel

BRAF mutations and response to BRAF inhibitors in non-small-cell lung cancer Case DM % tumor cells HS 45 BRAF V600E HS 1,25 p.E746_A750Del EGFR Normal BRAF EGFR Unknown

A patient story in 2014 Off-label vemurafenib was started in March 2014 at a dose of 720 mg administered twice per day After 10 days of administration she had an admission to emergency room for an epileptic episode with aphasia associated to hyperpyrexia (38°C). CT scan of the brain showed perilesional edema and she underwent to whole brain radiotherapy After one month of treatment a total body CT scan showed: dramatic reduction of the process of the right middle lobe (8x6.7 cm vs 12x7.7 cm) and reduction in number and dimension of limphnodes and other metastatic sites The patient suddenly died on May 2014 due to progression of cerebral lesions

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

Survival of patients with driver mutations and matched therapy Kris JAMA 2014 Tsimberidou CCR 2014

Molecularly targeted therapy vs conventional therapy: the SHIVA trial Le Tourneau Lancet Oncol 2015

BRAF mutations in melanoma, CRC and NSCLC CancerFrequencyPrognosticPredictive § %V600E Melanoma50%Y/NY90% CRC10%YN90% NSCLC3% Y*Y50% *only V600E § for response to cIass I RAF kinase inhibitors

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

OS of EGFR mutant NSCLC patients treated with afatinib Yang Lancet Oncol 2015 Exon 19 deletionsL858R

BRAF mutations and prognosis in melanoma Bucheit Cancer 2013 All BRAF mutant patientsBRAF mutant treated with BRAFi or MEKi BRAF mutant not treated with BRAFi or MEKiNRAS mutant patients

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

Heterogeneity of EGFR mutations 16/42 (38%) EGFR mutant samples consisted of only EGFR-mutated cells, whereas the other 26 samples consisted of cells with both wild type and mutated EGFR, with the proportion of EGFR-mutant cells ranging from 30% to 90% Among the 37 wild type cases, four (10.8%) showed low mutant frequency ranging from 7.69% to 20.83% Bai JCO 2012

Detection of EGFR mutations in NSCLC Sequencing vs Therascreen Zhou JCO 2011

PFS according to mutant allele frequency (MAF) of p.L858R EGFR mutation Ono Ann Onc 2014 ORR was significantly higher in the group with MAF >9% (79.1%) than in the group with MAF ≤9% (20%) (P = 0.022, Fisher’s exact test).

The Allelic Frequencies of Baseline BRAF V600 Mutations Did Not Impact PFS The median AF of BRAF V600 mutations is 33.9% in coBRIM PFS Based on BRAF Allele Frequency Days Percentage Progression Free Vem AF ≤33.9% (n = 101) Vem AF >33.9% (n = 105) Vem + cobi AF ≤33.9% (n = 99) Vem + cobi AF >33.9% (n = 95) st Quartile 2nd Quartile 3rd Quartile 4th Quartile Days PFS by BRAF AF Quantile (all patients) 26 Percentage Progression Free AF, allelic frequency; PFS< progression-free survival; WT, wild type.

The heterogeneity score (HS) was obtained by normalizing the frequency of mutant alleles for the fraction of neoplastic cells Heterogeneity Score (HS) in mCRC patients enrolled in the CAPRI trial Normanno N, et al. Ann Oncol 2015;26:1710–1714

Genotype of low (<33) KRAS HS tumors ID Patient KRAS HS Score Additional Mutations ,00NONE ,29PIK3CA ex 20, BRAF V600E, FBXW ,14PIK3CA ex 9 and 20, ERBB2, TP ,33NONE ,86PIK3CA ex9, TP53 (2 different mutations) ,71PIK3CA ex9, BRAF ex11, TP ,57NONE ,00PIK3CA ex9, TP ,00FGFR ,00TP53 Normanno et al Ann Oncol 2015

Heterogeneity Score (HS) and efficacy of treatment in the CAPRI trial Case 177 (SD, PFS 5,9 mo) 70% tumor cells HS 14,29 KRAS G13D HS 17,14 PIK3CA ex20 HS 54,29 BRAF V600E HS FBXW7 R465H 48,6 Normal BRAF KRAS PIK3CA Case 118 (PR, PFS 3,9 mo) 70% tumor cells HS 22,86 KRAS G12D HS 74,29 PIK3CA ex9 HS 57,14 TP53 Normal TP53 KRAS PIK3CA FBXW7 Normanno et al Ann Oncol 2015

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

Camidge Nat Rev Clin Oncol 2014 Mechanisms of acquired biological resistance to EGFR TKIs in NSCLC

The "three levels" system of resistance to anti-EGFR moAbs in mCRC Normanno Nat Rev Clin Oncol 2009 Level 1: the antenna  EGFR S492R mutation  ERBB2 gene amplification  MET gene amplification ErbB-2 MET Level 2: the main switch  KRAS mutations  NRAS mutations Level 3: the internal circuits  BRAF?  PIK3CA?

Melanoma escape pathways during disease progression on BRAF inhibitor therapy Shi Cancer Discov 2015

Resistant cells are selected by treatment with target based agents Sensitive Resistant Before Target Therapy After Target Therapy

Multiple resistance mechanisms to EGFR TKIs Blakely ASCO 2013

Melanoma heterogeneity and branched evolution during the acquisition of BRAF inhibitor resistance Shi Cancer Discov 2015

Detection of different mechanismsm of resistance to anti-EGFR moAbs in plasma of CRC patients Siravegna Nat Med 2015Bettegowda Sci Transl Med 2014

Challenges in genomics-driven oncology Alterations of “actionable” oncogenic pathways are not always predictive of response to targeted agents Different molecular alterations of the same oncogene may not be equivalent Intra-tumor heterogeneity may affect response to targeted agents The molecular profile of solid tumors may significantly change following treatment with target based agents Genomic testing programs should be strongly linked to matched clinical trials

Clinical trial enrollment after genomic testing Meric-Bernstam JCO 2015 Challenges to trial accrual included:  patient preference of noninvestigational treatment or local treatment  poor performance status or other reasons for trial ineligibility  lack of trials/slots  insurance denial

Genomics-Driven Oncology Garraway JCO 2013 Surgeon Endoscopist Radiologist Surgeon Endoscopist Radiologist Medical Oncologist Medical Oncologist Pathologist, Molecular Biologist, Geneticist Medical Oncologist Surgeon Radiotherapist