1. MIR Medicare Coverage with Evidence Development for PET: NOPR and Beyond Barry A. Siegel, M.D. Mallinckrodt Institute of Radiology

2. Barry A. Siegel, M.D. Disclosures Advisory Board –GE Healthcare Consulting –ImaginAb –Blue Earth Diagnostics Lecture Honoraria –Siemens

3. MIR Dissemination of PET into clinical practice has been very slow! Barriers Expensive technology Slow acceptance by clinicians Reliable supply of radiopharmaceuticals Regulation of radiopharmaceutical production Variable and restrictive coverage policies by government and private payers (reflecting lack of definitive evidence of utility)

4. MIR PET Reimbursement in the USA Dependent on FDA approval of PET drugs –Unique approach, aided by legislation Reimbursable clinical indications –Determined by technology assessment panels of third-party payers –Process dominated by Centers for Medicare & Medicaid Services (CMS)

5. MIR Medicare Coverage of PET Standard for reimbursement is “reasonable and necessary” In 1990s, CMS adopted a new evidence-based approach for making coverage determinations –Requires peer-reviewed scientific evidence to document that new technology leads to changes in patient management and to improved health outcomes for Medicare beneficiaries Prompted in part by poor quality of evidence used to support MRI coverage

6. The MRI Backlash PET Became the “Whipping Boy” of High Technology Medicine PET Payers MRI http://www.sportsofboston.com/wordpress/wp-content/uploads/2009/07

7. MIR Challenges with Diagnostics Determining “value” is a barrier for all diagnostics –Traditional blood assays, genetic profiling, or imaging Testing is a single node in a chain of diagnostic and therapeutic interventions Can one attribute improvements in health outcomes directly back to any single event in the chain, let alone a diagnostic imaging test? Usefulness of a diagnostic is constrained by the (non)availability of good therapies

8. MIR Medicare Coverage of Oncologic PET CMS elected not to consider oncologic indications for PET broadly Rather evaluated the evidence on a cancer- specific and indication-specific basis Problematic because the specific evidence typically has not been very robust “Catch 22”

9. MIR Does PET Improve Health Outcomes in Patients with Cancer? This has been very difficult to demonstrate Vast majority of PET clinical trials –Single-institution –Pilot  phase II –Small patient numbers (<50) A major reason for unfavorable technology assessments of PET (and for limited coverage)

10. MIR Does PET Improve Health Outcomes in Patients with Cancer? Evidence accruing in the last few years Randomized controlled trials –All done in countries with highly restricted PET coverage Practice-based evidence (e.g., registries) Change in management largely used as a “surrogate” for improved outcome (especially avoidance of futile therapies)

11. RCTs: FDG-PET in Oncology Cancer (Indication) No. RCTs Results NSCLC (preoperative staging) 5Mixed, but favor reduction in futile thoracotomy Colorectal cancer (liver metastasis resection) 2Conflicting results with respect to reduction in futile surgery Colorectal cancer (recurrence detection) 1Earlier detection and increased likelihood of complete resection of recurrence Cervical cancer (Treatment of extrapelvic disease guided by PET) 1No improvement in OS or DFS ITT Analyses: no improvements in survival. But should a diagnostic test be expected to improve survival? No practical way to fund RCTs for each tumor/indication!

12. MIR Medicare Coverage of Oncologic PET 1998Evaluation of solitary pulmonary nodules and initial staging of NSCLC 1999Suspected recurrent colorectal cancer, lymphoma, 2001Further coverage for 6 prevalent cancers 2002Individual requests submitted for several other cancers 2004 Unwieldy Approach Proposed mechanism for expanded coverage

13. MIR Coverage with Evidence Development (CED) An option for coverage of promising drugs, biologics, devices, diagnostics, and procedures that would not otherwise meet Medicare’s evidentiary standards for being “reasonable and necessary” CED links coverage to requirement that patients participate in a registry or clinical trial Goal of longitudinal data collection to improve understanding of the new technology First applied to biologic therapies for colon cancer, implantable cardiac defibrillators, and oncologic PET

