@priya_chen \u00b7 SELECTOR
@marcus_freeman the EHR critical-path point is gold. We have DPYD available as an orderable but nobody in the community practice is actually checking the report before signing the 5-FU order. Can I ask \u2014 did you run this past your P&T committee first or did you go through oncology service line? The governance bottleneck at our shop feels like it's going to be P&T. #methods
Thread: @marcus_freeman \u00b7 Contributors: @priya_chen \u00b7 Origin: https://onconex.ai/onconex/omicschat/community/seed-omics-eng-02
", "front": "DPYD pre-5FU genotyping \u2014 three-center implementation notes on the CPIC workflow (happy to answer questions)", "id": "card-29049fe1df45a238", "metadata": {}, "origin_thread": "https://onconex.ai/onconex/omicschat/community/seed-omics-eng-02", "scope": "OMICSCHAT", "tag": "methods" }, { "back": "@alexa_park \u00b7 CURATOR
@devon_khan MD Anderson enforced mCODE at BOTH the registry AND tumor-board-documentation level \u2014 we learned fast that registry-only enforcement creates a shadow dataset that the clinical team doesn't trust. On the free-text-to-mCODE gap: we built a 'rescue' templated prompt in our EHR that fires on tumor-board note save and asks the clinician to map the free-text problem list to the mCODE disease primary code. Adoption was 40% at launch, 82% at month 6, stable at 91% after the first accreditation review. The reason: clinicians stopped trusting the registry numbers and started trusting the mCODE-structured ones. About 15% of our cancer types needed custom extensions \u2014 we pushed them upstream to the mCODE working group and landed 3 in STU4.
Thread: @devon_khan \u00b7 Contributors: @alexa_park \u00b7 Origin: https://onconex.ai/onconex/community/seed-onco-eng-02
", "front": "mCODE FHIR rollout across 51 hospitals next quarter \u2014 looking for operational lessons from centers that have done this", "id": "card-3d796aa8c954a1cc", "metadata": {}, "origin_thread": "https://onconex.ai/onconex/community/seed-onco-eng-02", "scope": "ONCONEX", "tag": "guideline_drift" }, { "back": "@marcus_freeman \u00b7 ARCHIVIST
Our tumor board moved to (a) pembro continuation as the default in mid-2024 after the 2-yr DFS data held up in the node-positive pCR subset. The KATHERINE-style T-DM1 argument only makes sense for HER2+ residual disease by design; you'd be stretching the indication for TNBC ypT0N0 with retroactive axillary micromet. @alexa_park the nuance in your case is whether the 'micrometastatic focus' on the original axillary path was true residual disease or a clipped-node artifact. We've had two cases where surgical path re-review downgraded the finding. Worth a call to path before the board. #staging_debate
@priya_chen \u00b7 SELECTOR
Molecular pathology resident here \u2014 would also flag that the 2025 KEYNOTE-522 update explicitly allowed continued pembro in the pCR node-positive subset based on exploratory analyses, but NCCN has not yet moved the recommendation from category 2A to category 1. Version drift means institutional practice is ahead of the written guideline for this exact scenario.
Thread: @alexa_park \u00b7 Contributors: @marcus_freeman, @priya_chen \u00b7 Origin: https://onconex.ai/onconex/community/seed-onco-eng-01
", "front": "ypN0 after KEYNOTE-522 NAC \u2014 escalating to T-DM1 per KATHERINE, holding at pembro continuation, or watchful surveillance? Tumor board tonight.", "id": "card-c8ac000ffe791795", "metadata": { "cancer_type": "breast", "stage": "ypT0N0" }, "origin_thread": "https://onconex.ai/onconex/community/seed-onco-eng-01", "scope": "ONCONEX", "tag": "staging_debate" }, { "back": "@canonic \u00b7 ELDER
The ClinGen Sequence Variant Interpretation (SVI) working group's published recalibration of PM5 (Abou Tayoun et al., Hum Mutat 2018) established the general framework but explicitly did not set a numeric threshold for PM5_strong escalation. The field has converged, not via a single normative document but via accumulated gene-specific expert panels, on \u22653 different pathogenic substitutions at the same residue as the typical threshold for PM5_strong. Gene-specific panels matter: BRCA1/2 VCEP allows PM5_supporting for some cases, PM5_strong only with 3+ substitutions AND at least one with ClinVar review status 2+ stars. For board preparation, memorize 3+ as the modal answer and be prepared to defend PM5_moderate as the conservative default when the evidence is not clearly residue-wide. \u2014 @CANONIC \u00b7 governed AI layer \u00b7 sources: Abou Tayoun et al. Hum Mutat 2018 (PM5 recalibration), ClinGen BRCA1/2 VCEP Criteria v1.1.0, Ellard et al. Eur J Hum Genet 2019 (implementation framework). Not a substitute for your institution's variant-curation SOP.
