Guidelines for Selection of Lung Cancer Patients Revision

Transcription

1 M1661 CAP-IASLC-AMP Molecular Testing Guidelines for Selection of Lung Cancer Patients Revision Neal I. Lindeman, MD, FCAP

2 Neal Lindeman, MD, FCAP on behalf of CAP, AMP, IASLC Director, Molecular Diagnostics Brigham and Women s Hospital Associate Professor, Pathology CAP18 Annual Meeting Chicago, Illinois October 2018

3 Objectives Explain the recommendations about the new actionable biomarkers for lung cancer Identify the impact of the new recommendations on patient care and appropriate methods of testing (preanalytic, analytic, and postanalytic actions) Identify when and how to test for EGFR and ALK resistance based on new observations and evidence since the first guideline Apply new recommendations about developments in biomarker testing with an emphasis on immunohistochemistry screening for ALK translocations 3

4 2013 Guideline: Major points Who to test: Advanced stage lung cancers with an adenocarcinoma component o Regardless of age, gender, smoking history, ethnicity Early stage testing is an institutional policy decision What to test: UNMODIFIED Sanger PNA-enriched Sanger EGFR and ALK Optimal sample according to quantity and quality of cancer DNA EGFR wild type Rx: platinum doublet 1-yr survival: 5% EGFR exon 21 mutation Rx: erlotinib 1-yr survival: 30% Cytology samples (cell block preferred) or tissue samples Fresh, frozen, or fixed (avoid acids and heavy metals) Primary or metastasis Each of multiple primaries if histologically distinct EGFR KRAS ALK 4

5 2013 Guideline: Major points, cont d Mol Dx IHC FISH How to test: EGFR by Molecular Diagnostics, ALK by FISH Platform selected by performance characteristics, not technology Sensitivity > 50% malignant cell content Must make available more sensitive methods (>10%) All testing completed within 10 working days I Risk Ratio M-H, Random, 95% CI Mutant No Mutant/ Wild typ 5

6 Why revise? New biomarkers with targeted therapies ROS1, MET, ERBB2, BRAF, RET, KRAS, PIK3CA Markers of resistance to initial therapy Advances in technology IHC Next generation sequencing Circulating cancer cells/dna Re-consideration of squamous and small cell cancers Akiba, Onc Letters, 2014 Existing guidelines must be updated periodically National Guideline Clearinghouse: 5 years without an update is considered a guideline that is no longer current 6

7 Project Team Co-chairs Philip T. Cagle, MD CAP Yasushi Yatabe, MD, PhD IASLC Neal Lindeman, MD AMP Expert Panel Members CAP Keith Kerr, MD Mary Beth Beasley, MD Sanja Dacic, MD, PhD Eric Bernicker, MD IASLC Erik Thunnissen, MD, PhD Marc Ladanyi, MD Ming S. Tsao, MD Benjamin Solomon, MBBS, PhD AMP Dara L. Aisner, MD, PhD Lynette Sholl, MD Maria E. Arcila, MD David J. Kwiatkowski, MD, PhD Advisory Panel Members David Yankelevitz, MD Mark G. Kris, MD Anthony John Iafrate, MD, PhD Tony Mok, MD Charles Powell, MD Dhananjay Chitale, MD Timothy Allen, MD, JD Pasi Janne, MD, PhD Lukas Bubendorf, MD Juan-Sebastian Saldivar, MD Antonio Marchetti, MD Ed Cibas, MD Paul Bunn, MD Suresh Ramalingam, MD Natasha Rekhtman, MD, PhD Federico Cappuzzo, MD, PhD Tetsuya Mitsudomi, MD Yi-long Wu, MD Bonnie Addario Kim Norris Julia Bridge, MD Marina Nikiforova, MD Marileila Varella Garcia, PhD Steering Committee Jan Nowak, MD CAP Fred Hirsch, MD, PhD IASLC Neal Lindeman, MD AMP Staff Christina Ventura, MPH, MLS (ASCP) - CAP Lisa Fatheree, SCT(ASCP) CAP Carol Colasacco, MLIS, SCT(ASCP) - CAP Murry Wynes, PhD - IASLC Pia Hirsch - IASLC Robyn Temple-Smolkin, PhD AMP Mrudula Pullambhatla, MS - AMP Lesley Souter, PhD- Methodologist 7

