Regulatory Open Forum

CLIA and FDA:  Where are we going? 

August 2019

Glen Paul Freiberg, RAC

 

After more than 30 years, I remain upset about the creation and evolution of CLIA 88, the Clinical Laboratory Improvement Amendments of 1988, and commensurate regulations, which revised and expanded the original Clinical Laboratory Improvement Act. However, the reasons for being upset have evolved.  When the law and regulations were first issued, my employer, Boehringer Mannheim (now part of Roche Diagnostics) had point-of-care product lines (capillary blood glucose monitoring), physicians' office test equipment (the ReflotronÒsystem) and large clinical laboratory analyzers, assays and controls.  It was clear from the first reading of the law and regulations that CLIA 88 would kill the physicians' office market and for the wrong reasons.

 

CLIA 88's proposed regulations were promulgated because of quality problems in cytology laboratories, primarily PAP smear testing, reading and interpretation.  

The process at the time was exclusively a manual, microscopic review where many cytology laboratories were operating under poor working and quality monitoring conditions. A valid concern.  But rather than government intervention addressing those manual testing labs, the CLIA update included all tests that were not handled in large traditional clinical labs with just a few exceptions.  At the time, there were no quality or complaint problems with physicians' office equipment like the Reflotron system.  The new lab regulations caused such testing to become cost prohibitive for small physician offices.

 

The result was to deny patients prompt results when a physician deemed such prompt results useful. In addition, the proposed regulations themselves made no sense because the test classification was based on analyte name, demonstrating an outdated government understanding of how point-of-care tests could be performed.   During the initial proposed regulations' comment period, I demonstrated the deficiency at a CLIA meeting in Washington, D.C. by taking a Reflotron system to the meeting. I placed Reflotron system on a table and set three vials on top of the machine showing the CLIA/CDC representatives that one vial contained test strips for glucose, a waived test.  One vial contained identical looking strips for another test, classified as "moderately complex" and a third vial contained identical looking strips for a test classified as "highly complex."  Without plugging the system in, I removed a strip from each vial and showed the group a capillary tube and a capillary pipette explaining that you poke a finger, fill the tube, squirt the sample on the test strip of choice and stick it in the machine.  That's it. Clinical chemistry happens and you get a result.

 

The CLIA/CDC representatives were clearly surprised and while I was not the only person who commented on the proposed regulations, it was clear that test categorization by analyte name would not work.  The regulations were later re-proposed based on the complexity of the steps to perform the test.  Even so, the new laboratory record keeping procedures were so burdensome that the apparent goal was achieved: testing at small physicians' labs mostly went away and the ability to promptly report the results of a thyroid, cholesterol and many other tests with a patient while they were in the office also went away.  To me this regulatory power grab was a travesty as there was never a quality problem with this type of testing and did not need improvement or government intervention with new burdensome laboratory regulations.

 

Concurrently, the large CLIA labs continued and still continue to create their own tests that are not typically regulated by FDA.  However, FDA has been invading the CLIA lab's test development evolution little by little under an internal process called "enforcement discretion."  As a term of art, I disagree with how this term is being used for CLIA labs, it should instead be called "enforcement inconsistency." Where enforcement discretion is appropriate is in howFDA regulates rather than whatthey regulate.  For example, after an FDA inspection when an Inspectional Observations Form, the FDA-483 is issued, choices are made: defer follow up until the next inspection, issue a Warning Letter, or take other regulatory action such as seizure or embargo. This is enforcement discretion.

 

Picking and choosing different activities or services in CLIA labs is an accusation by FDA that the CLIA regulations in place, the certifications received and the competence of the already highly regulated labs and physicians using the data, are either ineffective or insufficient.  Therefore, it's time for a court case to further clarify the limit of FDA jurisdiction and FDA's claim to enforcement discretion on what they choose to regulate, that is, picking and choosing certain CLIA lab created services over others. In other words, I recommend the definition of a device be further clarified to require kit or assay distribution for FDA jurisdiction.