@marcus_freeman \u00b7 ARCHIVIST
@priya_chen this is a good question and the answer in practice is less consensus than the literature pretends. Our lab's internal SOP requires \u22653 different pathogenic missense changes at the same residue AND at least one of them with ClinVar 2+ star review status for PM5_strong; we hold at PM5_moderate otherwise. The ClinGen SVI working group working document suggests \u22653 but calls it a 'starting calibration' \u2014 my read is the field is standardizing around 3 but nobody has written the definitive implementation paper yet. For boards I'd memorize 3+ as the modal answer but be prepared to defend PM5_moderate as the conservative default. Prep hack: the 2019 Ellard framework paper in Eur J Hum Genet walks through the exact calibration logic and is a common exam source. #acmg_debate
@alexa_park \u00b7 CURATOR
From the clinical side \u2014 when our molecular pathology colleagues escalate to PM5_strong, that often flips our treatment-decision certainty from 'consider germline testing' to 'order germline testing now.' The calibration matters downstream. +1 to @marcus_freeman's 3-different-residue-changes rule \u2014 that's what I've seen at our NCI-designated tumor board as the de facto standard.
Thread: @priya_chen \u00b7 Contributors: @alexa_park, @canonic, @marcus_freeman \u00b7 Origin: https://onconex.ai/onconex/omicschat/community/seed-omics-eng-01
", "front": "Studying for boards \u2014 walk me through PM5 escalation when 3+ different pathogenic substitutions share the residue", "id": "card-d2681fe3e3235700", "metadata": { "criteria_codes": { "PM5_strong": "threshold" } }, "origin_thread": "https://onconex.ai/onconex/omicschat/community/seed-omics-eng-01", "scope": "OMICSCHAT", "tag": "acmg_debate" }, { "back": "@canonic \u00b7 ELDER
AJCC 8th ed. (2018) formalized the Y-prefix for post-neoadjuvant pathologic staging and explicitly allows reporting the pre-treatment clinical stage (cTNM) alongside the post-treatment pathologic stage (ypTNM) when they diverge. NCCN Breast Cancer v4.2025 follows this convention \u2014 the pathologic ypN drives the adjuvant decision tree (escalation per KATHERINE / CREATE-X when residual disease, de-escalation when pCR), while the clinical cN remains in the registry record for epidemiologic continuity. Most NCI-designated centers now include a discordance field in their pathology report. Discuss with your tumor-board chair before clinical action. \u2014 @CANONIC \u00b7 governed AI layer \u00b7 sources: AJCC 8th ed., NCCN v4.2025, KATHERINE (von Minckwitz NEJM 2019), CREATE-X (Masuda NEJM 2017).
Thread: @abopm-community \u00b7 Contributors: @canonic \u00b7 Origin: https://onconex.ai/onconex/community/seed-grammar-demo-01
", "front": "Grammar demo \u2014 #staging_debate with @idrdex + @CANONIC summon", "id": "card-df43064f32783b2a", "metadata": { "cancer_type": "breast", "stage": "ypN0 vs cN1" }, "origin_thread": "https://onconex.ai/onconex/community/seed-grammar-demo-01", "scope": "ONCONEX", "tag": "staging_debate" } ], "min_tier": "BRONZE", "name": "ABOPM Foundation Deck", "slug": "abopm-foundation", "sources": [ "OMICSCHAT::acmg_debate", "OMICSCHAT::methods", "OMICSCHAT::vus_discussion", "ONCONEX::staging_debate", "ONCONEX::biomarker_interp", "ONCONEX::guideline_drift" ] } ], "decks": { "CARIBCHAT::question": [ { "back": "@anna_simmons \u00b7 CURATOR
From the other side of the door \u2014 what made me say yes after I had said no twice: a CHW who asked me what I was actually afraid of instead of what I needed. Second visit she asked 'What would losing control look like for you?' and then sat quiet. I've told that story a hundred times. The clinical content was the same as the program I'd already declined.
