Regulatory uncertainty in "home-brew" lab testing

Laboratory-Developed Tests: A Legislative and Regulatory Review

In vitro diagnostics regulation requires continuous adaptation to technologic innovation. Unfortunately, there is a lack of understanding that such a crucial task should be performed efficiently. Europe has waited 23 years for a new regulation!. Anyway, US is under the same trend. Laboratory developed tests were initially regulated 25 years ago and there are still pending issues in the new draft legislation. If you want to know the details, an article in Clinical Chemistry explains the whole issue.

A quarter of a century after the FDA first asserted regulatory authority over LDTs in a draft guidance document, rules and/or guidance regarding LDT oversight have not been implemented. As such, legal questions regarding MDA authority over LDTs and the FDA draft guidance approach have neither been escalated to nor resolved by the judiciary. In addition, many questions central to this debate have not been answered. Are clinical laboratories manufacturers? Should laboratory devices and procedures be regulated similarly? Are there always clear limits between laboratory operations and the practice of laboratory medicine? Any future LDTregulatory or legislative efforts will need to balance and address these concerns if they are to be successful. It is unlikely that interpretation of current statutes and regulations can fully resolve these issues.

 Josep Moscardó, Barcelona landscape

Machine learning for clinical labs

 Machine Learning Takes Laboratory Automation to the Next Level

Good article on ML applications for microbiology lab.

There are two commercially available Food and Drug Administration (FDA)-approved microbiology laboratory automation platforms in the United States, namely, WASPLab (Copan Diagnostics Inc.) and Kiestra (Becton Dickinson) (6). Each system is highly customizable and consists of front-end processing, “smart” incubation according to laboratory protocol, and plate imaging. The processing unit performs medium selection, application of patient information and barcodes for tracking, medium inoculation, and plate streaking. Automation of these processes cuts down on and improves the consistency of repetitive tasks previously performed by technologists.

Image analysis software is not currently FDA approved, so the algorithm it deploys qualifies as a high-complexity laboratory-developed test when used to make definitive calls about microorganism presence/absence or culture significance. In this context, the end user need not understand the internal workings any more than they understand the inner workings of most computers. Additionally, as with most laboratory software, manufacturer assistance is provided in training the algorithm. Labs may, therefore, validate performance according to familiar sensitivity and specificity (for significant growth), precision and accuracy (for quantification), and procedural variation (coefficients of variation, Kappa statistics). As with any test, revalidation must be performed if components of the test change. The number of samples needed to train the algorithm (hundreds to thousands) will be algorithm dependent but easily available due to their common nature, facilitating both initial and revalidation using new plate images. Validation of machine learning image analysis for laboratory automation may, overall, be comparable to that performed for whole-slide imaging as used in histopathology, where the object of validation is a process as much as a machine (12) and where modest interobserver agreement may set a similarly modest benchmark for machine learning performance.

 Eivissa autèntica, Joaquim Gomis

Provant, provant (2)

Laboratory-Developed Tests in the New European Union 2017/746 Regulation: Opportunities and Risks 

 Les proves de laboratori són un ingredient fonamental del diagnòstic clínic. La Unió Europea va canviar la regulació després de 25 anys sense fer-ho i encara cal aplicar-la plenament. Ho hem explicat altres vegades. 

La nova regulació europea ha deixat ombres i àmbits sense donar-hi resposta i aquest és el cas dels que s'anomenen Laboratory Developed Tests o Inhouse Tests. És a dir, proves que no són produïdes a escala industrial. Vegem que diu aquest article:

First, in-house IVDs are exempt from most provisions of the IVDR only if they are not manufactured on an “industrial scale” (1); however, this term is not specified. Process automation, as currently applied in many laboratories, does not qualify as production on an industrial scale in our opinion.

Arguably the most problematic point of the IVDR with respect to in-house diagnostic devices is subparagraph (d) of Article 5 (5) IVDR. The section outlines that in-house diagnostic devices may only be used when the target patient group’s specific needs cannot be met by an equivalent marketed product. However, what if a commercial IVD exists and an in-house test differs in aspects of performance? Then, the entire section hinges on the interpretation of the term “equivalent” and “the target patient group’s specific needs.”

