Thursday, December 13, 2018

Very Brief Blog: AMA Posts Agenda for February 2019 Mtg; Dullsville for Lab Tests

The next AMA CPT meeting will be February 7-9 at the Westin La Paloma in Tucson (overall agenda here). 

Thursday February 7 is the CPT annual meeting, mostly policy updates and discussions.  February 8-9 is the Editorial Panel session for new codes.

The new codes have been posted.   If you were interested in commenting on lab codes, the public posting was November 19 and comments to November 30.   If you are interested in seeing and perhaps commenting on other codes, they are posted December 7 and comment open to January 17.  See the code level agenda here.

For the lab industry, not much.  Tab 13 deletes administrative MAAA code 0009M.  Stay awake, there's more.  Tabs 14 and 15 elevate Tier 2 codes to Tier 1, which is good news for fans of Palb2 or PIK3CA.   Tab 24 would create a Category III code for a stem cell cytotoxicity assay (my guess is this could have been proposed either as Cat III or PLA). 

About 50 Category III codes (as old as 0058T and as new as 0405T) are scheduled for a sundown discussion (more like sunset, I'd say).   

Wednesday, December 12, 2018

Very Brief Blog: My Brief Adventure with DTC Testing

Having worked in genomics policy for years, I have never explored DTC testing.  I ordered panel testing from Color and microbiome testing from Ubiome. 


Ubiome testing was easier than I thought (you tap the paper with a swab).  A report took several weeks.  (A) I had no "pathogens," (B) my normal flora was "normal," and a range of I guess (C) particularly healthy bacilli (various lactobacilli) were reported separately.


Kit. Color sent a simple kit with an empty vial to spit into.   After about 30 days, it provided a hereditary cancer panel, a heart health panel, and so far at 6 weeks has reported 2 of about 14 CYP genes. 

Cancer & Cardio.  I had no hits on the cancer panel (recall that the general prevalence of e.g. BRCA is well below 1%).

This wasn't surprising; Color "warns" you, so to speak, that most people will be negative in their tests.  And my extended maternal/paternal history is unusually negative for any cancers. 

I also had no hits on the cardio panel either (and I have no family history for A-Fib, for sudden death, or for familial hypercholesterolemia). 

CYP.  The first two available CYP genes came with strong click-through warnings not to change any meds without talking to a physician.   I was normal (intermediate) for CYP2C19 and CYP2D6. 

Color tells me that twelve more pharmacogenetic genes will be reported, but in up to six months.  I'm curious if I'm slow for CYP1A2, since I'm hypersensitive to caffeine.

Caveats.  I assume these tests look for the more common variants and if they aren't there, the default is "normal."

  • Despite clicking around, I couldn't find a list of the 12 (yet to be reported) CYP genes.
  • Despite clicking on a cardio link that promised "learn more about these genes," I didn't learn anything.  
    • So I'd be left googling Wikipedia to figure out what cardio gene MYH7 was.   
    • (I can guess that KCN and SCN are K+ and Na+ channels, but maybe my PhD in neuroscience helps there.)
  • The "learn more" click for cancer was more helpful; for example, I learned if my APC gene was positive I'd have a very high risk of CRC. 
  • Based on the various direct-to-consumer email traffic to me from each company, I believe each test was "ordered by a physician," somewhere, based on my age/sex/and very basic clinical checkboxes during enrollment.





Aside.  While my Color cardio panel was negative, I do have strong maternal/paternal loading for common hypertension of aging, both parents having had poorly controlled hypertension and dying of cardiovascular causes (father 60, mother 80). 

Accordingly, from my early 50s to late 50s my blood pressure switched from around 90/60 to around 150/100, and it seemed completely unrelated to sodium.  A daily generic tablet keeps my BP around 115 instead of 150, so I'm lucky.  I confirm BP control periodically with a Qardio bluetooth arm monitor linked to my iPhone.   But none of this issue is flagged by the Color cardio panel, which looks for rare things like potassium channel mutations.

At the podcast Tech Tonics, Harvard professor Sek Kathiresan recently discussed polygenic modeling of cardiac risk (here).  I might have that; but my hypertension-of-aging isn't something picked up by single disease genes like SCN5A.

Very Brief Blog: Gottlieb TweetWatch: Tweetorial on IVD, Happy-Tweets with CMS

In the past week, Washington news on IVD/LDT policy has included:

  • December 6, Policy outlook/overview by Gottleib and Shuren at FDA, preceding new legislative draft. Here.
  • December 10, Hill releases new 200 page model for FDA legislation.  Here.
    • I've heard that Hill is taking comments til early February.
Perhaps lost in the whirlwind events - today, December 12 Gottleib announced biosimilar policy for insulin - does this guy ever sleep? - were some equally interesting but less high profile Gottleib tweets.  