14. National Oncologic PET Registry: A Nationwide Collaborative Program Sponsored by Managed by Advisor Endorsed by Chair, Bruce Hillner, MD, Virginia Commonwealth University Co-chair, Barry A. Siegel, MD, Washington University Co-chair, R. Edward Coleman, MD, Duke University Co-chair, Anthony Shields, MD, PhD Wayne State University Statisticians: Dawei Liu, PhD, Fenghai Duan, PhD, Ilana Gareen, PhD,, Lucy Hanna, MS, Brown University

15. Objective Assess the effect of PET on referring physicians’ plans of intended patient management –across a wide spectrum of cancer indications for PET not currently covered by Medicare Hypothesis PET will lead to change of patient management as often for non-covered as reported for covered cancers

16. Goals Provide access to the service (PET) Minimize the burden to patients, PET facilities, and referring physicians Generate evidence of reasonable quality to help CMS decide whether to expand coverage of PET Registry to be financially self-supporting

17. Referring MD requests PET Referring MD requests PET Pre-PET Form Pre-PET Form PET done PET done PET interpreted & reported PET interpreted & reported Post-PET Form sent, including question for referring MD consent Post-PET Form sent, including question for referring MD consent Post-PET Form completed. Claim submitted Post-PET Form completed. Claim submitted Ongoing patient management NOPR Workflow Ask patient for consent Ask patient for consent

18. Pre-PET Form: Intended Patient Management  Observation (with close follow-up)  Additional imaging (CT, MRI) or other non-invasive diagnostic tests  Tissue biopsy (surgical, percutaneous, or endoscopic).  Treatment (if treatment is selected, then also complete the following) Treatment Goal: (check one)  Curative  Palliative Type(s): (check all that apply) –  Surgical  Chemotherapy (including biologic modifiers) –  Radiation  Other  Supportive care If PET were not available, your current management strategy would be (select one)? Intended management, given PET findings, asked on post-PET form

19. Key NOPR Results (Before 2009 NCD) Overall Impact on Patient Management –Diagnosis, Staging, Restaging, Recurrence –Data on 22,975 scans from May 8, 2006 – May 7, 2007 –J Clin Oncol 2008; 26:2155-61 Impact on Patient Management by Cancer Type –Confirmed Cancers –Staging, Restaging, Recurrence –Data on 40,863 scans from May 8, 2006 – May 7, 2008 –J Nucl Med 2008; 49:1928-35 Treatment Monitoring –Data on 10,447 scans from May 8, 2006 – Dec 31, 2007 –Cancer 2009:115:410-18

20. Cohort Profile First year of NOPR (5/8/06 to 5/7/07) 22,975 “consented” cases from 1,519 facilities Technology profile –84% PET/CT –71% non-hospital –76% fixed sites Hillner et al., J Clin Oncol 2008

21. PET Changed Intended Management in 36.5% of Cases Non-TreatTreat23.231.628.629.228.3 TreatNon-Treat7.9 7.59.78.2 Patients with change post-PET (%) 31.139.536.139.0 36.5 Hillner et al., J Clin Oncol 2008; 26:2155, Hillner et al., J Nucl Med 2008; 49:1928 Clinical Indication for PET Study (Percent) Pre-Pet Plan Post-PET Plan Dx n=5,616 Staging n=6,464 Restaging n=5,607 Recurrence n=5,388 All n=22,975 TreatSame16.046.515.820.425.5 Non-TreatSame52.914.048.040.737.9 Essentially uniform across different cancer types

22. Impact of FDG-PET for Treatment Monitoring Chemotherapy 82%, chemoRT 12%, RT 6% Ovarian, pancreas, NSCLC, SCLC most frequent Metastatic disease in 54% PET findings led to: –Switch to another therapy in 26% –Adjust dose or duration of therapy in 17% –Switch from therapy to observation/supportive care in 6% Management change more often if post-PET prognosis worse rather than improved/unchanged (70% vs. 40%) Hillner et al., Cancer 2009; 115:410

23. 2006National Oncologic PET Registry (NOPR) begins data collection 2008Initial NOPR results published and expanded coverage requested 2009National Coverage Determination (NCD) PET for initial treatment strategy (one scan) covered for nearly all cancers and some new cancers covered for subsequent treatment strategy Medicare CED and Oncologic PET Over 90% of US PET facilities participated Complete data for nearly 133,000 scans

24. NOPR-2009 Data collection continued for subsequent treatment strategy of remaining cancers (with minor CRF modifications) 155,540 scans with complete data submission Primary analysis: comparison of NOPR-2006 and NOPR- 2009 cohorts ( J Nucl Med 2012; 53:831-7) –Restaging, recurrence or treatment monitoring known cancers –Data on 41,145 scans (2006) and 70,358 scans (2009) –“Results strongly suggest it is unlikely that new useful information will be obtained by extending the coverage of certain cancer types and indications only under CED.”