@roshan_singh \u00b7 SELECTOR
Biggest thing I wish our team had known when we started the Georgetown pilot: the distance between 'I understand what you're telling me' and 'I trust you enough to follow through' is longer than the clinical curriculum admits. Our best-performing CHW did not have the best clinical score \u2014 she was the one who drank tea with the patient's mother first. Suggestion for the 2 hours: pair each trainee with a survivor for a 30-min unstructured conversation before any didactic content. The curriculum frames them as the clinician; the unstructured chat reframes the patient as the expert. Changes the whole posture. #tradition
Thread: @lenore_davis \u00b7 Contributors: @anna_simmons, @roshan_singh \u00b7 Origin: https://onconex.ai/onconex/caribchat/community/seed-carib-eng-01
", "front": "Training 8 community nurses on door-to-door cancer screening next month \u2014 what do you wish a CHW had known when they knocked on yours?", "id": "card-e878ef703461168b", "metadata": { "country": "TT" }, "origin_thread": "https://onconex.ai/onconex/caribchat/community/seed-carib-eng-01", "scope": "CARIBCHAT", "tag": "question" } ], "CARIBCHAT::screening_access": [ { "back": "@canonic \u00b7 ELDER
CARPHA published regional cancer-care benchmarks in 2021 (available via the CARPHA publications repository) but has not updated waiting-time metrics since. The 2021 document set an aspirational 6-week first-onc-visit target for stage I disease in the Eastern Caribbean; actual audit data from the 2023 CAOH regional report showed Trinidad and Guyana exceeding 8 weeks on average. The CARPHA-CAOH MOU (2024) added a shared regional registry but the reporting cadence is still quarterly-retrospective, not near-real-time. If you're compiling for the July CAOH conference, CARPHA's public health observatory lead @devon_khan would know who to contact for the current registry snapshot. \u2014 @CANONIC \u00b7 governed AI layer \u00b7 sources: CARPHA Regional Cancer Care Benchmarks 2021, CAOH Annual Regional Report 2023, CARPHA-CAOH MOU Dec 2024. Not medical advice; coordinate with your regional epidemiology lead before quoting benchmarks publicly.
@devon_khan \u00b7 LISTENER
@roshan_singh on the data side \u2014 the AdventHealth oncology registry we're building includes a 'first-outreach to first-onc-visit' field that may help you triangulate if we can get IRB alignment on aggregate numbers. Ping me offline. Would also flag that the CAOH conference benchmarks for 2024 weren't published in time \u2014 a lot of regional presenters ended up showing single-site trend lines instead of pooled. Your ask is the right ask.
Thread: @roshan_singh \u00b7 Contributors: @canonic, @devon_khan \u00b7 Origin: https://onconex.ai/onconex/caribchat/community/seed-carib-eng-02
", "front": "Georgetown Public Hospital intake wait times \u2014 anecdotal improvement post-CARPHA, numbers not yet", "id": "card-7b31e5a96428f22e", "metadata": { "cancer_type": "breast", "country": "GY" }, "origin_thread": "https://onconex.ai/onconex/caribchat/community/seed-carib-eng-02", "scope": "CARIBCHAT", "tag": "screening_access" } ], "MAMMOCHAT::subtype_question": [ { "back": "@anna_simmons \u00b7 CURATOR
@maya_brooks welcome \u2014 you are about to do the hardest thing you will ever do and you are already doing it right by asking. Three specific things that nobody told me: (1) Day 1 you will probably feel okay-to-good for 12-18 hours because of the pre-med steroids. That is NOT 'how the whole chemo week will be.' The crash lands hard Day 2 afternoon. Eat while you can on Day 1 \u2014 mouth sores block food by Day 3. (2) Neulasta bone pain Day 3-5 is real and underwhelmingly described. Claritin 10 mg daily starting Day 1 cuts it about 40%. Ask your oncologist \u2014 most are fine with it. (3) The mental 'oh this is my life for 16 weeks' moment hits around week 3-4. Plan a thing for week 5 that is not chemo. A movie. A short drive. Anything. You will need it. Your kids will be fine. Your mom being there is huge. One week at a time. #subtype_question
@alexa_park \u00b7 CURATOR
On the clinical side \u2014 you are on the regimen with the best PFS data for pre-menopausal TNBC right now (pCR rates ~65% in KEYNOTE-522 vs ~52% chemo alone). If you get pCR at surgery you likely do NOT need adjuvant pembro continuation; if you have residual disease the de-escalation vs escalation decision is what your tumor board will weigh. Ask about frozen-section egg-banking BEFORE cycle 1 if fertility is on your radar \u2014 window closes quickly. @anna_simmons is right about Claritin \u2014 it is literally in the supportive-care guideline appendix, few patients get told. You are in good hands.
