Colorectal cancer - comprehensive genomic profile

Synthetic data · demonstration only

“Josephine Stellar” is a fictional patient. Every clinical event, laboratory value, specimen, document and molecular result in this report is invented for demonstration. No real patient data appears here, and no part of this document should be used to inform the care of any person. The analytical pipeline, evidence sources and drug labels referenced are the production ones.

Patient

Josephine Stellar

Date of birth

12 Apr 1979

Age / Sex

47 y · female

Diagnosis

Colorectal cancer

Specimen

tissue · colon (sigmoid)

Tumour purity

70%

Report date

18 May 2026

Ordering clinician

Dr. Elena Marsh

Assay FoundationOne CDx · Testing lab Foundation Medicine · Testing date 18 May 2026

Drivers

Alterations

FDA-matched treatments

Trials matched

Section 1 Clinician synthesis 3 drivers identified · 12 options reviewed · 4 prioritised by the board

Molecular tumour board roadmap

Treatment horizon

First line

Pembrolizumab (Keytruda)

On progression

Sotorasib + Panitumumab

Later line

Regorafenib (Stivarga)

Throughout

Track emerging resistance clones

Backbone runner FOLFIRI available at any line

What to do today

Start now Pembrolizumab (Keytruda) Standard of careMICROSATELLITE INSTABILITY

A PD-1-blocking antibody with tumour-agnostic approvals in MSI-H/dMMR and TMB-high disease. FDA-approved for microsatellite instability -high (MSI-H) or mismatch repair deficient (dMMR) cancer that has spread to other parts of the body or cannot be removed by surgery.

Confirm before starting: MSI-High currently rests on sequencing alone. Mismatch-repair immunohistochemistry (MLH1, PMS2, MSH2, MSH6) was requested on block S26-041992B on 4 May and is still pending — expect loss of MSH2/MSH6, concordant with the MSH2 p.A636P allele. Chase this result before the first dose; see Testing. Baseline immune-toxicity bloods are listed there as well.

Equivalent alternative: Dostarlimab-gxly (Jemperli) — tumour-agnostic dMMR approval. Not equivalent: durvalumab (Imfinzi) is approved for dMMR endometrial cancer; use here would be off-label and is not recommended by this board.

On progression Sotorasib (Lumakras) + Panitumumab (Vectibix) Standard of careKRAS

KRAS G12C inhibitor combined with an anti-EGFR antibody; FDA-approved for chemorefractory KRAS G12C-mutated colorectal cancer (CodeBreaK 300). The board ranked this the preferred option on progression through checkpoint blockade. Alternative: adagrasib (Krazati) + cetuximab (Erbitux).

Later line Regorafenib (Stivarga) Tumour-type standard of care

An oral multi-kinase inhibitor with antiangiogenic and stromal activity. FDA-approved for metastatic colorectal cancer previously treated with fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy and an anti-VEGF agent. Selected on tumour type and line of therapy; no predictive biomarker applies. Held for the refractory setting after the options above.

Backbone therapy FOLFIRI Standard of care

Tumour-type standard of care: first- or second-line cytotoxic backbone for metastatic colorectal cancer. Selected by cancer type, line and fitness — not matched to a molecular biomarker. Regimen: folinic acid + fluorouracil + irinotecan.

Order first: DPYD and UGT1A1 genotyping are required before this backbone only — they do not gate the precision options above. See Testing.

Do not use Fam-Trastuzumab Deruxtecan-nxki - Contraindicated in this molecular context.

Do not use Tucatinib - Contraindicated in this molecular context.

Clinical situation

Ms Stellar is a 47-year-old woman with metastatic adenocarcinoma of the sigmoid colon, diagnosed in December 2025 following a four-month history of altered bowel habit and iron-deficiency anaemia. Staging CT showed five hepatic metastases with involved regional nodes and no thoracic disease (M1a). A near-obstructing primary was managed with an endoscopic self-expanding metal stent in December 2025; she did not undergo resection.

She received first-line FOLFOX with bevacizumab from January 2026, with a partial response at first restaging (target sum 96 → 61 mm, 5 March) and a fall in CEA from 62.4 to 14.0 ng/mL. CEA then doubled to 33.0 ng/mL by 10 April. On that biochemical signal alone — six weeks before any radiological change — her oncologist arranged a repeat biopsy of the stented primary, performed on 4 May and taken in preference to a liver biopsy because she remained on bevacizumab.