25. 2013National Coverage Determination (NCD) –Ended FDG-PET data collection –Further expanded coverage for subsequent treatment strategy to all cancers (but with a 3-scan limit) –3-scan limit intended to curtail use of PET for surveillance (statutorily non-covered) Medicare CED and Oncologic PET

26. NaF-PET vs. Bone Scintigraphy Higher bone uptake and better pharmacokinetics Comparable radiation dose Superior image quality with improved sensitivity Higher specificity (with PET/CT), since many benign processes have characteristic CT appearances Both allow total skeletal survey NaF-PET not vulnerable to Mo-99 supply

27. Medicare CED and Oncologic PET 2010 NCD for NaF-PET; CED required 2011NOPR opens NaF-PET registry 2014NOPR submits request for coverage of NaF-PET (5/15/2014)

28. NOPR (NaF-PET): Results Prostate cancer (Hillner et al., J Nucl Med 2014;55:574) ‒ 68% of all patients ‒ 1,024Initial staging ‒ 1,997 Suspected first osseous metastasis ‒ 510 Suspected progression of osseous metastases ‒ Treat vs. non-treat change in intended management in 44% to 52% (imaging-adjusted 12% to 16%) Other cancers (Hillner et al., J Nucl Med 2014; 55:1054) ‒ Similar results (lower impact with suspected first osseous metastasis than for prostate cancer) Treatment monitoring (Hillner et al., J Nucl Med 2015; 56:222) ‒ 40% overall change in management (59% new therapy with evidence of new or progressive disease)

29. NaF-PET: CMS Draft Decision September 16, 2015 Evidence is sufficient to determine that use of NaF- PET to identify bone metastasis is not reasonable and necessary to diagnose or treat an illness or injury... Continue CED for 12 mo from final decision date Confirmatory analyses needed to show that NaF-PET: –Changes management to more appropriate palliative or curative care –Improves QOL or survival

30. NaF-PET: Next Steps NOPR commented on draft decision, strongly disagreeing with CMS Await final decision Prepared to do claims linkage to demonstrate that intended management concordant with actual management Cannot evaluate the outcome measures requested by CMS

31. Strengths of the NOPR Data “Real world” data Timely data Very large patient cohorts Current technology (≥ 85% PET/CT) Good observational studies usually match controlled studies in magnitude and direction of effect (Concato NEJM 2000; Benson NEJM 2000; Ionnanidis JAMA 2001) Results similar to more tightly managed single-institution studies (e.g., Hillner 2004) and to Australian studies with outcome validation

32. NOPR Limitations Data “quality” Potential that physicians may have been influenced by the knowledge that future Medicare reimbursement might be influenced by their responses No control group –A fundamental problem with observational studies –Neither historical nor contemporaneous controls adequate Collected change in “intended” management, not actual management –Partially addressed by linking NOPR plans to claims data

33. NOPR Limitations Unknown if management changes were in the correct direction or improve long-term outcomes –Using management change as surrogate requires prior data on test accuracy and value of therapies Defining the relevant long-term outcomes for a diagnostic (instead of therapeutic) procedure is controversial NOPR does not address: –Whether PET should be used in lieu of or as a complement to other imaging techniques –The optimal sequencing of CT, MRI and PET. –How much ‘better’ PET is than next best met hod

34. Major Problem with the NOPR Paradigm Tradeoff between data quantity/quality and access Consequence of the self-funded model with non- engaged participants (You get what you pay for!) Possible solutions –Funding of participating sites/referring MDs More detailed clinical data Information about actual management/outcomes –Better data quality will require engaged/educated participants