 I després s'apunta a la responsabilitat de cada estat en la qüestió. Sabent que això pot afectar més del que ens imaginem, caldria una solució clara. Un altre article recent diu que pot afectar a fins el 50% de les proves diagnòstiques, em sembla molt i deuen parlar de laboratoris molt específics i proposa l'aplicació de la ISO 15189.  En qualsevol cas, tenim més incertesa, i més ombres que llum a la regulació europea de diagnòstic in vitro que algú haurà de corregir ben aviat.

Jordi Bernadó a la Galeria Senda

Enough is enough

If there is a grey area in medical devices and services regulation, this is the Laboratory Developed Tests one. Up to now, FDA has refused to define the rules of the game for 11,000 diagnostic tests performed at 2,000 labs in USA. This means that no official or external reviewer has analysed the clinical validity and clinical utility as it is done in any reagent and instrument. I can't understand why we have arrived at such a situation.
Fortunately NYT reports that on July 31st, FDA announced that this will change.

The agency said on Thursday that such discretion must end because circumstances had changed. Lab-developed tests once were fairly simple, often developed by a hospital for tests on its own patients. Now the tests can be complex and are being developed by  companies and marketed widely.

Some widely used commercial tests have never had to be reviewed by the agency. These include Myriad Genetics’ breast cancer risk test, the subject of a Supreme Court patent decision last year; the Oncotype DX test from Genomic Health, which is used to determine if women with early-stage breast cancer need chemotherapy; and noninvasive prenatal tests for Down syndrome that are rapidly catching on.

In this blog I have supported several times for a clear regulation of these tests . Just the other day when looking at the statements of FDA commissioner, I was astonished:

Just as drugs need to be safe and effective for treating diseases, medical devices used to help diagnose disease and direct therapy also need to be safe and effective, Faulty test results could lead patients to seek unnecessary treatment or to delay or to forgo treatment altogether.

These statement raise more concerns about what US regulator has done after all these years. And european regulation is still worse in this sense. I have explained such disaster previously and up to now there is no news. Some times I wonder why do we pay taxes, why do we have to be part of Europe. Enough is enough.

Validesa i utilitat de les proves genòmiques

Genome-Based Diagnostics: Clarifying Pathways to Clinical Use: Workshop Summary

En teoria de jocs es diu que tenim un punt focal quan trobem una solució que la gent tendeix a utilitzar per coordinar-se quan falta comunicació, perquè sembla que satisfà tothom. El concepte va ser introduït per l'economista guanyador del Premi Nobel Thomas Schelling en el llibre The Strategy of Conflict (1960). El que aquest punt focal o equilibri sigui eficient o no ho sigui ja són figues d'un altre paner..
Tot d'una, quan llegia un l'informe de l'IOM sobre tests genètics, m'ha vingut a la memòria Schelling i els seus punts focals. Resulta que s'explica com davant dels biomarcadors s'ha encetat un cercle viciós (punt focal) que només es pot trencar mitjançant un canvi regulatori i d'incentius. Afegeixo un paràgraf d'interès:

The basic problem is that there has been relatively little consistency regarding which biomarkers have been introduced into clinical practice. Very few cancer biomarkers with demonstrated clinical utility have been introduced over the past 30 years. Even among those tests that have been integrated into practice, their use in certain settings has not always been supported by evidence of benefit, such as the use of prostate-specific antigen (PSA) as a screening test (Andriole et al., 2009), said Hayes. This has helped to create what Hayes has termed a “vicious cycle” in which tumor biomarkers are systematically undervalued (Figure 2-1). This undervaluation has led to limited use of these diagnostics by health care providers and poor reimbursement when a marker has been able to navigate the regulatory environment to be brought to market. Lack of use and reimbursement in turn leads to limited funding for biomarker research because the return on investment is low. The perception that markers have little utility has also led to an environment of lower academic  recognition for developing biomarker-based tests. The overall result is reduced ability and incentive to conduct properly designed clinical trials to generate high-quality evidence of clinical utility. In return, there is reduced data certainty, higher skepticism, and few recommendations for clinical use, said Hayes, which completes the cycle by contributing to the poor valuation of marker utility. Hayes focused his recommendations for breaking the “vicious cycle” of undervalued tumor biomarkers on two areas: the regulatory environment and marker reimbursement.