That biopsy produced the substantive finding. The FoundationOne CDx report of 18 May returned MSI-High with a tumour mutational burden of 41 mut/Mb, driven by an MSH2 p.A636P founder allele, alongside KRAS p.G12C, PIK3CA p.H1047R, APC p.R1450*, a BRCA2 frameshift and ERBB2 amplification. None of this was known when first-line therapy was selected: mismatch-repair immunohistochemistry was not requested on the December diagnostic biopsy, and the March plasma assay reported MSI-High as not detected, a recognised limitation of cell-free DNA testing. MMR immunohistochemistry on the May block and a confirmatory germline panel were both still pending at the time of this board.

Restaging on 22 May, four days after that report, confirmed progression — two new hepatic deposits and a 38% increase from nadir, with CEA 91.4 ng/mL and a cholestatic pattern of liver function. First-line therapy was stopped. The genomic result was therefore already in hand when progression was confirmed, which is why the board convened on a second-line question rather than an investigative one.

She remains ECOG 1 and fully independent, with grade 1 oxaliplatin-related peripheral neuropathy. She is motivated for active treatment, including a clinical trial.

Therapy exposure, clinical events and tumour-marker trends

Progression Response Surgery / radiation / other

Bottom line · clinician synthesis

dMMR/MSI-H with a tumour mutational burden of 41 mut/Mb is the dominant finding and outranks every other alteration on this panel. MSH2 p.A636P is the likely mechanism and, at 46% allele fraction, is probably germline — Lynch syndrome, with implications for the family as well as the patient. She is ECOG 1, 47, and has never received checkpoint blockade, so pembrolizumab is the board’s first recommendation; KRAS, PIK3CA and ERBB2 do not predict immunotherapy benefit in MSI-H disease and should not pull rank here. KRAS G12C is held in reserve — sotorasib with panitumumab preferred on progression, adagrasib with cetuximab the alternative. ERBB2 amplification is not actionable: HER2-directed therapy in colorectal cancer requires RAS wild-type. BRCA2 frameshift with elevated gLOH is not actionable outside a trial — genomic-instability metrics are not validated in colorectal cancer and are confounded by mismatch-repair deficiency.

Treatment Eligible treatment options Evaluate for this patient · confirm conditions & interactions before prescribing

Pembrolizumab (Keytruda) MICROSATELLITE INSTABILITY Standard of care Do first FDA-approved

Owner · Treating oncologist When · At next review

Full dosing & evidence · §2

Dostarlimab-gxly (Jemperli) MISMATCH REPAIR Standard of care Alternative FDA-approved

Owner · Treating oncologist When · At next review

A PD-1-blocking antibody exploiting the high neoantigen load of mismatch-repair-deficient tumours. FDA-approved for solid tumors that are dMMR, have spread or come back, and got worse during or after other treatment and cannot be treated with other therapies.

Full dosing & evidence · §2

Regorafenib (Stivarga) Tumour-type SoC Standard of care Reserve · later-line FDA-approved

Dose · 160 mg STIVARGA (four 40 mg tablets) taken orally once daily for the first 21 days of each 28-day cycle

Owner · Treating oncologist When · At next review

An oral multi-kinase inhibitor with antiangiogenic and stromal activity. Selected on tumour type and line, not on a biomarker — FDA-approved for metastatic colorectal cancer previously treated with fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy and an anti-VEGF agent. Not a precision match to KRAS; the gene shown in the match tables reflects the filter row only.

Class-level drug interactions for this agent (CYP3A4 inducers and inhibitors, BCRP substrates) are listed in Appendix F. None is currently charted for this patient.

Full dosing & evidence · §2

FOLFIRI SoC chemo Test required first Non-precision

Owner · Treating oncologist When · At next review

Tumour-type standard of care: first- or second-line cytotoxic backbone for metastatic colorectal cancer. Selected by cancer type, line and fitness - not matched to a molecular biomarker.

↳ Regimen: Folinic acid + Fluorouracil + Irinotecan.

↳ UGT1A1 poor metaboliser (*28/*28 or Gilbert syndrome) - reduce irinotecan dose

↳ Requires DPYD genotyping and UGT1A1 genotyping before use - see Testing.