 

With my background in FDA regulatory affairs, many would think I would support FDA efforts in going after select CLIA lab services.  Here is why I do not:  as noted above, my belief is that the FDA jurisdiction is for diagnostic medical devices meeting the definition of the Act that are distributed, not for lab services.   And equally important, FDA cannot keep up with the patient benefitting innovations of the CLIA labs.

 

FDA regulated tests are standardized with set test designs and specifications following the FDA's Quality System manufacturing regulations, 21 CFR 820.  The tests created under CLIA are not distributed medical devices and instead are lab tests that meet CLIA requirements, a separate set of regulations.  And most important, current laboratory science especially in genetic testing, allows for continuous improvements, while the FDA's submission and review process tends to inhibit that.    

 

What attracted my attention to this process have been recent articles on personalized medicine, for example, "Personalizing Cancer" in the August 31, 2019 issue of "Forbes."  Similarly, the opportunity for next-generation sequencing to identify multiple gene issues leading to more effective combination therapies will move too quickly for the FDA processes.  As an example of recent FDA intervention in a CLIA lab, there was a recent Warning letter to a company called Inova Genomics Laboratory.  Here is the reference:

 

https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/inova-genomics-laboratory-577422-04042019

 

It's easy to see why FDA objected to the Inova lab service because the lab was providing information for drug selection and on dosing for drugs' off-label uses.  FDA has claimed the potential for hazards in the company's service without evidence of harm.  While it is accepted that physicians can prescribe drugs off-label and modify dosing, FDA believes that using genetics, even with peer-reviewed publications to back up recommendations, to help with off-label dosing is objectionable.   So, while peer-reviewed publications are typically acceptable for use in supporting an FDA application, using them in this instance apparently is not, since some of the drugs are being used off-label.

 

Inova responded to the Warning Letter by stopping the practices that FDA found objectionable rather than submitting an FDA application or going to court.  Had they submitted an application for just FDA approved drug indications, the potential clearance would have been detrimental to patients the moment a new peer-reviewed publication became available for something not covered by the clearance since the CLIA laboratory could not, under FDA rules, promote this new information.  The point being - how is FDA's interference in the Inova service of benefit to patients and what are the risks outside a challenge to FDA's regulatory process?  Can we not trust the CLIA lab director, CLIA Medical Director and attending physician to use the data appropriately?  Or might this Warning Letter be nothing more than turf protection?  And why is this application of genetic data more important to regulate than all other laboratory created CLIA tests?

 

A few words about the evolution of the use of the FDA "Warning Letter" and as coercion in the Inova case.  When I began work as an FDA Investigator, primary initial inspection enforcement was a "Notice of Adverse Findings Letter" and a "Regulatory Letter."  The former was used to ensure that top management was aware of the FDA-483.  (Sometimes top management was at a location other than the facility being inspected.) A Regulatory Letter meant that FDA had the data and evidence necessary to take the inspected firm to court if FDA did not receive an adequate response to the Regulatory Letter within a specified time frame.   Clearly, having the evidence in place to act and promising to act is both a commitment and a lot of work.  So why generate all the legal preparation work if a threat would be adequate, especially when a Warning letter could achieve change by simple coercion?   The coercion is especially effective if the targeted company is a public company.  Turns out this newer approach has been quite effective, but we need to ask if the approach is being abused regarding CLIA lab services and we probably need to ask through the court system since FDA is not about to change their position on discretion.

 

Therefore, my recommendation is that the next time FDA uses a Warning Letter as coercion in a CLIA lab case, the firm should confront the practice in court and attempt to resolve FDA's jurisdiction limitations and its consistent application of the regulations.  FDA has a terrible record in court so this may be worthy of the investment.  If the court finds that FDA can regulate CLIA labs as FDA believes, then FDA needs to establish a policy and timeline that calls for submissions of all tests they declare are not exempt from submission and establish parity for distributed kits.  FDA will also need to add manufacturing requirements and find a way to establish safety and effectiveness clinical data parity between lab created tests that are used in-house and those tests that are distributed as kits.  And while they are at it, FDA will need to address variability limitations among different labs for tests of the same name/intended use, and, of course, increase availability of proficiency testing.  A lot to consider: how much government do we need? Perhaps just considering some enhancement of the CLIA requirements and exclusion of FDA from lab developed tests will be enough.