Actualment als USA (i aquí encara menys) no hi ha un procés de revisió de la FDA per a les proves diagnòstiques de laboratori del tipus Laboratory Developed Tests LDT on s'avaluï la validesa analítica, validesa clínica i utilitat clínica. Es regula per la llei CLIA tal com s'explica al text. La proposta és doncs que la FDA prengui part del procés de revisió i es reformuli la regulació existent. I des de la vessant dels incentius, se suggereix que s'estableixin anàlisis cost-efectivitat de les proves que permetin situar el seu preu en funció del valor que aporten.
Ens trobem doncs en un llibre clau per a un moment clau. I qui tingui ulls i vulgui que el llegeixi, són tant sols 105 pàgines fonamentals per entendre una de les qüestions determinants de la medicina del futur.
Em costa admetre que el punt focal per aquí aprop es redueixi a veure passar els dies inexorablement i la innovació tecnològica resti sense avaluar, un equilibri ineficient. En Schelling diria que cal comprometre's de forma creïble per sortir-ne, però per ara i pel que fa al regulador, no ho sé veure per enlloc.

PS. Miss-selling drugs, a The Economist.

PS. La sindicatura emet informe sobre l'Hospital Clínic. Enmig del desori observo que s'han deixat de cobrar 40 milions d'euros amb trasplantaments a forasters!

PS. A DM trobareu alguns detalls sobre el nou sistema de pagament.  Esperem més informació en el futur.

PS. Les autopistes sense cotxes ens constaran 290 milions el 2012, l'any passat van costar 80 milions.

MABS in history of medicine

The Lock and Key of Medicine Monoclonal Antibodies and the Transformation of Healthcare

While reading FT this summer I came across an article quoting a unique book on history of monoclonal antibodies (MABS). Right now there are more than 30 drugs in the market based on hybridoma technology that was created in 1975.
The birth of MABS is explained with full details, how the creators finally didn't patented it and why, the difficulties for research in an unconnected world, etc... An exciting story that is worth reading. Right now, it would be completely different, commercialization of research and medicine has raised considerably.

That a British company spearheaded the first marketing of Mabs, a technology devised in a British laboratory by an émigré Argentinian scientist with his German colleague, highlights the international nature of biotechnology commercialization. Sera- Lab’s venture to sell Mabs took place in the midst of the excitement generated by the founding of Genentech in 1976. The emergence of Genentech, which had been set up
to market recombinant DNA products, galvanized numerous alliances among academics, entrepreneurs, and venture capitalists to launch new companies to commercialize biotechnology. Most of the early enterprises set up in the wake of Genentech’s birth were dedicated to exploiting recombinant DNA for the mass production of natural products such as interferon and insulin for drugs. But the early germination of the modern biotechnology industry did not rest solely on recombinant DNA. By the 1970s a number of pioneering companies were developing Mab products, including Sera- Lab and two startups: Hybritech in San Diego and Centocor in Philadelphia. Entrepreneurs who risked entry into the field had no guarantee of success and were entering totally uncharted
territory. Such individuals faced major fi nancial, personal, professional, and regulatory challenges as well as a great deal of hostility, pessimism, and litigation.

Milstein with Köhler at the time of their receiving the Nobel Prize in 1984 together with Nils Jerne.

Mabs have had their strongest therapeutic impact in the field of cancer. The first Mab to reach the market for cancer was edrecolomab (Panorex), which was granted German regulatory approval in 1995 for the treatment of postoperative colorectal cancer. Developed by Centocor in partnership with the Wistar Institute, it was withdrawn in 2001 because of its poor effi cacy in comparison with other drugs. Since 1997, however, the U.S. Food and Drug Administration (FDA) has approved twelve Mab drugs for cancer treatment, including rituximab (Rituxan), approved in 1998 for the treatment of non- Hodgkin’s lymphoma. By 2012 there were over 160 candidates in clinical trials for cancer, with seventy of them in phase III trials, the stage before a drug is submitted for regulatory approval.

Mabs have enabled the identification and characterization of cancerous tumors previously difficult to detect and diff erentiate from other tumors, thereby providing a better understanding of cancer. They have also opened a path to more personalized medical treatment. Trastuzumab (Herceptin), for example, was specifically developed to target HER2/neu, a protein overexpressed by tumors found in 25 percent of newly diagnosed breast- cancer patients