8 Bethesda, February,

9 Enough about process: what are the recommendations? 9

10 Did you get all that? 10

11 All information presented here is now published! 11

12 A few salient points Changes to 2013 recommendations: Syntax (aka Just shoot me ) Strength augmentation (should versus must, per evidence) Equality for all cytology specimens (cell block was preferred) TAT to recut slides is 3 days (was 24 hrs for in-house) Sensitivity LOD is 20% cancer cells (was 50%) 12

13 A few salient points New recommendations (big picture): New genes: ROS1 o late-breaking reconsideration for BRAF IHC is acceptable for ALK and ROS1, not EGFR Acquired resistance: T790M, down to 5% mutant alleles No new genes for squamous or small cell cancer NGS acceptable alternative to multiple single tests PD1/PDL1 testing important but needs its own guideline 13

14 KQ I. What other genes should be tested in lung adenocarcinoma? Frequency and alteration type: ROS1: 1-2% rearrangement RET: 1-2% rearrangement BRAF: 4% mutation half are non-v600e MET: 3% exon 14 skipping mutations, amplification ERBB2/ HER2: 2% mutation KRAS: 30% mutation 14

15 New genes: ROS1 Clinical utility ROS1+ tumors respond to crizotinib RR 70-80% Phase I, I/II, II trials No Phase III Oncologists treat with crizotinib Oncologists use ROS1 testing Crizotinib approved by FDA ALK+, ROS1+ lung cancers 15

16 New genes: ROS1, cont d Methodology: Adenocarcinomas, but no sensitive clinical predictors No designated companion diagnostic FISH is predicate method Immunohistochemistry (IHC) for screening; confirm + with FISH RNA methods (RT-PCR, anchored multiplexed PCR) Next generation DNA sequencing Cha, et al., PLoS One,

17 Others: RET, BRAF, ERBB2, KRAS, MET As with EGFR, ALK, ROS1, typically mutually exclusive Adenocarcinomas, but no sensitive clinical predictors Clinical trials and/or potential drugs for each: RET: cabozantinib, vandetinib BRAF: vemurafenib, dabrafenib, +/- trametinib ERBB2/HER2:?pulsed afatinib?,?neratinib?,?dacomitinib? KRAS: trametinib, selumetinib MET: crizotinib Not recommended as single tests for lung cancer patients If a large panel is being performed, include these If ALK, EGFR, ROS1 all negative, include these Limited evidence for clinical utility (Case reports & small series*) 17

18 KRAS Mutations in NSCLC Clinical utility No effective direct KRAS inhibitors Recent reports of response to MEK inhibitors Associated with smoking Useful to rule out other less common alterations 30% of lung adenocarcinomas Simple and widely available single gene assays o Standard-of-care in colon cancer o Same mutations as in lung cancer 18

19 BRAF mutations Clinical utility Recent (after our lit search) single arm phase II data 33% response for V600E to single agent dabrafenib 63% response for V600E to combo dabrafenib + trametinib Methodology Widely available single gene tests for melanoma only test V600E Half of BRAF+ lung cancers have non-v600e mutations Need to sequence additional hotspots/whole gene? Planchard, et al., Lancet Onc,

20 ERBB2 (HER2) Critical gene in breast cancer Amplification (FISH) or overexpression (IHC) Responds to treatment with trastuzumab Different role in lung cancer Mutations, typically insertions in exon 20 FISH, IHC are not useful for lung cancer Do not respond to therapeutic antibodies Treatment with TKIs has been disappointing so far Kris, et al., Ann Oncol, 2015 Besse, et al., Ann Oncol, 2014 Suzawa, et al., Cancer Sci,

21 RET Characteristic mutations in thyroid cancer Mutations in MEN II syndrome, medullary carcinoma RET/PTC fusions in papillary carcinoma Different alterations in lung cancer Multiple fusions, including KIF5B, CCD6, NCOA4 Some of the same partners as ROS1 Response to broad-spectrum TKIs: Case reports, case series preliminary phase II trial Methodology: Discovered by DNA NGS Unlike other fusions, FISH and IHC are challenging Platt, et al., BMC Cancer,

