Tuesday, April 30, 2019

Webinar May 21, 2019: NGS Testing in Cancer, US and Europe

Boston Healthcare and Bruce Quinn Associates present a webinar on NGS policy for cancer patients on Tuesday, May 21, 2019 at 11 ET.   Both US policy (including the CMS NGS NCD) and European policies and markets will be discussed.

Find the website for registration here.

Very Brief Blog: United Nations Issues Urgent Report on Antibiotic Crisis

On April 29, 2019, the United Nations issued an urgent report on the antibiotic crisis and the emergence of antibiotic resistant-bacteria.

  • United Nations report here.
  • New York Times coverage here.
  • CNN here.  Business Week here.
In addition to the summary report, there are a number of additional supplemental and research reports at the UN page.  The workgroup is "UN Interagency Coordination Group on AMR," or IACG.


There are a few remarks about diagnostics, e.g. " The Organization is also working to improve the accessibility and use of diagnostics and laboratory services to reduce diagnostic uncertainty and inform treatment choices."   They refer to "the inadequacy of the clinical pipeline of new vaccines, medicines and diagnostics..."  Diagnostics "form a core element of the response in helping to select appropriate antimicrobials to treat a disease." Adding:  "Diagnostics are often perceived as an additional expense to the cost of treatment, even though their use may be associated with significant cost savings and efficiency downstream."

Additional Policy Notes

Medicare.  In Medicare policy, just a few weeks ago the President's Counsel on Combatting Antibiotic-Resistant Bacteria held a special meeting resulting in a vote, to urge CMS to finalize mothballed rules requiring US hospitals to have antibiotic stewardship committees.   Here.

Pew and Wellcome.  See links to very recent Op-Eds by groups like the Pew Foundation and Wellcome Trust on the topic, hereFDA.  This topic was also a high priority of recently departed FDA commissioner Scott Gottlieb (here).

NIH and Speech.  In related news, earlier this week, WSJ and SCIENCE reported that NIH had "barred" several scientists from speaking about the sepsis treatment crisis, here and here and here.  One of them, Eichacker, had written two recent articles that were quite critical of CMS sepsis policy ("SEP-1"), here and here.  (Other authors outside NIH had been equally critical, here.)

CMS NTAP & ABx Dx.  I believe for the first time ever, CMS is considering granting a diagnostic test add-on payment in its current inpatient policy rulemaking; see discussion of T2 septicemia test, here.

Monday, April 29, 2019

CMS Reopens NGS Only For: Reconsideration of Evidence of Germline Tests for Targeted Treatments

On April 29, 2019, CMS posted an NCD reopening notice for its March 2018 NCD on uses of NGS in cancer patients.

Find the reopening notice here:

The NCD - as currently in force - has several rules, basically, cancer patients get automatic coverage for FDA-approved NGS tests like Foundation Medicine F1 CDx, for one time use, on-label, in patients with advanced cancer.   For other one-time tests in advanced cancer, LCDs may provide coverage even if the test isn't FDA-approved.

This NCD structure means patients with non-advanced cancer (e.g. breast cancer lumpectomy patients) seem to be excluded from NGS testing, and tests for multiple uses (minimal residual disease in leukemia) seem to be excluded.

There have been stakeholder concerns since January 2019 about the various domains of implied "non coverage," typically areas not actually reviewed and discussed in the text of the NCD.   See e.g. Genomeweb here.  MedTechDive here.

I've also raised concerns on innovation, meaning simple innovation, like simply migrated accepted platforms to NGS; here.

NCD Reopening is Narrow

The reopening domain of interest is limited SOLELY to "germline tests to identify those with hereditary cancer that may benefit from targeted therapies."   Any other comments will be considered off-topic.

Ironically, for example, this headline came in my inbox almost literally adjacent to the NCD reopening:
  • Novel NGS Method for Minimal Residual Disease Monitoring in AML   
  • Here.
  • But the CMS NGS NCD doesn't cover MRD monitoring now (more than 1 test), and the NCD reopening blocks this topic from the accepted comments, since it isn't related to germline testing.
CMS text below the break.   Comment period is open 30 days, from April 29 to May 29, 2019.   The revised NCD is due by October 29, in six months.

For an initial trade journal reading of the new NCD reopening, Susan Kelly at MedTechDive here.  Genomeweb here.

What Will Labs Think?

Myriad's next investor call is Tuesday, May 7, 2019, 4:30 ET, and so is Invitae's.

Very Brief Blog: LabCorp, OmniSeq - LDT to Become FDA-Approved Tumor Panel

In August 2017, LabCorp and OmniSeq, a genomics reference lab in Buffalo, announced a distribution commitment and investment from LabCorp, here.  In June 2018, they jointly rolled out OmniSeq Advanced, here.

In April 2019, additional investments from LabCorp to OmniSeq were announced, here and here.  Especially, the new resources will fund the ability to pull OmniSeq tests through the FDA pipeline.   This is likely spurred, at least in part, by Medicare's National Coverage Decision to give preferential coverage to FDA-cleared and -approved tumor gene tests.

