Disclosures Institutional Research Grants: Roche- Genentech, - - PDF document

11/5/19 1

David R. Gandara, MD University of California Davis Comprehensive Cancer Center

Emerging Role of Liquid Biopsy in Precision Medicine: Non-Small Cell Lung Cancer as a Model

1

Disclosures

• Institutional Research Grants: Roche-

Genentech, Novartis, Merck

• Consultant/Advisory Board: AstraZeneca,

Celgene, CellMax, FujiFilm, Roche-Genentech, Guardant Health, Inivata, IO Biotech, Lilly, Merck, Samsung Bioepis

2

11/5/19 2

NSCLC

100 patients with Advanced Stage Non-Small Cell Lung Cancer (NSCLC): They all look alike, but they are not

3

Non-SqCC

100 patients with Advanced Stage Non-Small Cell Lung Cancer (NSCLC): They all look alike, but they are not

SqCC 43 y/o female Never-Smoker Adenoca EGFR mt 55 y/o male Former Smoker Adenoca ROS1 fusion 56 y/o female Former Smoker Adenoca MET ex14 mt 34 y/o female Never Smoker Squamous HER2 mt 65 y/o male Current Smoker Adenoca BRAF V600E mt 52 y/o male Former Smoker Adenoca ALK fusion In 2019 most Oncologist would agree that these are different malignancies. In 2019 most Oncologists would agree that these patients need different therapies

4

11/5/19 3 Evolution & Expanding List of Guideline Recommendations for Genomic Testing in NSCLC

“The NCCN NSCLC Guidelines Panel strongly endorses broader molecular profiling with the goal of identifying rare driver mutations for which effective drugs may already be available, or to appropriately counsel patients regarding the availability of clinical trials. Broad molecular profiling is a key component of the improvement of care of patients with NSCLC).”

Genomic Alteration (i.e. driver event) Available targeted agents with activity against driver event in lung cancer* EGFR mutations

• simertinib, erlotinib, gefitinib, afatinib, dacomitinib

ALK rearrangements alectinib, brigatinib, ceritinib, crizotinib, lorlatinib HER2 mutations ado-trastuzumab emtansine, afatinib BRAF V600E mutations dabrafenib + trametinib, vemurafenib MET amplification/mutation crizotinib ROS1 rearrangements crizotinib, ceritinib RET rearrangements cabozantinib, vandetanib NTRK rearrangements entrectinib, larotrectinib

NCCN Clinical Practice Guidelines. NSCLC. v3.2019.

5

ALK: 65% RR to crizotinib; ~70% RR to 2nd-gen TKI; Ceritinib in resistant cancers Alectinib 1st line MET ex14: 30-40% RR Crizotinib Cabozantinib HER2 mutation: ~15% RR -Afatinib; ~20% -Dacomitinib 44% -Ado Trastuzumab BRAF (V600E): >60% RR to BRAF + MEK inhibitor combo ROS1: 70% RR to Crizotinib

• Ceritinib

HER2 mutation EGFR: RR>70% to 1st-2nd–Gen TKIs; ~60% RR to 3rd-Gen TKIs in resistant cancers

Growing Number of Oncogene-driven NSCLCs with Active Targeted Therapies

RET: Cabozantinib: RR=40% Alectinib NTRK: Larotrectinib 71% RR MET ex14 Capmatinib 60% RR Tepotinib 59% RR KRAS G12C AMG510 48% (11/23) responders Pending: LOXO-292: RR=77% BLU-667: RR = 50%

Gandara: Lung Cancer Summit. ESMO19

6

11/5/19 4

Schema for Multidisciplinary Integration of Biomarker Testing in Advanced Stage NSCLC: Looking for “Actionable” Oncogenes

Pulmonologist Interventional Radiologist Surgeon Identify Target Lesion Biopsy Histology Evaluation Determine Therapy Pathologist Oncologist Multidisciplinary Team (Tumor Board) Molecular Biomarker Testing Identify Patient Referring Physician Med Oncologist Thoracic Surgeon Radiation Oncologist Pulmonologist Radiologist Pathologist When Progression à Re-Biopsy Treat Determine New Therapy