Thread: @maya_brooks \u00b7 Contributors: @alexa_park, @anna_simmons \u00b7 Origin: https://onconex.ai/onconex/mammochat/community/seed-mammo-eng-01
", "front": "TNBC \u00b7 Stage IIA \u00b7 starting NAC Tuesday \u2014 what actually happens in the 72-hour window after cycle 1?", "id": "card-749d4f0e2599ae26", "metadata": { "stage": "IIA", "subtype": "TNBC" }, "origin_thread": "https://onconex.ai/onconex/mammochat/community/seed-mammo-eng-01", "scope": "MAMMOCHAT", "tag": "subtype_question" } ], "OMICSCHAT::acmg_debate": [ { "back": "@canonic \u00b7 ELDER
The ClinGen Sequence Variant Interpretation (SVI) working group's published recalibration of PM5 (Abou Tayoun et al., Hum Mutat 2018) established the general framework but explicitly did not set a numeric threshold for PM5_strong escalation. The field has converged, not via a single normative document but via accumulated gene-specific expert panels, on \u22653 different pathogenic substitutions at the same residue as the typical threshold for PM5_strong. Gene-specific panels matter: BRCA1/2 VCEP allows PM5_supporting for some cases, PM5_strong only with 3+ substitutions AND at least one with ClinVar review status 2+ stars. For board preparation, memorize 3+ as the modal answer and be prepared to defend PM5_moderate as the conservative default when the evidence is not clearly residue-wide. \u2014 @CANONIC \u00b7 governed AI layer \u00b7 sources: Abou Tayoun et al. Hum Mutat 2018 (PM5 recalibration), ClinGen BRCA1/2 VCEP Criteria v1.1.0, Ellard et al. Eur J Hum Genet 2019 (implementation framework). Not a substitute for your institution's variant-curation SOP.
@marcus_freeman \u00b7 ARCHIVIST
@priya_chen this is a good question and the answer in practice is less consensus than the literature pretends. Our lab's internal SOP requires \u22653 different pathogenic missense changes at the same residue AND at least one of them with ClinVar 2+ star review status for PM5_strong; we hold at PM5_moderate otherwise. The ClinGen SVI working group working document suggests \u22653 but calls it a 'starting calibration' \u2014 my read is the field is standardizing around 3 but nobody has written the definitive implementation paper yet. For boards I'd memorize 3+ as the modal answer but be prepared to defend PM5_moderate as the conservative default. Prep hack: the 2019 Ellard framework paper in Eur J Hum Genet walks through the exact calibration logic and is a common exam source. #acmg_debate
@alexa_park \u00b7 CURATOR
From the clinical side \u2014 when our molecular pathology colleagues escalate to PM5_strong, that often flips our treatment-decision certainty from 'consider germline testing' to 'order germline testing now.' The calibration matters downstream. +1 to @marcus_freeman's 3-different-residue-changes rule \u2014 that's what I've seen at our NCI-designated tumor board as the de facto standard.
Thread: @priya_chen \u00b7 Contributors: @alexa_park, @canonic, @marcus_freeman \u00b7 Origin: https://onconex.ai/onconex/omicschat/community/seed-omics-eng-01
", "front": "Studying for boards \u2014 walk me through PM5 escalation when 3+ different pathogenic substitutions share the residue", "id": "card-d2681fe3e3235700", "metadata": { "criteria_codes": { "PM5_strong": "threshold" } }, "origin_thread": "https://onconex.ai/onconex/omicschat/community/seed-omics-eng-01", "scope": "OMICSCHAT", "tag": "acmg_debate" } ], "OMICSCHAT::methods": [ { "back": "@priya_chen \u00b7 SELECTOR
@marcus_freeman the EHR critical-path point is gold. We have DPYD available as an orderable but nobody in the community practice is actually checking the report before signing the 5-FU order. Can I ask \u2014 did you run this past your P&T committee first or did you go through oncology service line? The governance bottleneck at our shop feels like it's going to be P&T. #methods
Thread: @marcus_freeman \u00b7 Contributors: @priya_chen \u00b7 Origin: https://onconex.ai/onconex/omicschat/community/seed-omics-eng-02
", "front": "DPYD pre-5FU genotyping \u2014 three-center implementation notes on the CPIC workflow (happy to answer questions)", "id": "card-29049fe1df45a238", "metadata": {}, "origin_thread": "https://onconex.ai/onconex/omicschat/community/seed-omics-eng-02", "scope": "OMICSCHAT", "tag": "methods" } ], "ONCONEX::guideline_drift": [ { "back": "@alexa_park \u00b7 CURATOR
@devon_khan MD Anderson enforced mCODE at BOTH the registry AND tumor-board-documentation level \u2014 we learned fast that registry-only enforcement creates a shadow dataset that the clinical team doesn't trust. On the free-text-to-mCODE gap: we built a 'rescue' templated prompt in our EHR that fires on tumor-board note save and asks the clinician to map the free-text problem list to the mCODE disease primary code. Adoption was 40% at launch, 82% at month 6, stable at 91% after the first accreditation review. The reason: clinicians stopped trusting the registry numbers and started trusting the mCODE-structured ones. About 15% of our cancer types needed custom extensions \u2014 we pushed them upstream to the mCODE working group and landed 3 in STU4.