See OmniSeq's home page here.

Caris has also prepared submissions for FDA (e.g. here).  Illumina has a large tumor panel under review at FDA with breakthrough status, here.  Guardant has a liquid biopsy test with FDA breakthrough status - here.

Medicare has proposed to offer special inpatient new-ech payment status in 2021 and later for breakthrough devices (here).

Very Brief Blog: PLA Codes Brought to Life - Rady and Whole Genome Sequencing in Infant ICU

CMS has posted the initial code list for its summer new lab test pricing meeting - here.  47 codes are already up for discussion, of which 41 are new PLA codes.  (And, I expect CMS to probably expand this list in late May with a couple dozen last minute codes coming out of the May 9 AMA CPT meeting).

One of the codes CMS has posted is 0094U, Genome (eg, unexplained constitutional or heritable disorder or syndrome), rapid sequence analysis.

2018: Faernes et al.

In 2018, Rady and its coauthors published Faernes et al., (open access here), a study of outcomes and economics in 42 infants with ICU-level medical crises of unknown etiology.

2019:  Clark et al. 

Newly, in April 2019, Clark et al. have published a detailed methods and phenotyping paper in Science Translational Medicine.   Get it here.  The full title is, "Diagnosis of genetic diseases in seriously ill children by rapid whole-genome sequencing and automated phenotyping and interpretation."  See coverage and interview at Genomeweb, here.

  • See more at the Rady Children's Institute for Genomic Medicine website, here.  
  • Their annual Frontiers in Pediatric Genomic Medicine conference is April 30-May 3 in La Jolla, here.

Federal Legislative Boost?

In 2018, a bill promoting WGS testing in ICU-level Medicaid infants was floated; here; it may reappear in 2019.

Very Brief Blog: Bach/Trusheim on Biosimilars Folly; New Book on Precision Medicine Economics

Two brief items from the horizon scan - Peter Bach and Mark Trusheim and colleagues blog on Health Affairs on why biosimilars aren't working well; and separately, a new academic textbook on precision medicine and economics.

Why Biosimilar's Aren't Effective (...Competition)

In a 2017 article in Forbes, Bach and Trusheim pointed out it would be cheaper for the feds to buy Gilead (even above market price) than pay for all the anticipated sales of Sovaldi.

With colleagues Preston Atteberry and Jennifer Ohn, they're back with a pair of long blogs in Health Affairs (Part 1 here, Part 2 here) on why biosimilars aren't and won't be an effective deterrent (or, a drug on the market) to the high prices of biologicals.

The articles have their own little firestorm on Twitter, and have been picked up in STAT and in ENDPOINTS.  Here and here.

For a recent 2019 piece in Forbes by Dan Mendelson on the same topic, here. For a 2018 view on "the biosimilars problem," also in Forbes, here.   

For a July 2019 response, as a blog in Health Affairs by Brill & Ippolito, here. 

  • Former FDA commissioner Scott Gottlieb was about as forthright as he could be, as Commissioner, in December 2018 in describing biosimilars problems, here.   And he was more outspoken after leaving office, here.

More Links:  Bach x 2

For Bach's testimony to the Senate Finance Committee in January 2019, here.  This is a quite interesting 11-page essay in its own right. 

For Bach's Sloan Kettering, "Center for Health Policy & Outcomes," here; for its Drug Pricing Lab (website) here.

New Book on Economics of Precision Medicine

Just released by the University of Chicago, and available as Ebook and textbook on Amazon (circa $100): Economic Dimensions of Personalized and Precision Medicine, about 270pp.

I've clipped the table of contents below the break.  It includes a chapter on precision medicine and game theory by Berndt & Trusheim.  (Be prepared to google Bertrand Competition).

Thursday, April 25, 2019

Very Brief Blog: Nature Reviews Cancer Publishes Go-To Review on "Cancer Over-Screening"

Screening for cancer is crucial to intercept cancers early and allow early successful treatment.  Some of the most successful screening methods are long-established (mammography, Pap smear, FIT or colonoscopy).   However, new methods are constantly coming on line (Exact Sciences ColoGuard, Epigenomics Epi proColon, low-dose CT for high risk lung cancer (LDCT), enhanced digital or MRI mammography, 4KScore, Prolaris, Decipher and others in prostate cancer, etc). 

Even with the oldest methods (mammography), the best rules and practices remain contentious (including start age, stop age, role of family history on practices, role of genetic risk burden, etc.)  Providing different screening rules based on risk or genetics seems especially murky.   Today in 2019, Medicare defines high-risk colon cancer screening solely by "family history" somewhere amongst your relatives, entirely ignoring whether or not you actually inherited the gene at risk or not (here).

New Review of (Over)-Screening:  Srivistava et al., 2019

Although it's not open access, Nature Review Cancer has published a comprehensive 10-page review that will probably be the go-to reference for the next couple years.   Find Srivistava et al., 2019, hereAuthors are from NIH, MD Anderson, Univ California, Hopkins.

