MétaCan
Menu
Back to cohort
Record W3016631762 · doi:10.1016/j.jtho.2020.03.029

Canadian Multicenter Project on Standardization of Programmed Death-Ligand 1 Immunohistochemistry 22C3 Laboratory-Developed Tests for Pembrolizumab Therapy in NSCLC

2020· article· en· W3016631762 on OpenAlexafffundabout
Emina Torlakovic, Roula Albadine, Gilbert Bigras, Alexander H. Boag, Anna Bojarski, Michael Cabanero, Sophie Camilleri‐Broët, Carol C. Cheung, Christian Couture, Kenneth J. Craddock, Jean‐Claude Cutz, Prashant Dhamanaskar, Pierre Fiset, Mohammad Hossain, David Hwang, Diana N. Ionescu, Doha Itani, Margaret M. Kelly, Keith Kwan, Hyun J. Lim, Søren Nielsen, Gefei Qing, Harman Sekhon, Alan Spatz, Ranjit Waghray, Hangjun Wang, Zhaolin Xu, Ming‐Sound Tsao

Bibliographic record

VenueJournal of Thoracic Oncology · 2020
Typearticle
Languageen
FieldMedicine
TopicCancer Immunotherapy and Biomarkers
Canadian institutionsNova Scotia Health AuthorityJewish General HospitalDalhousie UniversityUniversity of OttawaOttawa HospitalUniversity of ManitobaMcGill UniversityLondon Health Sciences CentreSaint John Regional HospitalUniversity of British ColumbiaSunnybrook Health Science CentreMcMaster UniversityManitoba HealthCredit Valley HospitalSouthlake Regional Health CenterTrillium Health CentreSt. Joseph’s Healthcare HamiltonInstitut universitaire de cardiologie et de pneumologie de QuébecMcGill University Health CentreRoyal University HospitalUniversity of TorontoUniversity of CalgaryHealth Sciences CentreUniversity of AlbertaUniversity Health NetworkHealth Sciences NorthQueen's UniversityKingston General HospitalCentre Hospitalier de l’Université de MontréalSaskatchewan Health AuthorityUniversité LavalUniversity of Saskatchewan
FundersEMD SeronoPfizer CanadaMerckMerck CanadaAstraZeneca CanadaAstraZenecaPfizerBristol-Myers Squibb
KeywordsPembrolizumabMedicineStandardizationImmunohistochemistryOncologyInternal medicineImmunotherapyCancer

Abstract

fetched live from OpenAlex

INTRODUCTION: The programmed death-ligand 1 (PD-L1) immunohistochemistry (IHC) assay is used to select patients for first or second-line pembrolizumab monotherapy in NSCLC. The PD-L1 IHC 22C3 pharmDx assay requires an Autostainer Link 48 instrument. Laboratories without this stainer have the option to develop a highly accurate 22C3 IHC laboratory-developed test (LDT) on other instruments. The Canadian 22C3 IHC LDT validation project was initiated to harmonize the quality of PD-L1 22C3 IHC LDT protocols across 20 Canadian pathology laboratories. METHODS: Centrally optimized 22C3 LDT protocols were distributed to participating laboratories. The LDT results were assessed against results using reference PD-L1 IHC 22C3 pharmDx. Analytical sensitivity and specificity were assessed using cell lines with varying PD-L1 expression levels (phase 1) and IHC critical assay performance controls (phase 2B). Diagnostic sensitivity and specificity were assessed using whole sections of 50 NSCLC cases (phase 2A) and tissue microarrays with an additional 50 NSCLC cases (phase 2C). RESULTS: In phase 1, 80% of participants reached acceptance criteria for analytical performance in the first attempt with disseminated protocols. However, in phase 2A, only 40% of participants reached the desired diagnostic accuracy for both 1% and 50% tumor proportion score cutoff. In phase 2B, further protocol modifications were conducted, which increased the number of successful laboratories to 75% in phase 2C. CONCLUSIONS: It is possible to harmonize highly accurate 22C3 LDTs for both 1% and 50% tumor proportion score in NSCLC across many laboratories with different platforms. However, despite a centralized approach, diagnostic validation of predictive IHC LDTs can be challenging and not always successful.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.065
metaresearch head score (Gemma)0.024
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.842
Threshold uncertainty score0.341

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0650.024
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.004
Science and technology studies0.0040.002
Scholarly communication0.0020.001
Open science0.0050.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.044
GPT teacher head0.423
Teacher spread0.379 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations21
Published2020
Admission routes3
Has abstractyes

Explore more

Same venueJournal of Thoracic OncologySame topicCancer Immunotherapy and BiomarkersFrench-language works237,207