35. CMS PET Registries What’s Next?

36. A Study to Evaluate the Clinical Utility of Amyloid PET in U.S. Medicare Beneficiaries Study Chair: Gil D. Rabinovici Co-chairs: Maria C. Carrillo, Constantine A. Gatsonis, Bruce E. Hillner, Barry A. Siegel, Rachel A. Whitmer

37. IDEAS-Study.org PET Amyloid Imaging Three agents approved by FDA as imaging biomarkers of amyloid deposits –April, 2012 18 F-florbetapir –October 2013 18 F-flutemetamol –March, 2014 18 F-florbetaben

38. IDEAS-Study.org Insufficient evidence of clinical utility to justify coverage of Aβ PET Reimbursement would be considered under coverage with evidence development (CED) in clinical studies designed to: Develop better treatments or prevention strategies for AD Identify subpopulations at risk for developing AD Resolve clinically difficult differential diagnoses (e.g., frontotemporal dementia versus AD) Must demonstrate Aβ PET improves health outcomes (short-term outcomes related to changes in management as well as longer-term dementia outcomes)

39. IDEAS-Study.org After a Two-Year Gestation: IDEAS An open-label, longitudinal cohort study under CED to assess the impact of amyloid PET on patient-oriented outcomes in individuals meeting Appropriate Use Criteria for amyloid PET (Johnson, et al. 2013) The primary hypothesis is that, in diagnostically uncertain cases, knowledge of amyloid PET status will lead to significant changes in patient management, and this will translate into improved medical outcomes

40. IDEAS-Study.org Inclusion Criteria – AUC+ Medicare beneficiary, age 65 or older, referred by dementia expert Board-certified in neurology, psychiatry, or geriatric medicine, devotes ≥25% of patient contact time to cognitive disorders New diagnosis of cognitive impairment (<2 years) MCI Dementia Etiology of cognitive impairment is uncertain after a comprehensive evaluation History & physical exam, mental status testing, labs, structural neuroimaging (CT or MRI) Knowledge of amyloid PET status is expected to alter diagnosis and management

41. IDEAS-Study.org Exclusion Criteria Normal cognition or subjective complaints only, non- medical purpose, other inappropriate uses per AUC Knowledge of amyloid status may cause significant psychological harm or otherwise negatively impact the patient or family Amyloid status known (previous CSF or PET) Previous participation in an anti-amyloid trial Cancer requiring active therapy (excluding non-melanoma skin cancer) Hip/pelvic fracture within 12 months Life expectancy less than 24 months Residence in a skilled nursing facility

42. IDEAS-Study.org To assess the impact of amyloid PET on patient management at 90 days Referring dementia expert completes CRFs: Pre-PET: intended management assuming no access to amyloid PET Post-PET: actual management at 90 days post-scan, incorporating amyloid PET results Also leading diagnosis, differential diagnosis, diagnostic confidence pre- and post-PET IDEAS: Aim 1

43. IDEAS-Study.org Test whether amyloid PET imaging will lead to a ≥ 30% change between pre-PET and post-PET patient management within ~90 days in a composite measure of at least one of the following: AD drug therapy; Other drug therapy; and Counseling about safety and future planning The hypothesis will be tested separately for MCI and dementia. Aim 1: Primary Objective

44. IDEAS-Study.org Aim 1: Rationale 31% change in AD drug therapy and 7% change in non AD drug therapy in 229 patients (intended management) 1 35% change in cholinesterase inhibitor or memantine use in 140 patients (retrospective) 2 37% change in patient management (prospective study of 211 patients) 3 Prospective study of 618 patients randomly assigned to immediate or delayed (1 y) disclosure: immediate disclosure led to significantly more changes in management (68%) vs. delayed disclosure (56%) 4 1.Grundmann et al. Alzheimer Dis Assoc Disord. 2013;(1):4-15. 2.Sanchez-Juan et al. Neurology. 2014 Jan 21;82(3):230-8. 3.Zwan et al., presented at 2015 AAIC 4.Pontecorvo et al., presented at 2015 AAIC.