I've clipped the abstract below the break.   Authors note that overdiagnosis rates vary a lot, they are not an issue in cervical or colon cancer, but rise to 25% in breast, 60% in prostate. 

In the conclusion, they view better molecular determinants as a path forward for better public healthcare in this area.

For a new article on budget impact modeling of cancer screening, Jahn et al. 2019, here.  For an article in Stat, tied to one in JAMA, on "excesssive use of testing," here. For an April 2019 article in The Economist on screening in lung cancer (and other purposes), here.

Wednesday, April 24, 2019

Very Brief Blog; CMS Launches Proposed Inpatient Rule (FY2020); Tweaks NTAP Payments

On April 23, 2019, CMS released its annual spring proposed rules for Inpatient Hospitals, including new technology add-on payments.
  • CMS press release here.
  • Kaiser Health News here.
  • Boston STAT (subscription) here.
  • Discussion of NTAP by law firm Hogan Lovells, here.
  • The Federal Register home page for the rule is here.  
  •    84 Fed Reg 19158-19677, May 3, 2019.  Here.
  •    Rule CMS-1716-P, comments til June 24, 2019.
  •    CMS IPPS CY2020 home page (additional files, data) here.
For the NTAP program, devices must currently meet several criteria including being not just literally newly approved, but more broadly a "new" form of technology, as well as having substantial health improvements in Medicare patients ,and substantial cost relative to its DRG.  Currently to be "new" means to fail several criteria that would make it "substantially similar" to an existing device.

Expedited Pathway Proposal(s)

CMS proposes to deem a device "new" if it comes out of an FDA expedited approval pathway (II.H.8).  They also proposed an expedited pathway device will be entirely exempted from the need to demonstrate "substantial" clinical improvement (p. 19161-2 and 19367ff).

The rules for becoming an FDA breakthrough device were published by FDA recently, in December 2018, 30pp, here.

Clarifying a "Substantial Clinical Improvement"

Separately, CMS is also proposing some changes to clarify what is "substantial" improvement. (Current phrasing uses examples that it reduces mortality, decreases hospitalization or physician visits, or reduces recovery time.)  See current rule summarized at 19273-5, and revisions presented at ff.  See notes at bottom of this blog.

Add-On Payment: Boost to 65%?

In addition, CMS proposed to boost the add-on payment dollar amount from 50% to 65% of the additional cost.   (At one time, NTAP payments were budget-neutral, but they no longer are.  However, NTAP payments are a tiny part of Part A spending, such as 0.03% (1/30 of 1%); see excellent recent review here.)

NTAP payments generally last 3 years from device introduction, then stop.  Comments on the existing cohort of NTAP payments (e.g. from 2019) at 19276ff.

Diagnostics - T2 Molecular Microbiology Test

For those interested in diagnostics, a discussion of add-on payment for T2 Biosystems rapid bacteria molecular panel is at p. 19356ff.  For example, CMS discusses whether it is "new" or "substantially similar" to prior microbiology systems.   "We note that the T2 test panel uses DNA to identify bacterial species...standard of care blood cultures a DNA test is also required...we invite public comments whether T2Bacteria Test Panel is "substantially similar..." to existing devices..."

Nerd note: HEOR?  Add-on payments are for products that increase hospital costs making the DRG inadequate.  T2's device may decrease hospital costs (shorter stays, etc).  CMS doesn't seem to deal with this; for the add on payment, it seems to be looking at increased hospital device costs, and not potential longer term savings like reduced length of stay that might be claimed to cancel out the device costs.  Indeed, faster discharges is a written criterion for "substantial clinical improvement" although that would tend to reduce or offset costs of the same new device.

LDT and NTAP:  For those who track FDA vs LDT issues, I don't think an LDT has ever even been proposed for an NTAP.


Key paginations related to this blog include:

19161-2, Overview
19272ff, NTAP section begins (section H)
19356ff  T2 dx device
19367ff, H.6., RFI for substantial improvement
19369ff, H.7., Specific suggestions re substantial improvement
19371ff, H.8., Breakthrough devices to get a "pass" from "new" and "substantial improvement"
19373ff, H.9., Raise payment from 50% to 65%
19672, Appendix A, Section O, no special pathway for drugs under expedited review, only devices

NTAP runs 19272-19373, or 110 pages.

Footnote: Substantial Clinical Improvement

Substantial clinical improvement (SCI) has been open to CMS interpretation, with some landmarks being "reduces mortality, decreases hospitalization or physician visits, or reduces recovery time."

I've seen cases where CMS grants Substantial Clinical Improvement in IPPS but not OPPS, but here, they seem to treat Substantial Clinical Improvement as a unitary concept for IPPS and OPPS.

At H.6 p. 19367ff and H.7 19369ff, CMS proposes some additional options for Substantial Clinical Improvement. 