Adapted from: Raez, Gandara et al Clin Lung Cancer 2016

T r e a t

When Progression à Re-Biopsy

Plasma ctDNA Plasma ctDNA

7

Near-Future Approach (Patient-Based Therapy):

Genomic profiling by high throughput next generation sequencing for decision-making in individual patients

Next Generation Sequencing (NGS):

• Whole Genome or Exome capture

Sequencing (DNA)

• Whole or Targeted Transcriptome

Sequencing (RNA)

• Epigenetic profiling

1. Histomorphological Diagnosis:

Cancerous

Evolving Approach (Target-Based Therapy V2.0):

Multiplexed molecular tests with increased sensitivity & output for decision-making in individual patients

Current Approach (Target-Based Therapy V1.0):

Single gene molecular testing for decision-making in individual patients

• 2. Molecular Diagnosis:

Multiplex, Hot Spot Mutation Tests:

• PCR-based SNapShot
• PCR-based Mass Array SNP
• Sequenom

Initial High-Throughput Technologies:

• SNP/CNV DNA microarray
• RNA microarray

Single Biomarker Tests:

• Sanger DNA Sequencing
• RT-PCR
• FISH
• IHC

Representative technologies:

Extract tumor nucleic acids: Archival cancer specimens Archival FFPE tumor specimens Macro- or Micro-dissection

• f Tumors

DNA and RNA

Empiric Approach (Past) (Compound-Based Therapy):

Clinical-histologic factors to select drugs for individual patients

Evolution of Biomarker Testing in NSCLC: Past, Current & Future

from Li, Gandara et al: J Clin Oncol , 2013

à Plasma ctDNA by NGS for Genomics & TMB

8

11/5/19 5

Terminology: Liquid Biopsy

Liquid biopsy (blood sample) CTCs (circulating tumor cells) cfDNA (cell-free DNA) ctDNA (circulating tumor DNA)

Crowley E, et al. Nat Rev Clin Oncol 2013;10:472–484.

9 What can Liquid Biopsy provide in November 2019 for NSCLC? Tumor Genomics & blood-based Tumor Mutational Burden (investigational) Advantages of plasma ctDNA over Tumor biopsy or re-biopsy:

• Indicated when tumor tissue not available or high risk (or “plasma-first” situations)
• Reflects shed tumor DNA into plasma from all tumor sites, providing a “global perspective”
• May abrogate the issue of tissue heterogeneity and undergenotyping due to small sample
• Can determine mechanism of resistance without biopsy, to guide subsequent therapy
• Can be repeated serially (longitudinal assessment) for response & early progressive disease
• Relatively non-invasive & high acceptance rate by patients
• Detection of Minimal Residual Disease (i.e. after surgical resection)

10

11/5/19 6

Patient with Advanced Treatment-naïve NSCLC

From Rolfo, Gandara et al: J Thorac Onc 2018. aEGFR, ALK, ROS1, and BRAF at minimum, but panel if available. bStrongly suggest tissue sparing to facilitate participation in clinical

• trials. cWhile NGS is preferred, based on availability, other validated assays are acceptable.

YES NO NO YES

Molecular profiling on all with non squamous, non squamous component, or if clinical features may suggest a molecular driver Surgical specimen is available Perform molecular analysisa

• n surgical specimenb;

NGS is preferredc; Treat with SOC therapy based on presence or absence of

• ncogenic driver; Perform

PD-L1 IHC as needed Perform molecular analysisa on liquid biopsy (ctDNA); NGS is preferredc Perform molecular analysisa

• n tissue biopsy specimensb;

NGS is preferredc; Treat with SOC therapy based on presence or absence of

• ncogenic driver; Perform

PD-L1 IHC as needed Tissue biopsy specimen is sufficient for molecular testing Therapeutic target positive Treat with SOC therapy based on presence of

• ncogenic driver

Therapeutic target negative Tissue re-biopsy Perform molecular analysisa on tissue biopsy specimensb; NGS is preferredc; Treat with SOC therapy based on presence or absence of oncogenic driver; Perform PD-L1 IHC as needed

11

Patient with NSCLC progressive or recurrent disease during therapy

from Rolfo, Gandara, et al: J Thorac Onc 2018.

acobas/ddPCR for EGFR mutation NGS preferred for ALK and ROS1. bStrongly suggest tissue sparing to facilitate participation in clinical

• trials. cWhile NGS is preferred, based on availability, other validated assays are acceptable.