45. IDEAS-Study.org IDEAS: Aim 2 To assess the impact of brain amyloid PET on hospital admissions and emergency room visits in study patients (amyloid PET-known) compared to matched patients not in the study (amyloid PET-naïve) over 12 months CMS claims data will be collected for all study participants and from concurrent controls matched according to a validated algorithm

46. IDEAS-Study.org Determine if amyloid PET is associated with ≥ 10% relative reduction in study patients compared to matched controls: Inpatient hospital admissions over 12 months Emergency room visits over 12 months Aim 2: Primary Objective

47. IDEAS-Study.org AIM 2: Rationale Individuals with dementia at increased risk for hospitalizations & ED visits (Feng 2014) Annual hospitalizations: 26.7% vs. 18.7% Annual ED visits: 34.5% vs. 24.5% Two-thirds deemed preventable CHF exacerbation, bacterial pneumonia, UTI Dementia associated with increased mortality and shorter survival after hospitalizations Preliminary data from Kaiser shows targeted dementia plan led to 18% reduction in ED visits and 11% reduction in hospitalizations (Whitmer, unpublished)

48. IDEAS-Study.org Secondary Objectives Assess impact on resource utilization Assess decrease in unnecessary treatments and procedures Use of AD meds in amyloid-neg patients Repeated, less definitive diagnostic tests Assess impact on clinical trial referrals Compare outcomes MCI vs. dementia Aβ+ vs. Aβ-

49. IDEAS-Study.org Much more complicated study than NOPR ‒ Will collect more detailed information from referring MDs, as well as images (for future analysis) ‒ Patient-centered outcomes (Aim 2) most important to CMS Estimated sample size ‒ Aim 1:11,050 subjects for 30% change in management composite endpoint ‒ Aim 2:18,448 subjects for 10% relative reduction in hospitalization, ER visits Expected study cost $20M (excluding cost of scans) Timeline to coverage: at least 5 years IDEAS Study

50. IDEAS-Study.org IDEAS Operational Model Patients PET Imaging Centers Dementia Specialists IDEAS Project Team IDEAS Project Team

51. IDEAS-Study.org IDEAS Operational Model PET Imaging Centers Dementia Specialists IDEAS Project Team IDEAS Project Team Screen Consent and register Pre-PET Intended management plan Order PET scan

52. IDEAS-Study.org IDEAS Operational Model PET Imaging Centers Dementia Specialists IDEAS Project Team PET Scan PET Completion Form PET Report PET Assessment Form Scans uploaded to ACR image archive Patients

53. IDEAS-Study.org IDEAS Operational Model PET Imaging Centers Dementia Specialists IDEAS Project Team IDEAS Project Team +90 Days Post-PET (ACTUAL management plan) Patients

54. IDEAS-Study.org IDEAS Operational Model PET Imaging Centers Dementia Specialists IDEAS Project Team IDEAS Project Team Data Analysis Impact on management plan 12 month outcomes assessment of longitudinal cohort and CMS matching cohort

55. IDEAS-Study.org What are the Incentives For Dementia Specialists and PET Facilities to Participate? A brain amyloid PET scan will be covered for their patients with cognitive decline in whom the diagnosis is not clear A stipend of $750 will be provided to participating Dementia Specialists for completion of required case report forms for each patient enrolled PET facilities will be reimbursed for the PET scans under CMS under coverage with evidence development (CED) which requires research study participation as a condition of payment

56. IDEAS-Study.org

57. Registration Update (opened Sept 30, 2015) Dementia ExpertsPET Facilities 202 practices pre-registered 65 practices completed registration (218 physicians) 226 facilities pre-registered 120 facilities completed registration (297 physicians)

58. IDEAS-Study@acr.org IDEAS-Study.org IDEAS Steering Committee Inaugural Meeting, June 18, 2015

59. MIR Preferred Road to Coverage of Diagnostic Tests Provide adequate evidence that: Incremental information is obtained with new diagnostic technology compared to alternatives Changes physician recommendations Results in changes in therapy Improves clinically meaningful health outcomes In patients similar to those in whom it will be used Modified from Jacques 2012

60. MIR Conclusions NOPR has successfully used one pathway to help achieve coverage for PET in cancer But pathway quite slow and burdensome Clinical trials of new molecular imaging tracers and methods must focus, from the outset, on obtaining evidence of improved patient outcomes