H.6 is in the form of an open ended request for information about S.C.I.  H.7 structures a separate request for feedback on specific written proposals from CMS.   These are:
  • SCI could be proven by reference to evidence of broad adoption.  If so, how define?
  • Positive clinical outcomes against existing technologies.  This provides a more firm agreement on what the comparison outcome is.
  • Evidence can include real-world evidence and does not necessarily have to be published in a peer reviewed journal before review.
  • Improvement may be defined more specifically to subsets of beneficiaries with certain preconditions, co-morbidities, etc.  However, since ICD-10 categories are crude, this could be hard for CMS to define (there's no code for ALK-positive DRG patients).  
  • SCI is possible without regard to FDA approval criteria; a device might be 510(k) for FDA but be different enough to have SCI.
H.8 proposes to automatically count breakthrough devices as "new" and "substantial clinical improvement" and H.9 proposed to raise NTAP payments from 50% to 65%.

For a 2010 PhD thesis by Bockstedt on the NTAP process and outcomes, here.  For some 2018 concerns from the pharma community, here.

Existing CY2019 NTAP payments include:  Defitelio (19276), Stelara (19276), Zinplava (19277), Kymriah/Yescarta (19278), Vyxeos (19279), Vabomere (19280),  Remede/Respircardia device system (19281), Zemdri (19281), Giapreza (19282), Claret/Sentinel cerebral device system (19282), Procept Aquabeam prostate system (19283), AnexXa (19283).   3/13 are devices.

New technology proposals for upcoming year CY2020 include (a) Azedra (19284), (b) Capblivi (19289), (c) CivaSheet [nuclear medicine] (19295), (d) Contepo (19300), (e) Duragraft device (19305), (f) Eluvia stent (19312), (g) Elzonris (19318), (h) Erdafitinib (19322), (i) Erleada (19325), (j) Spravato (19329), (k) Xospata (19335), (l) GammaTile (19339), (m) IMI/REL injection (19343), (n) Jakafi (19346), (o) Downstream Oxygen System Therox (19352), (p) T2 Bacteria  Test Panel (19356).  At 19367, the proposed modifications to NTAP begin, as shown above.  The list includes 5 devices and 1 test along with 9 drugs.

Saturday, April 20, 2019

CMS NCD on NGS in Cancer Doesn't Fit FDA Approvals and Blocks Healthy Platform Migration

In March 2018, CMS finalized a National Coverage Determination on Next Generation Sequencing when used in cancer patients.  Although the effective part of the NCD is just a few sentences out of the 80-page total text, those several sentences have powerful implications for the development of cancer care and precision medicine as a whole. 

This essay highlights two key issues. 

On-Label Uses of NGS Testing

First, there are already examples where the NCD for NGS blocks use of on-label FDA-approved testing for drug management. 

These include blocking use of Her-2 genomics for on-label adjuvant therapy in breast cancer with on-label uses of Herceptin and related drugs.   Another problem area is on-label, on-guideline use of leukemia/lymphoma chemotherapy (or decisions for bone marrow transplant), when NGS platforms are used for minimal residual disease detection. 

These blockades - which will multiply with time on a rolling basis - are not seen in the proposed NCD for CAR-T therapy, which allows coverage for FDA-approved uses and on a future basis for NCCN-endorsed uses of CAR-T as they appear.

Validated Platform Transition to NGS Platforms

The second key issue is blocking the transition of tests that are approved on one platform, from use with equivalent or better results on a new platform (NGS).

For example, Agilent and others now have high-efficiency, accurate RNASeq platforms that many qRNA MAAA tests could transition to (see example here and here).  However, if the RNASeq is viewed as a form of NGS, then coverage stops as soon as the transition or bridging has been fully validated.   Various classes of precision medicine oncology test, including FISH and IHC, become more accurate when translated to NGS platforms (see Lin et al. 2019, here.)

Blocking these transitions is a really bad idea.  It's as if NIH said, you can have this grant, but you have to use rotary-dial telephones, not push-button telephones.

Sanger (Left), NGS (Right)


Snapshot: The NCD Rules

In the 1940s, the prolific author Isaac Asimov introduced "Three Laws of Robotics," which were very simple but led to many ramifications explored in subsequent short stories and novels.   In brief, the First Law, a robot may not harm a human being.  Second Law, a robot must obey orders, except when it conflicts with the first law.  Third Law, a robot must protect its own existence, except not in conflict with first or second laws.

The NCD can also be summarized quickly along the same lines as Asimov's science fiction.   First Law, an NGS test shall only be used in patients with recurrent or metastatic cancer.  Second Law, an NGS test shall only be used once per patient.  Third Law, an NGS test is covered if approved by FDA as a CDx, and also, used for an on-label patient indication, but only if it doesn't conflict with the first or second laws.     Fourth Law, an NGS test may be covered by LCDs, but only if it doesn't conflict with the first or second laws.