Targetable resistance mutation absent Perform molecular analysisa

• n tissue biopsy specimenb;

NGS is preferredc; Treat with SOC therapy based on presence or absence of

• ncogenic driver; Perform

PD-L1 IHC as needed Perform molecular analysisa on liquid biopsy (ctDNA) Tissue re-biopsy Targetable resistance mutation present Feasible Not Feasible Evaluate the potential benefit

• f other therapy for marker

unknown or best supportive care Treat with SOC therapy based

• n presence of
• ncogenic driver

12

11/5/19 7

High Circulat ating Tumor (ct ct)D )DNA Det etect ection Rat ate ac across Multiple Can ancer Ty Types (N=21,807)

Guardant360 plasma NGS assay for detection of somatic alterations in 21,807 cancer patients 85% detection rate across all cancers 93% SCLC 87% NSCLC Median VAF: 0.41% (range 0.03-97.6)

NSCLC

Zill, Mack, Gandara, Landman et al, CCR 2018

13

Aggarwal et al: JAMA Oncol 2018

PENN2 Study: Response to Targeted Therapy is Independent of Plasma Mutation Allelic Fraction

Progression of Disease Partial Response

n=42; R=-0.121; p=0.45

14

11/5/19 8 High ORR in Patients With Tumor- or Plasma-Detected, T790M-Positive NSCLC Treated With Osimertinib

Oxnard et al. J Clin Oncol. 2016;34:3375.

15 What can Liquid Biopsy provide in November 2019 for NSCLC? Tumor Genomics & blood-based Tumor Mutational Burden (investigational) Advantages of plasma ctDNA over Tumor biopsy or re-biopsy:

• Indicated when tumor tissue not available or high risk (or “plasma-first” situations)
• Reflects shed tumor DNA into plasma from all tumor sites, providing a “global perspective”
• May abrogate the issue of tissue heterogeneity and undergenotyping due to small sample
• Can determine mechanism of resistance without biopsy, to guide subsequent therapy
• Can be repeated serially (longitudinal assessment) for response & early progressive disease
• Relatively non-invasive & high acceptance rate by patients
• Detection of Minimal Residual Disease (i.e. after surgical resection)

16

11/5/19 9

ctDNA Utility in Under-Genotyped/QNS Non-Squamous NSCLC

383,$30% 879,$68% 26,$2% Tissue$Biomarker$ Positive$ Tissue$QNS$or$UG Tissue$Fully$ Genotyped,$ Biomarker$Negative

Tissue Genotyping Stat atus (N (N=1 =1288) 383 383 of 1288 1288 (30% 30%) Biomar arker Positive fo for Drive ver On Oncogene 879 (68%) Quan antity In Insufficient (QNS NS) or

• r

Un Undergenotyped (UG UG) ctDNA NGS Increas ased Biomar arker Yield ctDNA an anal alysis identified 252 252 ad additional al ac actionab able biomar arkers (19% of 1288) (2 (29% of f 879) no not prev evious usly det etect ected ed in n tissue QNS/UG cas ases

Biomarker N in ctDNA* EGFR 42 KRAS 127 ALK fusion 3 ROS1 fusion 2 RET fusion 14 BRAF V600E 13 MET amp 23 MET E14 7 HER2 mutation 21 TOTAL 252

383,$30% 252,$19% 627,$49% 26,$2% Tissue$Biomarker$ Positive Tissue$QNS/UG,$ ctDNA$Pos$ Tissue$QNS/PG,$ ctDNA$Neg$ Tissue$Fully$ Genotyped,$ Biomarker$Negative

Adapted from Zill, Mack, Gandara, Landman et al, CCR 2018

!