------------------------------
Glen Freiberg RAC
President, RCQ Consulting
Rancho Santa Fe CA
United States
------------------------------

The introduction and eventual enforcement of the Verifying Accurate, Leading-edge IVCT Development (VALID) Act should help to begin to address most of these concerns and ideally begin to lay the framework for alignment with IVDR as well.

From my limited experience with IVDs, it seems that enforcement discretion was a logical when the IVD regulations were enacted. LDTs were designed and validated within small, high complexity labs to detect rare diseases that could not be addressed through a cleared or approved device. Additionally, accreditation by the CMS, technical expertise and experience of those lab individuals, and final review by pathologists or MDs was sufficient to allow FDA to rely on such a strategy. In the present, the term "LDT" is at times used more expansively and increasingly relies on materials and instruments not necessarily marked for clinical use, additionally advancement and reliance on software further complicates the issue. (Source) 

This creates a situation that could lead to patients undergoing unnecessary medical procedures or not receiving proper treatment. The FDA has "identified problems with several high-risk LDTs including: claims that are not adequately supported with evidence; lack of appropriate controls yielding erroneous results; and falsification of data. The FDA is concerned that people could initiate unnecessary treatment or delay or forego treatment altogether for a health condition, which could result in illness or death. The FDA is aware of faulty LDTs that could have led to: patients being over- or undertreated for heart disease; cancer patients being exposed to inappropriate therapies or not getting effective therapies; incorrect diagnosis of autism; unnecessary antibiotic treatments; and exposure to unnecessary, harmful treatments for certain diseases such as Lyme disease." (Source)

With respect to establishing a quality management system that would allow for laboratories to meet the same (or similar) requirements as those within 21 CFR 820, it would benefit labs to consider certification under the Medical Device Single Audit Program (MDSAP). Doing so addresses the quality systems of participating countries, with ISO 13485 as the backbone of the program. It does require knowledgeable persons to help implement such systems, but the result is a robust and fairly dynamic quality system that can allow a lab to proactively address issues or even prevent them before they affect a patient. Amongst the features identified by the FDA in comments regarding LDTs from labs were:

• A risk-based approach to oversight;
• Risk classification activities;
• Adverse event reporting;
• A robust laboratory quality system.

In order to achieve these objectives, a more robust quality system would be required. In my small amount of experience I have noticed CAP requirements used simply as a checklist activity, rather than as the basis on which to build a more effective quality management system. Requiring a risk based framework around design and development planning, as well as change control and control of processes would allow labs to consider the potential impact before testing.

It does seem that change is coming in some form or another, but it is also an opportunity to help provide focus on the types of testing that would warrant additional scrutiny (using a risk-based approach). Also, given the apparent upheaval caused by companies not recognizing/ignoring the approach of MDR/IVDR in the EU and Health Canada requiring MDSAP certification, it is a chance to align the industry in the same direction as the rest of the international market before things diverge too much.

------------------------------
Christopher Erwin
Scottsdale AZ
United States
------------------------------

Original Message:
Sent: 07-Aug-2019 17:33
From: Glen Freiberg
Subject: CLIA and FDA: Where are we going? August 2019

CLIA and FDA:  Where are we going? 

August 2019

Glen Paul Freiberg, RAC

 

After more than 30 years, I remain upset about the creation and evolution of CLIA 88, the Clinical Laboratory Improvement Amendments of 1988, and commensurate regulations, which revised and expanded the original Clinical Laboratory Improvement Act. However, the reasons for being upset have evolved.  When the law and regulations were first issued, my employer, Boehringer Mannheim (now part of Roche Diagnostics) had point-of-care product lines (capillary blood glucose monitoring), physicians' office test equipment (the ReflotronÒsystem) and large clinical laboratory analyzers, assays and controls.  It was clear from the first reading of the law and regulations that CLIA 88 would kill the physicians' office market and for the wrong reasons.