The First Law and On-Label Uses

Here, I've extracted the coverage rules in the NCD so that the First Law is that the NGS tests shall only be used in recurrent or metastatic cancer (e.g. stage III, IV cancer).   There are already examples where FDA approvals overrun this rule.

The most prominent is for Herceptin-class biologicals used as adjuvant therapy in breast cancer.   The FDA has approved the Foundation Medicine test for Her-2-neu genomics, based on an FDA-validated bridging study to an FDA-approved FISH test.  See here.    FMI presented bridging accuracy studies in 317 blocks, 125 positive, 192 negative (various analyses, TP, FP, etc, had 80%-96% agreement).   Herceptin is approved as a major adjuvant therapy for early-stage breast cancer that has, or has not, spread to lymph nodes (here).   The 2017 labeling is here.   Hint: That labeling is nowhere cited by the NCD. 

The adjuvant studies were specifically based on gene amplification, to which the FMI labeling and approval are bridged by the FDA.  Tumors were equal to (or larger than) T1c, e.g. T1 tumors in the 10-20mm range.  They weren't stage 3 or 4.  These fall outside the tumors allowed by the NCD's First Law requiring advanced-stage disease. 

FDA labeling for the drug specifically refers to use of "FDA-approved tests" for Her-2 overexpression, thus clearly including the FMI test by reference and providing labeled uses in Stage 1-2 cancer patients.

Adding a targeted biological may improve adjuvant therapy for early-stage resected cancer, but not always (e.g. not for cetuximab and colorectal cancer, here).  However, adding targeted drugs can have a substantial value for clinical outcomes in early-stage disease, including on-label uses, as shown in the case of breast cancer.  And MSI-family mutations (e.g. MLH-1, MSH-2 status that are accurately diagnosed by NGS including report on the FMI test) can be used in stage 1-2 adjuvant therapy decisions (see UpToDate here; Tougeron et al. here.)

The Second Law and Multiple Uses

One of the most important concepts in hematopoeitic cancers is the detection of minimal residual disease.   This was an early use of flow cytometry, but has migrated to molecular methods (e.g. molecular BCR-ABL with  serial tests) and now to FDA-authorized NGS methods (the ClonoSEQ test, here).   FDA has increasing moved towards using MRD as a fundamental outcome (here and here), but it is already universally used as a decision point for leukemia management.

Within Medicare MACs, the MolDx system covers ClonoSEQ in Medicare patients in the form of one, one-time test cycle with up to four MRD assays (A56322, here). 

As more tests in more geographies migrate to NGS platforms for MRD, it is imperative the NCD be updated to allow more than one test per lifetime per patient in these leukemia and lymphoma patients.   This would bring the NCD into consistency with new and constantly evolving FDA labeling for cancer care.

Platform Migration

It's easier to see what is blocked by the NCD that what is prevented from development.   

With advances by Agilent and others, it's now directly possible to migrate qRNA tests (most MAAA tests) onto RNASeq platforms (here and here.)   Almost the same day that the NCD was released controlling NGS test use in the US, such tests were being released with approval of European authorities (here).   But the First and Second Law of the NCD block use of the tests solely because they are migrated onto NGS platforms.   This may prevent tests from being migrated to more efficient, faster, or more accurate technology platforms.  (Hence my example earlier, comparing rotary dial and pushbutton phones). 

For example, a recent major breakthrough has been FDA-approved blood tests for rare RBC antigens, tests that are especially important in African-America and other minority populations.  If such tests (example here) were migrated to NGS platforms for efficiency and cost, the NCD would make them non-covered.  Or rather, they'd be non-covered in all Americans with a diagnosis of cancer, even a small skin cancer, but they would be covered in a dying patient with advanced metastatic cancer, which makes no sense.

Tests should be valued for their impact on care - when they are reasonable and necessary - not what platform they are run on.

A Note on Platforms

The NCD seems to treat NGS platforms as "a device" rather than as a tool or modality (like light, pumps, or wheels).     

While there is one FDA category - for FMI-like tests - that defines Category II NGS platforms for use in tumors to give reports (510(k) reports) to physicians that are not drug-specific - most FDA sequencing test categories are not method specific.    For example, FDA device category 21 CFR 866.6080 is NGS-specific, it's a 510(k) category not a CDx or a PMA category, and more importantly, many FDA sequencing test categories like 866.5940, 866.5900, 866.6100, 866.3365, are not sequencing technology specific. 

Similarly, the FDA PMA category for germline cancer tests isn't specific to NGS either (here; Myriad BRCA CDx approval here).

NGS is no more an FDA device category than are "pumps" or "things that use light."  The former includes heart pumps, bedside saline pumps, pain pumps, and so on.   A bedside drug pump has no inherent medical necessity separate from what it is pumping, and an NGS platform has no inherent medical necessity (or approval path) separate from what it is sequencing (MRSA bugs, BRCA genes, KRAS, etc.)   Nor are "things that use light" an FDA category - see ophthalmoscopes, colonoscopy devices, laparoscopic devices, AI retinal imaging devices, psoriasis UV therapy lights, and so on.   There is no reason for an NCD on the FMI test in oncology to act so broadly as to even block the use of microbiology NGS tests in septic cancer patients, although that's how the NCD was written.