61. IDEAS-Study.org Supplemental Slides

62. IDEAS-Study.org Sample Size Aim 1. The projected prospectively-recruited sample size is 11,050 30% change in composite endpoint Assumes 60% MCI, 40% dementia β=0.80, α=0.025 (Bonferroni corrected) Planned futility analyses at 1/3 and 2/3 accrual Aim 2. The projected prospectively-recruited sample size is 18,488 (additional 7,438 beyond Aim 1) 10% relative reduction in hospitalizations, ED visits Similar number of controls, identified in CMS database β=0.90, α=0.025 (Bonferroni corrected) Assumes 35% in Medicare Advantage plans excluded from Aim 2 (no claims data)

63. IDEAS-Study.org How Do Patients Enroll in the IDEAS Study? Patients may be enrolled through a Dementia Specialist who is participating in the study. Once enrollment begins (expected Jan 2016), a list of participating Dementia Specialists and Imaging Sites will be posted on the IDEAS Study website.

64. IDEAS-Study.org Patients referred by Dementia Specialists must have access to a study PET Imaging facility. PET Imaging facility must be within 3-4 hours of an amyloid tracer supplier. Dementia Specialists and Imaging Sites will be posted on ideas-study.org upon patient enrollment (January 2016). Clinical Site Locations

65. IDEAS-Study.org Amyloid PET Radiotracer Locations PET Imaging Centers located within 3-4 hours of location may have access to radiotracers. (Updated 10.20.2015.) Beta Amyloid Tracer Availability PhoenixAZChicagoIL HackensackNJ TempeAZRomeovilleIL TotowaNJ Colton (Los Angeles)CAIndianapolisIN AlbanyNY Culver City (Los Angeles)CANew OrleansLA ColumbusOH Palo AltoCABostonMA ClevelandOH SacramentoCAWoburnMA PhiladelphiaPA GilroyCAHaverhillMA GrayTN DenverCOBeltsvilleMD DallasTX East HartfordCTDetroitMI HoustonTX Ft. LauderdaleFLEast LansingMI CharlottesvilleVA JacksonvilleFLMinneapolisMN SterlingVA TampaFLKansas CityMO SeattleWA SanfordFLCharlotteNC MorgantownWV AtlantaGA Raleigh/DurhamNC

66. IDEAS-Study.org Where are amyloid agents available in USA?

67. IDEAS-Study.org Physicians who self-identify as being trained and board certified in neurology, psychiatry, or geriatric medicine Physician also must devote a substantial proportion of patient contact time (≥25%) to the evaluation and care of adults with acquired cognitive impairment or dementia 1,2 Required online training (Alzheimer’s Assn site) Informed Consent (investigator is research subject) Other online training on eCRF will be offered IDEAS: Dementia Specialist Criteria (also known as Referring Physicians) 1.Johnson et al. J Nucl Med 2013;54:476-490 2.Johnson et al. Alzheimers Dement 2013:9:e1-e16

68. IDEAS-Study.org PET Facility Qualifications Experience with brain PET imaging Free-standing PET facilities – accreditation for PET brain imaging by the ACR, Intersocietal Accreditation Commission (IAC), or RadSite PET facilities located in hospitals – Medicare – approved hospital accreditation, such as Joint Commission Interpreting physicians: board-certified and completion of vendor-provided reader training

69. IDEAS-Study.org IDEAS Study Timeline CMS Approval Develop Phase Accrual Phase Last POS final claim CMS Claims data for last POS Completed Analysis Publish/ Request Reconsideration CMS Coverage Decision March 2015 9 months 42+ months 48+ months 54+ months 66+ months 30 months 50+ months

70. IDEAS-Study.org If Yes At Post-PET Scan doctor visit, study participants will be asked if they wish to participate in additional IDEAS related studies Information on all Add-On Studies provided TrialMatch (TM) contact the volunteer to register If interested in 1 or more additional study: Brain Health Registry (BHR) will register volunteers and obtain informed consent for studies of interest CARE- IDEAS DNA Banking Online Cognitive Testing (BHR) GAP TRC-PAD Additional Clinical Trials via TM IDEAS Add-On Studies