Biomarker N in Tissue EGFR 256 KRAS 61 ALK fusion 27 ROS1 fusion 9 RET fusion 4 BRAF V600E 10 MET amp 10 MET E14 4 HER2 mutation 1 FGFR3 fusion 1 TOTAL 383

17 What can Liquid Biopsy provide in November 2019 for NSCLC? Tumor Genomics & blood-based Tumor Mutational Burden (investigational) Advantages of plasma ctDNA over Tumor biopsy or re-biopsy:

• Indicated when tumor tissue not available or biopsy is high risk (or “plasma-first” situations)
• Reflects shed tumor DNA into plasma from all tumor sites, providing a “global perspective”
• May abrogate the issue of tissue heterogeneity and undergenotyping due to small sample
• Can determine mechanism of resistance without biopsy, to guide subsequent therapy
• Can be repeated serially (longitudinal assessment) for response & early progressive disease
• Relatively non-invasive & high acceptance rate by patients
• Detection of Minimal Residual Disease (i.e. after surgical resection)

18

11/5/19 10 FLAURA: Osimertinib vs Gefitinib/Erlotinib in 1st line therapy of EGFR-mutated NSCLC

Soria et al: NEJM 2017 & Ramalingam et al: ESMO 2019

OS

19

FLAURA: Acquired Resistance Mechanisms in Comparator EGFR TKIs (n=129)

*Resistance mechanism reported may overlap with another; #Acquired T790M + C797S + L718Q: 1%; †PIK3CA + T790M (n=1), PIK3CA + T790M + C797S (n=1), and PIK3CA (n=1)

BRAF D594N: 1% KRAS G12C: 1% NRAS G12D: 1%

Acquired T790M: 47%

Other EGFR mutations*: 1%

HER2 amplification: 2%

MET amplification: 4%

MET amplification + T790M: 2%

PIK3CA mutations: 3%#

HER2 MET MET MET MET EGFR mTOR AKT p53 BIM BCL2

Survival Apoptosis

PIK3CA MEK RAF RAS ERK

Proliferation

CCDC6 RET CCDC6-RET: 2% HER2 HER2 HER2 EGFR

The most common acquired resistance mechanisms were T790M mutation (47%), MET amplification and HER2 amplification

20

11/5/19 11

FLAURA: Acquired Resistance Mechanisms after Osimertinib first-line therapy (n=91)a

• No cases of acquired EGFR T790M
• The most common resistance mechanisms were MET amplification (15%) and EGFR C797X mutation (10%)

– Other mechanisms included HER2 amplification/mutation (3%), PIK3CA(7%), RAS/RAF mutations and ALK transformation

aResistance mechanism reported may overlap with another; bTwo patients had de novo T790M mutations at baseline of whom one acquired C797S at progression

Ramalingam SS, et al. ESMO 2018. Abstract LBA50. Secondary EGFR mutations:b

C797X: 7%; L718Q+C797S: 1%;

L718Q + ex20ins: 1%; S768I: 1%

HER2 amplification: 2%

HER2 mutation: 1%

MET amplification: 15%

mTOR AKT p53 BIM BCL2 PIK3CA MEK RAF RAS ERK MET MET MET MET BRAF mutations (V600E): 3% KRAS mutations (G12D/C, A146T): 3% SPTBN1 ALK SPTBN1-ALK: 1% EGFR EGFR HER2 HER2 HER2 HER2

Survival Apoptosis Proliferation

PIK3CA mutations: 7%

Cell cycle gene alterations CCND amps: 3% CCNE1 amps: 2% CDK4/6 amps: 5%

21

52 yo female, never-smoker, with new Dx of stage IV lung adenocarcinoma, May 2016, lymphangitic spread-R. lung, bone mets

• EGFR Ex19del identified (May 2016).

Initially treated with erlotinib à PR.

• Progressive disease in Feb 2017,

found to be T790M+. Treated with osimertinib à PR.

• In Jan 2018 & May 2018, monitoring

shows absence of Ex19del & T790M.

• In Jan 2019, new cough & SOB.

CT scan read as “stable”, but possible “pneumonia”.