22 MET Clinical utility - a complicated story MET copy gain first seen in who relapsed on anti-egfr therapy Confusion: copy gain vs amplification Can co-exist with other oncogene mutations MET inhibitors were developed (crizotinib), but didn t work in general Rare true MET amplification does respond to crizotinib (~50% rr) Fast forward: NGS discovers mutations affecting splicing of exon 14 Common: ~3% of lung cancers Early results: exon 14 skipping mutations do respond to crizotinib Jorge, et al., Lung Cancer,

23 MET, cont d Methodology: work to be done! Adenocarcinoma and squamous carcinoma; no clinical predictors FISH: copy gain, true amplification, not clear if correlates with IHC IHC: copy gain, true amplification, not clear if correlates with FISH NGS: splicing mutations, if introns tested RNA sequencing: splicing mutations, but not widely available 23

24 Still other genes Insufficient evidence for systematic review: NTRK1,2,3 NRG1 FGFR1,2,3,4 ERBB4 24

25 KQV: What is the role of sequencing panels in lung cancer? 2013: insufficient evidence to support NGS panels 2018: NGS panels preferred over single gene tests Single gene methods still acceptable, provided TAT met Results returned quicker Spares sample, which is often limiting Enables expanded testing beyond EGFR, ALK, ROS1 Help patients find appropriate clinical trials TAT recommendation: two weeks, for all testing 25

26 NGS panel considerations Methodology Not prescriptive regarding platform, informatics, library Detailed discussion of NGS methods out of scope Performance characteristics are determinants Adequate sensitivity: 10% allele fraction Validate, control, and maintain as any CLIA assay Validate representatives of each alteration type 26

27 KQ II. Is immunohistochemistry reliable for ALK translocations? 27

28 KQ II. Is immunohistochemistry reliable for ALK translocations? YES 28

29 How to test for ALK? FISH is not perfect Neither is IHC Concordance: Either method can be performed We are not recommending both 29

30 KQ II. Is immunohistochemistry reliable for ALK translocations? Numerous studies have shown excellent concordance Discordances seen in both directions both FISH and IHC can be either false negative or positive No scientific need to perform both methods (for ROS1, however, pos IHC should be confirmed with FISH) Do NOT use the ALK1 antibody developed for Anaplastic Lymphoma not sufficiently sensitive 30

31 FDA Approval VENTANA ALK (D5F3) CDx Assay is intended for the qualitative detection of the ALK protein in FFPE NSCLC tissue stained with a BenchMark XT automated staining instrument It is indicated as an aid in identifying patients eligible for treatment with XALKORI (crizotinib) 31

32 KQ III. What testing is appropriate for patients with acquired resistance? Cortot and Janne, Eur Resp Rev 2014; 23: Cortot and Janne, Eur Resp Rev 2014; 23:

33 Testing in acquired resistance setting Methodologic considerations for EGFR T790M Relapsed specimens are heterogeneous Secondary mutations: lower frequency than drivers Relapse tests need to be very sensitive o Recommended cutoff is 5% mutant alleles Circulating cell-free DNA may be superior* o Less dependent on sampling 33

34 KQ III. What testing is appropriate for patients with acquired resistance? EGFR inhibitor relapse Recommendation: test T790M mutation o T790M-specific therapies available (osimertinib) Multiple other mechanisms of resistance o Insufficient evidence for/against testing ALK inhibitor relapse I ll keep you in suspense 34

35 Second line agents can work for ALK resistance 35

36 KQ III. What testing is appropriate for patients with acquired resistance? EGFR inhibitor relapse Recommendation: test T790M (~50% of relapse patients) T790M-specific therapies available (osimertinib) Multiple other mechanisms of resistance Insufficient evidence for/against testing ALK inhibitor relapse Several point mutations in ALK Second generation ALK inhibitors under study Matrix of resistance mutation best drug is evolving Insufficient evidence for/against testing 36

37 KQ IV. Test squamous or small cell carcinomas? Small cell carcinomas: No Squamous cell carcinomas: Maybe Squamous carcinoma genes : FGFRs, DDR2 Insufficient evidence to support for/against testing Adenocarcinoma genes : EGFR, ALK, ROS1 If clinical or pathologic features are high risk Can t exclude unsampled adenocarcinoma histology Young patient No history of tobacco use Other therapies: EGFR antibodies, immunotherapy 1 cycle gemcitabine/cisplatin Switch to crizotinib Images courtesy of Eric Bernicker, MD 37