FDA doesn't attempt to cover all uses of "pumps" or "light" in one guidance document, nor is there one payer or CMS NCD policy written to handle "pumps" or "light."  FDA doesn't remotely attempt to cover all uses of NGS testing in healthcare in any one guidance document, and CMS shouldn't try to do so in one NCD either.   CMS ends up with something like the "Four Rules of NGS Testing" I've utilized here, and they quickly run into contradictions or insufficiencies.  It's like trying to have "The Three Rules of Chess" - it won't ever work.

Friday, April 19, 2019

Systematic Review Pummels Diagnostics RCTs - Because Docs Know If They See Test Results

From time to time everyone sees "Hierarchies of Evidence" that start with case reports at the bottom and rise through RCTs to the highest point of evidence, meta-analyses of RCTs.

Everyone should be aware that RCTs that are pivoted on a diagnostic test can be an inefficient way to study the impact of diagnostics.  For example, in a drug trial between arms A and B, everyone in Arm A gets Drug A and everyone in Arm B gets Drug B.  In a diagnostic trial, if you use the diagnostic only in Arm B, maybe 20% get a change because of an especially low result, and 20% get a change because of an especially high result.  But in this example, 60% of the patients in arms A and B of the diagnostic trial are treated exactly the same way and should have exactly the same result in both arms.  This dilutes the apparent impact of using the diagnostic.  (See longer viewpoint here.)

Along with others, I've pointed out for years that an RCT can't get a perfect score with a diagnostic, because it can never be blinded as to whether doctors in the diagnostic arm know the result of the diagnostic.   (For example, you can't give fictional placebo randomly positive or negative cancer PET scan reports in one arm, and real cancer PET scans in the other arm.  Shiver.)

However, I just ran across a meta-analysis from NIH and Harvard that dings a diagnostic test RCT because the doctors weren't blinded to the fact a diagnostic was used and thus became part of their decision.   Good lord.

The article is Pepper et al. and looks at RCT results across more than a dozen studies in critical care sepsis patients where procalcitonin is used to monitor infection and assist management.  (Procalcitonin rises and falls with the severity of bacterial infection.)    The meta-analysis concludes that procalcitonin added to standard of care reduces antibiotic dosing days by one to two full days, and may also reduce mortality, but not with a strong effect. 

The abstract and body of the study repeatedly discusses the scientific problem of bias in the studies.  However, at only one point do they clearly describe what that bias is, and one of the main charges is that doctors were not blinded and knew that they knew procalcitonin results in the intervention arm.  How on earth else would you design a diagnostic study??   The authors state at five different points that the results were marred by "high risk of bias," but only at one point that this included primarily non-blinded doctors given PCT test results.  Some formal estimators of bias (e.g. funnel plots for publication bias) were negative.


Extra credit...

Recent papers on predictive and prognostic test evaluations.

Wolff et al. (2019)  PROBAST: A tool to assess the risk of bias and applicability of Prediction Model studies.  Annals Intern Med 170:51-8.

Moons et al. (2019)  PROBAST: ...Explanation and Elaboration.  Annals Intern Med 170:W1-W33.


Riley et al. (2019)  A guide to systematic review and meta-analysis of prognostic factor studies.  BMJ 364:K4597.

Flurry of News: AI in Medicine and Creative Destruction

German economist Joseph Schumpeter coined the term Creative Destruction (schöpferische Zerstörung) to describe the growth and collapse of businesses and industries under capitalism.  (Today "creative destruction" is sometimes used enthusiastically to describe dynamic change and progress.  In his day, Schumpeter saw this less as upward growth and more as an eventual downward spiral, a "fall-of-the-west" or "implosion-of-capitalism" viewpoint.) 

Today with AI (and other new technologies in healthcare) we often construe "creative destruction" as an interesting idea, investment, bubble, hype, and then disappointment a la the Gartner Hype Cycle.   Is IBM Watson becoming an example?

First Example:  Derek Lowe's Assessment of IBM Watson in Drug Discovery

See an open access article by Derek Lowe at his Science blog, In the Pipeline.  He reports on April 18 that STAT reports that IBM has has canned its much-vaunted "Watson for Drug Discovery" now.   But here, Watson for Drug Discovery can probably stand in, almost with canned text, for many efforts to turn big data into useful health outcomes.   IBM wrote...
Watson for Drug Discovery reveals connections and relationships among genes, drugs, diseases and other entities by analyzing multiple sets of life sciences knowledge. Researchers can generate new hypotheses using the resulting dynamic visualizations and evidence-backed predictions. . .Pharmaceutical companies, biotech and academic institutions use Watson for Drug Discovery to assist with new drug target identification and drug repurposing. Connect your in-house data with public data for a rich set of life sciences knowledge. Shorten the drug discovery process and increase the likelihood of your scientific breakthroughs.
Clearly, this could be easily edited into "Harvard for Drug Discovery" or "Mayo for Health Outcomes" or "Optum for Public Health" -- to give some wholly hypothetical names to applications for AI where the existence of real projects wouldn't surprise you.