• Plasma ctDNA identifies EGFR C797S resistance mutation.
• Re-read of CT scan consistent with recurrent lymphangitic spread.

Navigational bronchoscopy shows recurrent adenocarcinoma.

• Started on therapy designed for C797S+ disease.
• June 2019: CT scan shows major response. C797S= non-detectable

Case Example: Serial plasma ctDNA monitoring

Case Report Courtesy of David Gandara MD, UC Davis

22

11/5/19 12

FLAURA: Early clearance of plasma EGFR mutations as a predictor of response to osimertinib and comparator EGFR-TKIs

aPresence of plasma EGFR mutations detected by ddPCR; ddPCR, droplet digital polymerase chain reaction

Zhou C, et al. ASCO 2019. Abstract 9020.

• This analysis of FLAURA confirms prior studies

showing that presence of EGFR mutation in plasma ctDNA at baseline is a poor prognostic factor

• Patients with plasma EGFR mutation clearance have

improved PFS

• Clearance of EGFR mutation from ctDNA favors
• simertinib in PFS

PFS based on detection of plasma EGFRma at baseline PFS based on detection of plasma EGFRma at week 6 following initiation of treatment 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 3 6 9 12 15 18 21 24 27 Time from randomization (months) Probability of PFS

Non-detectable Detectable

• No. at risk

147 352 135 323 124 266 114 217 101 162 84 126 42 63 14 20 2 3

3 6 9 12 15 18 21 24 27 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Time from randomization (months) Probability of PFS Osimertinib Comparator EGFR-TKI

134 124 131 118 117 99 108 76 90 47 74 35 40 14 15 3 2 1 Osimertinib Comparator

• No. at risk

PFS in patients with clearance of plasma EGFRma at week 6 after initiation of treatment 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 3 6 9 12 15 18 21 24 27 Time from randomization (months) Probability of PFS Non-detectable Detectable

• No. at risk

258 70 249 63 216 46 184 29 137 21 109 13 54 8 18 2 3 Non−detectable Detectable

Non-detectable Detectable

23

BFAST: Phase 2/3 screening trial in patients with treatment-naïve NSCLC: Initial results from the ALK+ cohort

Gadgeel S, et al. ESMO 2019. Abstract LBA81_PR. Sample (-) for BFAST alteration Sample (+) for BFAST alteration Patients not enrolled in treatment cohorts Complete Closed Alectinib 600 mg PO BID until PD (n=78 planned; 87 actual)

ALK+

Alectinib PO at 900, 1200, or 750 mg BID (n=50–62 planned; 8 actual) RET+ Atezolizumab 1200 mg IV q3w until PD or loss of clinical benefit bTMB+ Randomised 1:1, n=440 Platinum-based chemotherapy for 4 or 6 cycles Entrectinib 600 mg PO daily until PD (n=50) ROS1+ Real World Data Cohort

Physicians will receive overall results from bSMP assay

*All cohorts have additional, treatment-specific inclusion/exclusion criteria

Blood to FMI for cfDNA testing (bSMP and bTMB assays)

Screening inclusion/exclusion criteria*

• Age > 18 years
• Unresectable, stage

IIIB or IV NSCLC

• Measurable disease
• Treatment naïve
• ECOG PS 0–2

Vem + Cobi + Atezo n=25 to futility; up to 80 for primary analysis BRAF+

24

11/5/19 13 BFAST ALK: Results using ctDNA alone

Gadgeel S, et al. ESMO 2019. Abstract LBA81_PR.

25

Causes of “False Negative” & “False Positive” Liquid Biopsy Results (assumes Tissue=Gold Standard)

Adapted from Oxnard et al: JCO 2019

“False Negatives” in Liquid Biopsy Insufficient DNA shed into plasma: (low tumor volume, eliminated by therapy) Technical Issues: Insufficient sensitivity in older assays “False Positives” in Liquid Biopsy Technical Factors: Sample differences (> 6 months from tissue to plasma sampling) WBC contamination: Germline Variants Clonal Hematopoiesis Tumor Heterogeneity: Positive Plasma & Negative Tissue (assumes tissue is “Gold standard”)