Standard-of-care chemotherapy · §1.4

Testing Follow-up testing Order now - each result would change management

DPYD genotyping forFOLFIRI Required before treatment

Owner · Molecular pathology / PGx lab When · Before fluoropyrimidine (5-FU/capecitabine)

DPYD deficiency causes severe/life-threatening fluoropyrimidine toxicity; pre-treatment genotyping is EMA-mandated (ESMO III,A). Variant carriers need ≥50% dose reduction. Applies to the FOLFIRI chemotherapy backbone only. It is not required before pembrolizumab, sotorasib + panitumumab, adagrasib + cetuximab or regorafenib, and must not delay them.

UGT1A1 genotyping forFOLFIRI Required before treatment

Owner · Molecular pathology / PGx lab When · Before irinotecan

UGT1A1*28/*28 (poor metaboliser) raises the risk of severe neutropenia/diarrhoea with irinotecan; consider a 30% starting dose reduction. Applies to the FOLFIRI chemotherapy backbone only — specifically the irinotecan component. It does not gate any of the precision options on this roadmap.

MMR IHC on block S26-041992B forPembrolizumab Added by clinician

Owner · Molecular pathology / PGx lab When · Before starting the proposed treatment

Confirms dMMR and shows which protein is lost (expect MSH2/MSH6, concordant with A636P). You're committing to IO on sequencing alone until it lands.

Germline Panel MSH2 Added by clinician

Owner · Molecular pathology / PGx lab When · Before starting the proposed treatment

MSH2 A636P at 46% and BRCA2 S1982Rfs*22 at 45% are both germline-range. Confirms Lynch → cascade testing for her two children, and her own endometrial/ovarian/urothelial surveillance.

B2m & JAK1/JAK2 status ERBB2 Added by clinician

Owner · Molecular pathology / PGx lab When · Before starting the proposed treatment

Canonical primary / acquired resistance mechanisms to checkpoint blockade in MSI-H CRC.

Pre-IO safety baseline forPembrolizumab Added by clinician

Owner · Molecular pathology / PGx lab When · Before starting the proposed treatment

TSH/free T4, morning cortisol, HbA1c, autoimmune history, hepatitis serology.

ctDNA test at IO Baseline and ~8 weeks Added by clinician

Owner · Molecular pathology / PGx lab When · Before starting the proposed treatment

gives an early molecular readout and shows the KRAS/PIK3CA subclone responding or not.

Adjunct agents Repurposed / adjunct agents Phase II+ evidence - discuss benefit-risk with patient before use

Aspirin MSH2 Phase III evidence

Aspirin - Aspirin reduced Lynch syndrome-associated colorectal cancer incidence by ~60% (CAPP2 Phase III RCT, 600 mg/day for >=2 years; Burn et al., Lancet 2011). Chemoprevention, not treatment of established cancer. COX-mediated prostaglandin signalling promotes proliferation and immune evasion in MMR-deficient colorectal mucosa; aspirin's irreversible COX inhibition counteracts this and may selectively promote apoptosis in MMR-deficient cells.

↳ Eligibility: Germline or somatic MMR gene inactivation (Lynch syndrome or MMR-deficient CRC); CAPP2 dose was 600 mg/day - assess cardiovascular and GI bleeding risk; CAPP3 is evaluating lower doses (100/300 mg) and may revise dosing.

↳ Discuss with: Medical genetics / gastroenterology / oncology in discussion with the patient.

PMID 22036019

Celecoxib (Celebrex) APC Phase III evidence

Celecoxib - Celecoxib significantly reduced colorectal adenoma burden vs. placebo in Phase III RCTs (PreSAP, NEJM 2006; APC trial). Chemoprevention (adenoma reduction), not treatment of invasive cancer. APC loss drives constitutive Wnt/beta-catenin signalling that upregulates COX-2; selective COX-2 inhibition removes a key proliferative and survival signal in APC-mutant colorectal mucosa.

↳ Eligibility: Germline APC (FAP/AFAP) chemoprevention context; the prior FDA FAP indication was voluntarily withdrawn in 2012 (confirmatory trial not completed - not for safety or efficacy reasons), so this use is now off-label; dose-dependent cardiovascular risk - review CV profile before use; not a substitute for colonoscopic surveillance.