First, he issued an extended TWEETORIAL on FDA and diagnostic tests, and the differences between IVD/LDT.   See PDF in cloud here.

Second, he traded happy back-patting tweets back and forth with Seema Verma of CMS:

Third, and this went viral, he conveyed a Dr Seuss style poem after CDC warned about hazards of unbaked cookie dough (making CDC the Debbie Downer of the federal twitter landscape):

Monday, December 10, 2018

Very Brief Blog: SUPPORT Act is Federal Legislation Affecting Lab Sales Practices to All Payers

In October 2018, Congress passed the SUPPORT Act, an omnibus opioid abuse legislative package.  An "11th hour" addition, Section 8122, could affect lab practices to all payers, not just Medicare and Medicaid.   Practices include compensation for blood draws, on-site phlebotomists, sales force compensation, etc.
  • Legislation here (Section 8122).
  • Covered in Dark Report, December 3, 2018 (subscription)
  • Open-access discussion at Epstein Becker Green, here.
  • Open-access discussion at Mintz law firm, here.
  • Memo on the ACLA website from Alston & Bird, November 26, 2018, here.
  • Letter from ACLA to HHS re: anti-kickback improvements; letter mentions Section 8122.  Here.
  • Washington Post here.  
At a glance, to me, I would have thought the law applied to relationships between opioid treatment centers and tox labs serving opioid treatment centers.  But it applies to opioid treatment centers OR labs, regardless of whom the lab is dealing with.

(According to Dark Report, p. 9, this is scope is "open to debate" in view of the legislative history and intent, or, "applies to all laboratories and to all services.")   Everyone agrees it's a criminal provision.


The law develops out of an earlier bill, HR6878, Eliminating Kickbacks in Recovery Act, EKRA, but the SUPPORT Section 8122 is not exactly the same.


Pop Quiz.

What is LEIE?  The HHS OIG list of excluded individuals and entities.  When was it last updated?  Today!     How many entities are on it?  70,847.  How many are entities (clinics, pharmacies, DME suppliers)? 3087.  How many are people?  67,760.  How many are labs?  45.  How many are hospice owners/employees?  35.  Related to infusion centers?  40.  Government employees related to the Health Resources Services Administration (HRSA/HHS)?  3.  Government employee of Substance Abuse and Mental Health Administration?  1.

By the way, if you want someone's name, address, and birthday, without breaking into Target's computers, they're here.  Youngest birthday seems to be April 29, 1990 (28).  There appear to be a couple dozen earlier than 1910, predominantly classed general practitioners. 

December 10, 2018: Legislators Release New 200-page Draft Diagnostics Legislation

Last Thursday, December 6, FDA released a rather quirky statement on its intention to encourage innovative diagnostics while regulating LDTs.  LDTs would be regulated by new methods, risk-based, and including "pre-certification" aka self-certification.   The statement, released in a series called "FDA Voices," opened with a statement by Commissioner Scott Gottlieb, then a statement by CDRH leader Jeff Shuren, then a statement from the legal team, Lauren Silvis, focusing on hazards and risks of unregulated LDTs (bad cop?).

December 7, as publicized by Genomeweb here, the Hill has released a new 200-page version of legislation that would greatly reform the FDA's approach to clinical diagnostic tests.

  • See the 200 page legislative online here.
  • It's now called VALID - "Verifying Accurate Leading-Edge IVCT Development Act of 2018."
  • This legislative version follows a prior draft legislative proposal in May 2017, and a 60-page FDA version of a better idea released in August 2018.
    • It includes pre-certification as law, something that FDA has been proposing in speeches and pilot programs.
  • Biocentury here, AACC here.  Sen. Hatch here.
  • I've heard from DC experts that legislators are welcoming comments until early February, and that the 200 page bill has some holes in it, as it was put together quickly.

Can such an act possibly pass in this section?  You can assume not (a couple legislative days left.)  Sen. Hatch (retiring) regrets he will not be around, in 2019, to see the possible future action on the bill.

Stay tuned for CY2019 and a new Congress.