26

11/5/19 14

Tumor Mutational Burden (TMB) as a Candidate Predictive Biomarker for Cancer Immunotherapy

• Somatic mutations in cancers are multifactorial (including DNA repair

defects, carcinogens & enzymatic alterations in DNA polymerases)

• These mutations produce neoantigens that induce anti-tumor immune

responses

• TMB is an emerging predictive biomarker for cancer checkpoint

immunotherapy (CIT)

• TMB can be estimated using whole-exome sequencing (WES) or

comprehensive genomic profiling by NGS (e.g., FoundationOne & FACT in blood[bTMB]) . MSK-IMPACT. Guardant OMNI1-8

• Studies show that TMB either by WES or CGP correlate with each other &

with efficacy of CPI therapy in multiple cancer types1-3

• Predicted neoantigen load (NAL), a component of TMB most closely

linked to immune response, correlates with F1 TMB4,5,7

• TMB identifies a distinct patient population not currently captured by

PD-L1 IHC or other immune biomarkers5,6

IHC, immunohistochemistry; PD-L1, programmed death-ligand 1; TMB, tumor mutational burden.

• 1. Yarchoan M, et al. N Engl J Med. 2017; 2. Chalmers ZR, et al. Genome Med. 2017; 3. Goodman AM, et al. Mol Cancer Ther. 2017;
• 4. Efremova M, et al. Front Immunol. 2017; 5. Topalian SL, et al. Nat Rev Cancer. 2016; 6. Kowanetz M, et al. WCLC 2017. 7. Mariathansan,

et al. Nature 2018. 8. Rizvi et al: ESMO IO 2018.

PD-L1 TMB

From Gandara, LeGrand et al: ASCO 2018

27

High Tissue TMB is associated with increased efficacy of Checkpoint Inhibitor Monotherapy

Gandara, Legrand et al: ASCO 2018 WES: CM-026 NSCLC (Nivo -high TMB) Carbone et al: NEJM 2017

PD-L1 + TMB

NGS Foundation-One: Multiple Tumor Types NGS -IMPACT: Multiple Tumor Types Samstein et al: NatGen 2019

28

11/5/19 15

Analytical & Clinical Validation of Tumor Mutational Burden in Blood (bTMB) in association with Atezolizumab efficacy in advanced NSCLC (POPLAR & OAK Trials)

Gandara DR, et al. Nature Med 2018.

OAK Study

29

B-F1RST: Proof of Principle trial for Blood Tumor Mutational Burden (bTMB) Selection of Atezolizumab Immunotherapy

Kim ES, et al. ESMO 2018. Abstract LBA55.

30

11/5/19 16

BFAST (Blood First Assay Screening Trial): Phase II/III in Advanced Treatment-naïve NSCLC

Key Endpoints Cohort C:

Primary = PFS by Investigators (hierarchical testing, bTMB≥16 first, then bTMB ≥10) Secondary = PFS by IRF, OS, ORR, PRO

31

MYSTIC: 1L durvalumab ± tremelimumab vs chemotherapy in metastatic NSCLC – bTMB

Rizvi NA, et al. ASCO 2019. Abstract 9016.

ctDNA by Guardant OMNI

32

11/5/19 17

What can Liquid Biopsy provide in November 2019 for NSCLC? Tumor Genomics & blood-based Tumor Mutational Burden (investigational) Advantages of plasma ctDNA over Tumor biopsy or re-biopsy:

• Indicated when tumor tissue not available or biopsy is high risk (or “plasma-first” situations)
• Reflects shed tumor DNA into plasma from all tumor sites, providing a “global perspective”
• May abrogate the issue of tissue heterogeneity and under-genotyping due to small sample
• Can determine mechanism of resistance without biopsy, to guide subsequent therapy
• Can be repeated serially (longitudinal assessment) for response & early progressive disease
• Relatively non-invasive & high acceptance rate by patients
• Detection of Minimal Residual Disease (i.e. after surgical resection)

33 Prediction of Relapse after Surgery for Early Stage NSCLC by plasma ctDNA

Swanton et al: Nature 2017

34