Lowe's blog has a lot of detail about IBM Watson for Drug Discovery, and if you don't subscribe to STAT, he also links to a recent open access article at IEEE Spectrum "How IBM Watson Overpromised and Underdelivered on AI Health Care," here.   (IEEE first chimed this note in 2015, here.)

"Overpromised and Underdelivered" is a truly impressive deep dive piece by IEEE senior editor Eliza Strickland, who also manages the May 2018 IEEE series, "Hacking the Human OS," which is a wealth of interesting open access articles on similar themes.   Find it here and track their running medtech blog here.

For addition IBM Watson bad news, see Forbes on a fiasco with MD Anderson, 2017, here.  See a 2017 negative STAT article on IBM Watson, here.  See a 2018 article on layoffs, here, and a 2018 article on scaling back the IBM Watson hospital services business here.  Over 5 years, IBM stock has slipped from $190 down to $140.  Microsoft is up from $45 to $120, and Apple up from around $100 to around $200.   During that time, Dow Jones is up from 16,000 to 26,000.  (GE has slipped from $25 to $10.  GE spunout its biopharma business to Danaher (here) but is working with ACR on AI in imaging...here...see next story.)

Second Example:  AI in Imaging

Creative destruction is a dynamic mix of upswings and downswings.   Here's an upswing.  At the same time, we read that AI medical imaging startup AIDOC has just raised a new $27M for medical imaging based on AI, here.

And institutions are buying in, including prestigious bodies like NIH and American College of Radiology.   See an article in MedTechDive here, that links to a press release from Radiological Society of North America, that leads to a new NIH/ACR position paper on the expected rapid and important growth of AI in imaging.  That last document is here, Langlotz et al., and it's $30.  (As mentioned earlier, GE is tying in with ACR re AI R&D, here.)

See also the recent trade press here, about a new 20 page FDA white paper and guidance document on AI in devices, here, which was released with a press release by Scott Gottlieb, here.

FDA has begun approving AI-driven devices, such as for retinopathy diagnostics (here).  Those approvals have been for locked-software devices; the new guidance moves FDA toward self-updating AI systems. 

Digital Pathology, Too

CAP Today ran a long cover story on digital pathology in February 2019 - by Karen Titus, here.   (Admittedly, focused more on digital storage and interpretation than machine learning or AI.)  But see a very interesting article that we don't need "digital pathology," but rather "intelligent digital pathology," by Acs and Rimm in JAMA Oncology in March 2018 (here).   See some additional autumn 2018 headlines in digital pathology at blog, here.

See a 2018 review article on AI and digital pathology by Tizhoosh and Pantanowitz, open access, here or here.  See a 2018 article in IEEE Spectrum, "The First Frontier for Medical AI is the Pathology Lab," here.  Similarly, see a trade press article in Healthcare-In-Europe, October 2018, here.

See a May 2019 article by consultancy DeciBio on Digital Pathology and Multiplex Spatial Tissue Analysis, here.

In another area of medical technology, Edwards Lifesciences inked a deal in December 2018 with Bay Labs for AI-assisted product development in cardiac devices (here).

If It's in NEJM, It's Probably True

For another institutional endorsement, see the brand new review on machine learning in medicine by Rajkomar et al. in New England Journal, April 4, 2019, here

For recent webinars from Rock Health about "How to Exit [financially] in Digital Health" and from Accenture/Medtronic on "How AI Can Change the Future of Healthcare" - see here and here, respectively.

If the UK National Health Service Is Doing It, We Hope the Economics Are Sound

For a "Topol Report" which is a 50-page roadmap to digital health plans for the NHS over the next decade - here.

Wednesday, April 17, 2019

Update: CMS Posts First Code List for June CLFS Crosswalk Meeting

In late March 2019, CMS announced its annual summer new lab test pricing meeting for June 24.   See the full discussion at my post a few days ago, here.

On April 15, CMS posted a code list of codes to be discussed.   Comments and participant registration are due by June 10 (see prior link).

3+3+41=47 Codes To Be Discussed

CMS posted three codes under reconsideration - BRCA1-2 code 81163, BRCA-1 sequence 81165, and 0046U, a tyrosine kinase gene.

CMS posted only three new regular pathology codes, 813X1 (PALB2), 813X2 (PALB-2 family variant), 8XX01 (PIK3CA, targeted analysis).   There seem to be no new non molecular lab codes, if this list is complete.

Finally, CMS posted a bonanza of 41 PLA codes, including those approved in Q12019.

Will CMS Add May 2019 PLA Codes?  Like Last Year?  Past as Prologue?

Last year, the late June CMS CLFS meeting included those PLA codes approved in early May 2018 by AMA. 