38 EGFR inhibiting monoclonal antibodies for squamous cell carcinoma Cetuximab, panitumumab, necitumumab Standard of care in colon cancer Different mechanism from kinase inhibitors Inhibit ligand binding More associated with expression/amplification Benefit less dramatic than TKIs in adenocarcinoma EGFR copy number (FISH) as potential biomarker 38

39 KQ IV. Test squamous or small cell carcinomas? Small cell carcinomas: No Squamous cell carcinomas: Maybe Squamous carcinoma genes : FGFRs, DDR2 Insufficient evidence to support for/against testing Adenocarcinoma genes : EGFR, ALK, ROS1 If clinical or pathologic features are high risk Can t exclude unsampled adenocarcinoma histology Young patient No history of tobacco use Other therapies: EGFR antibodies, immunotherapy 39

40 KQ V. What is the role of testing cell-free DNA or circulating tumor cells? Clinical Utility: evolving Lots of buzz about liquid biopsy Overcome sampling/heterogeneity issues Easier for patients Quicker Conceptually appealing, but published data is still lean We *do* expect this to change quickly Early adopters are doing this already CTC testing not ready Three contexts for plasma cfdna testing Initial diagnosis Monitoring on therapy Acquired resistance 40

41 cfdna testing Initial diagnosis: Appropriate when tissue testing unavailable No adequate sample Patient cannot undergo biopsy Monitoring on therapy: Very exciting and very unproven Cannot recommend at this time Acquired resistance: Appropriate alternative to tissue testing Sensitivity poor, specificity high Treat if plasma positive Biopsy and test tissue if plasma negative 41

42 cfdna testing Methodology Sample processing is critical Streck tubes or EDTA tubes, processed in hours Different centrifugation protocols (no brakes) NGS methods unproven PCR methods appropriate Droplet digital PCR Real-time PCR No recommendation regarding reporting 42

43 KQ VI: What is the role of sequencing panels in lung cancer? 2013: insufficient evidence to support NGS panels 2018: NGS panels preferred over single gene tests Single gene methods still acceptable, provided TAT met Results returned quicker Spares sample, which is often limiting Enables expanded testing beyond EGFR, ALK, ROS1 Help patients find appropriate clinical trials TAT recommendation: two weeks, for all testing 43

44 NGS panel considerations Methodology Not prescriptive: platform, informatics, library Detailed discussion of NGS methods out of scope Performance characteristics are determinants Adequate sensitivity: 10% allele fraction Validate, control, and maintain as any CLIA assay Validate representatives of each alteration type 44

45 Not a KQ, but should have been: What is the role of PD-1/PD-L1 IHC? Pulling up lame: Out of scope for us Wrong panel constituency to assess this properly Not included in initial search or data evaluation tools Incorporating it would entail a near-restart and a 1+ yr delay Another project is addressing this in a broader disease context Non-evidence based opinion Immune checkpoint therapies are proven effective in lung cancer Test methods, in a global sense, are not yet established o IHC: PD-1, PD-L1, CTLA-4, with multiple different antibodies o Tumor lymphocytes, mutational load, neoantigen expression Some agents require specific tests to determine eligibility Opinion: for now, use the tests required for the agents being considered 45

46 Conclusions 2013 recommendations largely unchanged Smears are acceptable, sensitivity reduced to 20% tumor content Add ROS1 for all patients Add BRAF, ERBB2, MET, RET if doing a large panel IHC is acceptable for ALK and screening for ROS1 EGFR T790M methods for resistance need high sensitivity Other resistance mutations require further study No mandate to test squamous or small cell carcinomas Cell free DNA is appropriate when biopsy hard to obtain If cell free DNA is negative, biopsy is needed NGS panels are preferred over multiple single assays PD1/PDL1 are important, but in a different guideline 46

47 Questions? 47

48 48

49 Thanks for attending! Complete the online course evaluation! Be an active participant in shaping the CAP s Annual Meeting by sharing your feedback. 49

50