Sunday, December 9, 2018

Steeply Falling Valuations: HLI, Nanthealth, Vermillion

Wall Street Journal has posted a story that Human Longevity Inc (HLI) is raising funds under the strictures of a "down round," which has lowered the valuation from $1.6B  eighteen months ago, to $310M today (negative 80%).   The round would raise $25M under rules (WSJ uses the word "onerous") that would protect new investors.   WSJ story here.  MedCity here.   (MedCity links to a cloud term sheet which I don't believe WSJ did (here).)

According to the Journal, employee headcount has fallen from 300 to 150, and "its chief executive officer, chief medical officer and chief operating officer all departed in 2017, according to their LinkedIn profiles."   Google suggests that Chief Technology Officer has also been a rapidly passed-along title.

Other Declining Valuations

Other declining valuations include NantHealth, which was valued at $18.54 on 5/28/2016 and 79 cents per share today (4.2%).   

Another example of downward valuation would be Vermillion, which markets ovarian cancer laboratory tests.  VRML started at $310, on 11/1/2000, during the original internet/biotech bubble.  However, after slipping to $30 in the 2002 selloff, it later rose to $123 in September 2003, falling to 30 cents in December 2008, and rising to $32 in February 2010, related to FDA and CMS approvals.   On December 7, 2018, it closed at 52 cents.  Over the past four years, sales in 2014-2017 at Vermillion are about $2-3M per year, GAAP losses around $10-20M per year.  If you roll back to quarterly investor calls around the time of FDA approval (9/2009) and Medicare LCD coverage (3/2010), I believe (from memory) the company's management forecast was for revenue very near term in $millions, then $10s of millions within a few years.

Tuesday, December 4, 2018

Very Brief Blog: FDA Recognizes ClinGen Database for FDA Genetic Tests

Continuing its effort to smooth new pathways for FDA review of genetic tests of all types, FDA announced it has recognized ClinGen / Expert Curated Human Genetic Data as a database that may be referenced in submissions.

FDA press release (December 4, 2018) here.
For a Twitter message chain, see #fdarecognizesclingen here.

FDA cites its April 2018 guidance on use of databases in genetic test submissions.

Journals and Trade Journals 

Genomeweb covers the December 4 ClinGen story here, and cites back to a recent article in Genomeweb here about a special issue of the journal Human Mutation, here.   Volume 39, Issue 11, November 2018, had a series of open-access articles on ClinGen and Clin Var.

Very Brief Blog: FDA Releases 20-page Device De Novo 510K Proposal

As FDA promised in recent weeks, it has released new policymaking that would update and codify the existing de novo classification process.  De novo requests have risen from 22 in 2012 to 100 in 2017.
  • The published copy of the rulemaking is online here.  83 FR 63127-46, 12/7/2018.  
    • Comment is open for 90 days (March 7).
  • See a detailed pre-release discussion by MedTech Dive here.
  • Press at Genomeweb here.  News at RAPS, here.   
  • Nice review at Forbes by Yiannis Mouratidis, here.
Rules will be placed at 21 CFR 860.1, and under a new Subpart D to Section 860 when finalized.

In part, FDA states that the new rulemaking will increase the clarity and efficiency of the process, and (my reading) place into formal regulations some rules that may be similar to various prior guidances (some of which which FDA cites).  Regulations also have the force of law, which guidance documents do not have.

The de novo process dates to 1997 and was "streamlined" in 2012 by allowing direct classification of some new devices into the de novo process, no longer requiring a rejected 510(k) application.   My sense is that FDA has proceeded for years by policymaking by "guidance" and "process" and has not kept 510(k) or "de novo" regulations up to date.  (See here; for more background here and here).  Here, FDA dusts off 1998 regulations lost to the sands of time at 21 CFR 860 and updates them to reflect current statute and processes.

See the 17 page October 2017 De Novo guidebook here.


Class II Exempt-from-Review Changes

In other 510(k) news this year, FDA published rules regarding when Class II devices would be exempted from review in June 2018 (final rule 83 FR 25910), here

This includes some direct to consumer genetic risk tests used for wellness purposes, for example.   See FDA's regulatory classification for genetic risk tests at 21 CFR 866.5950 (issued November 7, 2017, revised June 5, 2018), here.  FDA received at least one comment that it was making genetic test review too easy.

Last Week's News: Making 510(k) More Rigorous

Today's FDA regulation on 510(k) de novo is separate from a press announcement on November 26 that FDA would seek to toughen up 510(k) predicate usage, here.   Takeaway: To the extent that 510(k) predicate usage is made more rigorous, more tests would be shunted into the 510(k) de novo pathway.   So they are trying to oil the path for de novo before those new devices are shunted into it.


Forbes author Mouratidis recently reviewed AI in pharma trials, here.