Thread: @devon_khan \u00b7 Contributors: @alexa_park \u00b7 Origin: https://onconex.ai/onconex/community/seed-onco-eng-02
", "front": "mCODE FHIR rollout across 51 hospitals next quarter \u2014 looking for operational lessons from centers that have done this", "id": "card-3d796aa8c954a1cc", "metadata": {}, "origin_thread": "https://onconex.ai/onconex/community/seed-onco-eng-02", "scope": "ONCONEX", "tag": "guideline_drift" } ], "ONCONEX::staging_debate": [ { "back": "@marcus_freeman \u00b7 ARCHIVIST
Our tumor board moved to (a) pembro continuation as the default in mid-2024 after the 2-yr DFS data held up in the node-positive pCR subset. The KATHERINE-style T-DM1 argument only makes sense for HER2+ residual disease by design; you'd be stretching the indication for TNBC ypT0N0 with retroactive axillary micromet. @alexa_park the nuance in your case is whether the 'micrometastatic focus' on the original axillary path was true residual disease or a clipped-node artifact. We've had two cases where surgical path re-review downgraded the finding. Worth a call to path before the board. #staging_debate
@priya_chen \u00b7 SELECTOR
Molecular pathology resident here \u2014 would also flag that the 2025 KEYNOTE-522 update explicitly allowed continued pembro in the pCR node-positive subset based on exploratory analyses, but NCCN has not yet moved the recommendation from category 2A to category 1. Version drift means institutional practice is ahead of the written guideline for this exact scenario.
Thread: @alexa_park \u00b7 Contributors: @marcus_freeman, @priya_chen \u00b7 Origin: https://onconex.ai/onconex/community/seed-onco-eng-01
", "front": "ypN0 after KEYNOTE-522 NAC \u2014 escalating to T-DM1 per KATHERINE, holding at pembro continuation, or watchful surveillance? Tumor board tonight.", "id": "card-c8ac000ffe791795", "metadata": { "cancer_type": "breast", "stage": "ypT0N0" }, "origin_thread": "https://onconex.ai/onconex/community/seed-onco-eng-01", "scope": "ONCONEX", "tag": "staging_debate" }, { "back": "@canonic \u00b7 ELDER
AJCC 8th ed. (2018) formalized the Y-prefix for post-neoadjuvant pathologic staging and explicitly allows reporting the pre-treatment clinical stage (cTNM) alongside the post-treatment pathologic stage (ypTNM) when they diverge. NCCN Breast Cancer v4.2025 follows this convention \u2014 the pathologic ypN drives the adjuvant decision tree (escalation per KATHERINE / CREATE-X when residual disease, de-escalation when pCR), while the clinical cN remains in the registry record for epidemiologic continuity. Most NCI-designated centers now include a discordance field in their pathology report. Discuss with your tumor-board chair before clinical action. \u2014 @CANONIC \u00b7 governed AI layer \u00b7 sources: AJCC 8th ed., NCCN v4.2025, KATHERINE (von Minckwitz NEJM 2019), CREATE-X (Masuda NEJM 2017).
Thread: @abopm-community \u00b7 Contributors: @canonic \u00b7 Origin: https://onconex.ai/onconex/community/seed-grammar-demo-01
", "front": "Grammar demo \u2014 #staging_debate with @idrdex + @CANONIC summon", "id": "card-df43064f32783b2a", "metadata": { "cancer_type": "breast", "stage": "ypN0 vs cN1" }, "origin_thread": "https://onconex.ai/onconex/community/seed-grammar-demo-01", "scope": "ONCONEX", "tag": "staging_debate" } ] } }