We expect that AMA will be approving about 30 PLA codes this year in early May, so it is quite possible, but not certain, that in late May CMS will add them to the June 24, 2019 agenda.   CMS needs to receive the May 2019 codes from AMA in time to post 30 days in advance of June 24, 2019.

Download the Spreadsheet

Go to the CMS CLFS Public Meeting Page, find CY2020 code list near the bottom.

2019 Gapfill Process

CMS has about 18 codes under CY2019 gapfill process, with prices being set this spring by MACs.  CMS should post proposed gapfill prices in April or May or June.  Here (see bottom section of that blog for gapfill roster).

Friday, April 12, 2019

Contractor Snapshot: MolDx Rules for Open LCD Meetings

On March 28, MolDx released a host of new proposed LCDs.  The public has a chance to comment during a public meeting to be held May 6 in Columbia, SC.   The deadline for meeting submissions is today, April 12.

The public meeting website is here.  Since this event webpage is probably temporary, I've put a cloud copy of the meeting rules here.

Some tidbits:

  • Comments are required 2 weeks after the LCDs were releaesed, and about 3 weeks ahead of the meeting.
  • Just to attend, advance registration is required, but this may be optional if there are still seats available on May 6.
  • The meeting will be recorded and posted on the MolDx website.  This is a new CMS rule regarding transparency.
  • They have a specific remark, if a stakeholder has negative comments on an LCD, please be specific and cite literature.
  • BYOL - bring your own laptop if you want to project PowerPoint.
  • There are 12 LCDs and 120 minutes, suggesting 10 minutes per LCD. 
MolDx LCDs are also released in many other jurisdictions, WPS, Noridian, etc, which have their own public meetings on their own cycles. 

Some of the rules, such as posting of video or transcript, follow new LCD rules released a few months ago by CMS - find them here.

Thursday, April 11, 2019

AMA Posts Over 30 Proposed PLA Codes (May CPT Vote)

On April 10, 2019, AMA posted over 30 proposed new PLA codes for public comment.  Public comment is brief; apply to AMA right away to get a copy of an application of interest to you and comment to the committee by April 18.   The AMA PLA committee will deliberate until April 24, and the codes will be voted on by the AMA CPT Editorial Panel in Chicago on May 10.   The final codes will be posted July 1 and active October 1.

  • PLA PDF agenda here.
  • PLA home page here.
  • PLA calendar page here.
  • The web page for the May CPT meeting is here, including the agenda of all regular (non PLA) codes proposed.

What are PLA Codes?

PLA codes are special rapidly issued AMA CPT codes for lab tests that are offered in the USA and are either proprietary, FDA-approved, or both.   

Will these codes enter the June 2019 CMS pricing meeting?

If last year is precedent, PLA codes from this cycle will be included in the June 2019 CMS crosswalk/gapfill meeting for pricing new lab codes.  (Here).  

How Many Recent PLA Codes?

From August 2018 to March 2019, AMA created approximately 52 PLA codes, and if all 35 new codes are created in May, as many as 87 PLA codes will enter the June crosswalk/gapfill meeting.

Monday, April 8, 2019

PAC CARB Votes: HHS Should Tell CMS to Finalize Antibiotic Stewardship Rules for Hospitals

We live in a world where antibiotic stewardship and antibiotic resistance are major public health topics.  Just this weekend the New York Times ran two stories on the antibiotic resistance crisis (here, here).

One major federal effort was announced by CMS in 2016 with some fanfare:  Requires all hospitals serving Medicare to have official Antibiotic Stewardship Programs in operation. 

However, the rule has never been finalized, and it will expire if not finalized by June 2019.   Stakeholders who think this requirement would be a good idea have a major forum:  PAC CARB, the President's Advisory Commission on Combating Antibiotic Resistant Bacteria.

Today, April 8, 2019, the PAC CARB held a special session and its commissioners voted that CMS should move forward with the ASP requirement.   All the commissioners voted "Yay."   The letter (as currently drafted) is online here

Several government bodies like FDA, NIH, CDC, BARDA, CMS, are assigned to the committee.  However, in today's meeting, like the January 30 meeting, the CMS delegate to PAC CARB did not attend the meeting.
  • See an unofficial meeting transcript in the cloud here.  

Get a window into stakeholder positions and politics here, at the website of APIC, the Association of Professionals in Infection Control.


I made a personal professional comment to the committee; here.


Some additional entry points into the topic of Antibiotic Stewardship Programs for hospitals...

  • In 2014, CDC released "Core Elements of ASP" (24pp), here, web here.
  • In 2016, Joint Commission released standards for ASP, see short summary 4pp here.  Register for 96pp PDF "Toolkit," here.  (Note - cobranded JC/Janssen).
  • In 2016, National Quality Partnership (NQP) of National Quality Forum (NQF) released a strategic consensus viewpoint, the 38pp "Playbook" for ASP, here.
  • IDSA has numerous subject-specific guidelines and has an accreditation for Centers of Excellence in ASP, here.