Challenges Solutions for Genomics in Clinical Medicine Girish

  • Slides: 12
Download presentation
Challenges & Solutions for Genomics in Clinical Medicine Girish Putcha, M. D. , Ph.

Challenges & Solutions for Genomics in Clinical Medicine Girish Putcha, M. D. , Ph. D. Director of Laboratory Science Mol. DX Palmetto GBA

Disclaimer • Any opinions expressed are my own and not necessarily those of Mol.

Disclaimer • Any opinions expressed are my own and not necessarily those of Mol. DX, Palmetto, or CMS.

Question Since reimbursement, even more so that regulatory, challenges are frequently cited as the

Question Since reimbursement, even more so that regulatory, challenges are frequently cited as the single most significant impediment to personalized/precision medicine, what can payers do to ensure the equitable availability of appropriately priced, high quality genomic tests that improve health and well-being?

The Vicious Cycle for Genomics In Clinical Medicine Regulatory “parallel universes” Weak and/or inappropriate

The Vicious Cycle for Genomics In Clinical Medicine Regulatory “parallel universes” Weak and/or inappropriate clinical use Poor coverage and reimbursement Poor evidence development

Reimbursement Challenges: Regulatory • Almost all commercially available genomic tests are laboratory-developed tests (LDTs).

Reimbursement Challenges: Regulatory • Almost all commercially available genomic tests are laboratory-developed tests (LDTs). • • Of 5921 tests registered with Mol. DX from inception to 9/25/2015, >93% (5512) are LDTs. Many effectively function as multiplexed companion diagnostics with similar, if not identical, intended use(s) and indication(s) for use as FDA-cleared or approved tests Limited to no oversight of marketing claims Limited to no transparency for providers, patients, or payers into their “quality” Evidentiary standards for commercialization are almost entirely self-determined with limited, generally non-transparent independent 3 rd party review (e. g. , CAP, NYSDOH, health technology assessment (HTA) groups, payers) Is this situation in the best interest of patients? Of innovation? Of healthcare in the US? Why is this appropriate for diagnostics but not for medical devices, drugs, etc?

Reimbursement Challenges: Coding, Coverage and Payment • Coding • • • Coverage • •

Reimbursement Challenges: Coding, Coverage and Payment • Coding • • • Coverage • • • CPT codes for genomic tests can be over-simplified, failing to recognize clinically important differences among these tests, resulting in their commoditization and undervaluation Obscure unambiguous identification of a test/service by allowing code stacking v 2. 0 (i. e. , by analyte vs methodology) Idiosyncratically variable despite presumably asking similar if not identical questions based on the same evidence Appropriately focused on clinical utility while inappropriately ignoring analytical and clinical validity (generally deferring instead to “regulatory approval”) Payment • • • Unlike some other areas of healthcare, pricing continues to be based on cost instead of market, let alone on value “Market-based” pricing imminent under PAMA though gaps will remain (e. g. , 81479) Core underlying assumption in PAMA is that commercial payers are “better” (i. e. , cheaper) at pricing tests than CMS

Solutions: Coding • Develop “better” codes that recognize clinically important differences among tests that

Solutions: Coding • Develop “better” codes that recognize clinically important differences among tests that impact their cost and value (because you cannot pay for “value” if you cannot identify it) • • PAMA: New advanced diagnostic laboratory test (ADLT) codes PAMA: Codes that distinguish FDA-cleared/approved tests from LDTs AMA: Proprietary laboratory assay (PLA) codes If necessary, payers can further clarify appropriate use of codes (e. g. , Mol. DX billing and coding guidelines for NGS-based somatic and germline panels–M 00127 and M 00130): • • • Do not stack Tier 1 or 2 codes for panels 81445, 81450 and 81455 apply only to tumor tissue-only panels (i. e. , not “liquid biopsies” or matched tumor-normal panels) reporting only somatic variants (i. e. , not germline), specifically SNVs and small indels (defined as ≤ 20 bp), not CNVs and rearrangements (i. e. , “hotspot” panels) If no codes accurately describe the test, code as 81479 and submit a dossier.

Solutions: Coverage • Transparently define and consistently apply criteria for analytical validity (AV), clinical

Solutions: Coverage • Transparently define and consistently apply criteria for analytical validity (AV), clinical validity (CV), and clinical utility (CU) that still allow adequate flexibility to consider nuances related to clinical context, evolving standards of care, patient and/or provider-specific factors, etc • Require and specify appropriate comparators: “Best practice” or “real world” • Apply CU metrics (e. g. , improvement in “net health outcomes”) transparently and consistently for all healthcare interventions (i. e. , diagnostics, therapeutics, physician services, hospital services, etc) so value -based trade-offs designed to optimize both the cost and effectiveness of care possible • Consider risk-sharing approaches (e. g. , coverage with evidence development) • Transparently define and consistently apply criteria to qualify for such approaches

Solutions: Coverage • Provide a brief, transparent rationale for coverage decisions, especially for any

Solutions: Coverage • Provide a brief, transparent rationale for coverage decisions, especially for any “deviations” from “standard” requirements • Provide a simple, robust and transparent appeals/reconsideration process • Develop a standardized application for coverage that includes assessment of AV, CV, and CU to facilitate efficiency and consistency in assessments, and to diminish administrative burdens on applicants and evaluators • Develop a realistic pre-submission process to enable test developers and payers to quickly and efficiently discuss the adequacy of a dossier in a “safe harbor” prior to full submission

Solutions: Reimbursement • Price all healthcare interventions (i. e. , diagnostics, therapeutics, physician services,

Solutions: Reimbursement • Price all healthcare interventions (i. e. , diagnostics, therapeutics, physician services, hospital services, etc) transparently using similar value-based (or at least market-based) metrics • Stop “silo-ing” money for different interventions so value-based trade-offs designed to optimize both the cost and effectiveness of care possible • If “silo-ing” is unavoidable, at least ensure that pricing within silos is internally consistent, and ideally reflects “value” or “market”, not simply “cost” • Consider risk-sharing approaches (e. g. , “reimbursement with evidence development”)

Non-invasive Prenatal Screens and Liquid Biopsies: Lessons Learned? • • • Tests exclusively LDTs

Non-invasive Prenatal Screens and Liquid Biopsies: Lessons Learned? • • • Tests exclusively LDTs (until recent FDA approval of 1 st liquid biopsy test – Roche’s EGFRv 2) Not validated in intended use patient population before commercialization Clinical studies have some (potentially significant) biases in design and suboptimal “real world” (not “best practice”) comparators • Impressive if mostly unregulated and unverified claims (“>99% accurate”) that can obscure. . . • Performance variations with alteration types • Clinical importance of predictive values versus “sensitivity” and “specificity” • Appropriate indications for use (e. g. , “diagnosis” vs “screening”) • Aggressive lobbying of clinical guideline bodies, leading to exuberant early recommendations followed by greater restraint • Health economic value propositions based on modeling with (accurate? ) assumptions “Good” tests. . . when limitations fully and transparently communicated and appropriately used

Thank you. gputchapmdx@gmail. com

Thank you. gputchapmdx@gmail. com