↳ Discuss with: Gastroenterology / colorectal surgery / oncology.

PMID 16943401

Trials Clinical trials Consider if standard options are exhausted

TAPUR ERBB2 Biomarker-matched (curated)

Owner · Treating oncologist When · If disease progresses

Trial enrolment - Targeted Agent and Profiling Utilization Registry (TAPUR) Study (Phase PHASE2) · matched on ERBB2. The purpose of the study is to learn from the real world practice of prescribing targeted therapies to patients with advanced cancer whose tumor harbors a genomic variant known to be a drug target or to predict sensitivity to a drug. NOTE: Due to character limits, the arms section does NOT include all TAPUR Study relevant biomarkers. For additional information, contact TAPUR@asco.org, or if a patient, your nearest participating TAPUR site (see participating centers)…

NCT02693535 · Appendix E

Monitor Monitoring & surveillance Non-drug surveillance to track over time - not a treatment

Track emerging resistance clones KRAS, PIK3CA Seeded · review

Serial testing shows the KRAS G12C (2.6% → 34%) and PIK3CA H1047R (1.7% → 29%) subclones expanding. Recommend serial ctDNA to track clonal evolution and detect emerging resistance before radiographic progression.

Cadence · With each response assessment (≈ every 8–12 weeks)

iRECIST response assessment IMAGING

At first restaging (8-9 weeks) with confirmatory imaging 4-8 weeks later before abandoning IO

TP53 serial ctDNA TP53

foundation one report noted absence from May panel may be due to sampling issues.

Acquired resistance monitoring via B2M and JAK1/JAK2 loss B2M, JAK1, JAK2

re-biopsy or ctDNA for B2M and JAK1/JAK2 loss, the canonical acquired-resistance mechanisms in MSI-H CRC.

Molecular rationale

Supporting evidence for the actions above - the variants driving each recommendation.

Clinically actionable variants

MAPK driver

KRAS G12C

Substitution locks KRAS in the GTP-bound active state, constitutively activating the RAS-MAPK cascade and driving uncontrolled tumour-cell proliferation.

Targetable Eligible to be evaluated for sotorasib or adagrasib (KRAS G12C inhibitors approved in CRC).

MMR deficiency

MSH2 A636P

Pathogenic missense disrupts MSH2:MSH6 (MutSα) mismatch repair complex, causing microsatellite instability and genome-wide hypermutation.

Targetable MSI-H/dMMR status; eligible to be evaluated for pembrolizumab or dostarlimab-gxly.

Potentially actionable variants

HRR loss

BRCA2 S1982Rfs*22

Frameshift truncates BRCA2, abrogating homologous recombination repair and MMEJ, leading to genomic instability and potential sensitivity to PARP inhibition.

Investigational Consider PARP inhibitor evaluation (e.g. niraparib or talazoparib) in trial context; germline testing warranted.

PI3K activating mutation

PIK3CA H1047R

Hotspot gain-of-function mutation in the kinase domain constitutively activates PI3K/AKT signalling, promoting tumour-cell survival and metabolic reprogramming.

Investigational No approved PI3K-directed therapy in CRC; relevant for clinical trial eligibility.

HER2 amplification

ERBB2 amplification

ERBB2 gene amplification overexpresses HER2 receptor, activating PI3K/AKT and MAPK signalling; however, co-occurring KRAS G12C mutation renders HER2-directed therapy ineffective.

Investigational HER2-directed therapy contraindicated in RAS-mutant CRC; amplification noted as context only.

Other variants of biological interest

WNT pathway loss

APC R1450* (truncating)

Nonsense mutation eliminates APC tumour suppressor function, constitutively activating WNT/β-catenin signalling and driving cellular plasticity and proliferation.

Context Canonical CRC driver; no approved targeted therapy; defines WNT-driven tumour biology.

Section 2 Therapeutic implications Evidence in Appendix D · dosing in Appendix F

Standard-of-care, biomarker-matched

FDA-approved for this indication and biomarker.