 

CLIA 88's proposed regulations were promulgated because of quality problems in cytology laboratories, primarily PAP smear testing, reading and interpretation.  

The process at the time was exclusively a manual, microscopic review where many cytology laboratories were operating under poor working and quality monitoring conditions. A valid concern.  But rather than government intervention addressing those manual testing labs, the CLIA update included all tests that were not handled in large traditional clinical labs with just a few exceptions.  At the time, there were no quality or complaint problems with physicians' office equipment like the Reflotron system.  The new lab regulations caused such testing to become cost prohibitive for small physician offices.

 

The result was to deny patients prompt results when a physician deemed such prompt results useful. In addition, the proposed regulations themselves made no sense because the test classification was based on analyte name, demonstrating an outdated government understanding of how point-of-care tests could be performed.   During the initial proposed regulations' comment period, I demonstrated the deficiency at a CLIA meeting in Washington, D.C. by taking a Reflotron system to the meeting. I placed Reflotron system on a table and set three vials on top of the machine showing the CLIA/CDC representatives that one vial contained test strips for glucose, a waived test.  One vial contained identical looking strips for another test, classified as "moderately complex" and a third vial contained identical looking strips for a test classified as "highly complex."  Without plugging the system in, I removed a strip from each vial and showed the group a capillary tube and a capillary pipette explaining that you poke a finger, fill the tube, squirt the sample on the test strip of choice and stick it in the machine.  That's it. Clinical chemistry happens and you get a result.

 

The CLIA/CDC representatives were clearly surprised and while I was not the only person who commented on the proposed regulations, it was clear that test categorization by analyte name would not work.  The regulations were later re-proposed based on the complexity of the steps to perform the test.  Even so, the new laboratory record keeping procedures were so burdensome that the apparent goal was achieved: testing at small physicians' labs mostly went away and the ability to promptly report the results of a thyroid, cholesterol and many other tests with a patient while they were in the office also went away.  To me this regulatory power grab was a travesty as there was never a quality problem with this type of testing and did not need improvement or government intervention with new burdensome laboratory regulations.

 

Concurrently, the large CLIA labs continued and still continue to create their own tests that are not typically regulated by FDA.  However, FDA has been invading the CLIA lab's test development evolution little by little under an internal process called "enforcement discretion."  As a term of art, I disagree with how this term is being used for CLIA labs, it should instead be called "enforcement inconsistency." Where enforcement discretion is appropriate is in howFDA regulates rather than whatthey regulate.  For example, after an FDA inspection when an Inspectional Observations Form, the FDA-483 is issued, choices are made: defer follow up until the next inspection, issue a Warning Letter, or take other regulatory action such as seizure or embargo. This is enforcement discretion.

 

Picking and choosing different activities or services in CLIA labs is an accusation by FDA that the CLIA regulations in place, the certifications received and the competence of the already highly regulated labs and physicians using the data, are either ineffective or insufficient.  Therefore, it's time for a court case to further clarify the limit of FDA jurisdiction and FDA's claim to enforcement discretion on what they choose to regulate, that is, picking and choosing certain CLIA lab created services over others. In other words, I recommend the definition of a device be further clarified to require kit or assay distribution for FDA jurisdiction.

 

With my background in FDA regulatory affairs, many would think I would support FDA efforts in going after select CLIA lab services.  Here is why I do not:  as noted above, my belief is that the FDA jurisdiction is for diagnostic medical devices meeting the definition of the Act that are distributed, not for lab services.   And equally important, FDA cannot keep up with the patient benefitting innovations of the CLIA labs.

 

FDA regulated tests are standardized with set test designs and specifications following the FDA's Quality System manufacturing regulations, 21 CFR 820.  The tests created under CLIA are not distributed medical devices and instead are lab tests that meet CLIA requirements, a separate set of regulations.  And most important, current laboratory science especially in genetic testing, allows for continuous improvements, while the FDA's submission and review process tends to inhibit that.    

 

What attracted my attention to this process have been recent articles on personalized medicine, for example, "Personalizing Cancer" in the August 31, 2019 issue of "Forbes."  Similarly, the opportunity for next-generation sequencing to identify multiple gene issues leading to more effective combination therapies will move too quickly for the FDA processes.  As an example of recent FDA intervention in a CLIA lab, there was a recent Warning letter to a company called Inova Genomics Laboratory.  Here is the reference:

 

https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/inova-genomics-laboratory-577422-04042019

 

It's easy to see why FDA objected to the Inova lab service because the lab was providing information for drug selection and on dosing for drugs' off-label uses.  FDA has claimed the potential for hazards in the company's service without evidence of harm.  While it is accepted that physicians can prescribe drugs off-label and modify dosing, FDA believes that using genetics, even with peer-reviewed publications to back up recommendations, to help with off-label dosing is objectionable.   So, while peer-reviewed publications are typically acceptable for use in supporting an FDA application, using them in this instance apparently is not, since some of the drugs are being used off-label.

 

Inova responded to the Warning Letter by stopping the practices that FDA found objectionable rather than submitting an FDA application or going to court.  Had they submitted an application for just FDA approved drug indications, the potential clearance would have been detrimental to patients the moment a new peer-reviewed publication became available for something not covered by the clearance since the CLIA laboratory could not, under FDA rules, promote this new information.  The point being - how is FDA's interference in the Inova service of benefit to patients and what are the risks outside a challenge to FDA's regulatory process?  Can we not trust the CLIA lab director, CLIA Medical Director and attending physician to use the data appropriately?  Or might this Warning Letter be nothing more than turf protection?  And why is this application of genetic data more important to regulate than all other laboratory created CLIA tests?

 

A few words about the evolution of the use of the FDA "Warning Letter" and as coercion in the Inova case.  When I began work as an FDA Investigator, primary initial inspection enforcement was a "Notice of Adverse Findings Letter" and a "Regulatory Letter."  The former was used to ensure that top management was aware of the FDA-483.  (Sometimes top management was at a location other than the facility being inspected.) A Regulatory Letter meant that FDA had the data and evidence necessary to take the inspected firm to court if FDA did not receive an adequate response to the Regulatory Letter within a specified time frame.   Clearly, having the evidence in place to act and promising to act is both a commitment and a lot of work.  So why generate all the legal preparation work if a threat would be adequate, especially when a Warning letter could achieve change by simple coercion?   The coercion is especially effective if the targeted company is a public company.  Turns out this newer approach has been quite effective, but we need to ask if the approach is being abused regarding CLIA lab services and we probably need to ask through the court system since FDA is not about to change their position on discretion.

 

Therefore, my recommendation is that the next time FDA uses a Warning Letter as coercion in a CLIA lab case, the firm should confront the practice in court and attempt to resolve FDA's jurisdiction limitations and its consistent application of the regulations.  FDA has a terrible record in court so this may be worthy of the investment.  If the court finds that FDA can regulate CLIA labs as FDA believes, then FDA needs to establish a policy and timeline that calls for submissions of all tests they declare are not exempt from submission and establish parity for distributed kits.  FDA will also need to add manufacturing requirements and find a way to establish safety and effectiveness clinical data parity between lab created tests that are used in-house and those tests that are distributed as kits.  And while they are at it, FDA will need to address variability limitations among different labs for tests of the same name/intended use, and, of course, increase availability of proficiency testing.  A lot to consider: how much government do we need? Perhaps just considering some enhancement of the CLIA requirements and exclusion of FDA from lab developed tests will be enough.

------------------------------
Glen Freiberg RAC
President, RCQ Consulting
Rancho Santa Fe CA
United States
------------------------------