Drug / class	Clinical use / why it matters	Evidence	Appendix refs
Regorafenib (Stivarga) Dose · 160 mg STIVARGA (four 40 mg tablets) taken orally once daily for the first 21 days of each 28-day cycle	On-label for this cancer type. Regorafenib is FDA-approved for metastatic colorectal cancer previously treated with fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy, an anti-VEGF therapy and, if RAS wild-type, an anti-EGFR therapy. Selected on tumour type and line rather than on a biomarker — the KRAS association in the match table reflects the filter row, not a predictive relationship. Confirm prior-therapy criteria before use.	Biomarker-supported	Evidence · Dosing FDA-LABEL
Adagrasib (Krazati) + Cetuximab (Erbitux)	On-label biomarker-matched option for KRAS / Colorectal cancer. Confirm line of therapy and biomarker requirements from the label. Accelerated approval; verify the current label and criteria	FDA-approved	Evidence · Dosing FDA-LABEL
Sotorasib (Lumakras) + Panitumumab (Vectibix)	On-label biomarker-matched option for KRAS / Colorectal cancer. Confirm line of therapy and biomarker requirements from the label	FDA-approved	Evidence · Dosing FDA-LABEL
Dostarlimab-gxly (Jemperli)	On-label biomarker-matched option for Mismatch Repair / Solid tumors. Label appears to be later-line or post-progression. Accelerated approval; verify the current label and criteria	FDA-approved	Evidence · Dosing FDA-LABEL
Pembrolizumab (Keytruda)	On-label biomarker-matched option for Microsatellite Instability / Microsatellite instability -high (MSI-H) or mismatch repair deficient (dMMR) cancer. Confirm line of therapy and biomarker requirements from the label	FDA-approved	Evidence · Dosing FDA-LABEL

Standard-of-care chemotherapy (tumour-type based)

Guideline-recommended cytotoxic regimens for this cancer type, line of therapy, and fitness - independent of molecular biomarkers.

Drug / class	Clinical use / why it matters	Evidence	Appendix refs
Folinic acid + Fluorouracil (Adrucil) + Irinotecan (Camptosar)	First- or second-line cytotoxic backbone for metastatic colorectal cancer (NCCN Colon Cancer, Category 1; PMID:10744089)	SoC chemotherapy	Evidence NCCN Colon Cancer, Category 1;

Investigational / trial-eligible

Active clinical trials matching this profile.

Drug / class	Clinical use / why it matters	Evidence	Appendix refs
Palbociclib, Sunitinib, Temsirolimus	NCT02693535 · PHASE2 · high-confidence match · matched on ERBB2 · 47 mi to Windham Hospital (WH)	Trial lead	Trials
FOLFIRI Protocol, ACTIVE SURVEILLANCE, Nivolumab Protocol	NCT03803553 · PHASE3 · nearby site · matched on MSH2 · 50 mi to Massachusetts General Hospital	Trial lead	Trials
APL-101 Oral Capsules	NCT03175224 · PHASE2 · nearby site · matched on KRAS · 50 mi to Dana Farber Cancer Institute	Trial lead	Trials

Drugs to avoid

Contraindicated / resistance-relevant - avoid based on the molecular profile.

Drug / class	Clinical use / why it matters	Evidence	Appendix refs
Fam-Trastuzumab Deruxtecan-nxki	Avoid Fam-Trastuzumab Deruxtecan-nxki (HER2-directed therapy): primary resistance resistance linked to KRAS.	Resistance	Evidence · Resistance
Tucatinib (Tukysa)	Avoid Tucatinib (HER2-directed therapy): primary resistance resistance linked to KRAS.	Resistance	Evidence · Resistance

Repurposed / adjunct agents which may be suitable for this patient

Phase II+ clinical trial evidence - not standard-of-care; discuss benefit-risk before use.

Aspirin MSH2 Phase III evidence

Reference: PMID 22036019 ↗

Celecoxib (Celebrex) APC Phase III evidence

Reference: PMID 16943401 ↗

Section 2AResistance and exclusionsWhy matched agents were ruled out

Three agents matched this tumour on biomarker and were nonetheless excluded. Each was ruled out by an alteration already on the panel — the same KRAS G12C that makes one regimen available is what makes the HER2-directed options unusable. Recording the exclusion and its basis matters as much as recording the recommendation: a board that cannot say why an obvious-looking option was rejected has not finished the analysis.

Agent / class	Resistance type	Driver	Basis
Tucatinib + trastuzumab HER2-directed	Primary · intrinsic	KRAS p.G12C	HER2-directed therapy in colorectal cancer is indicated only in RAS wild-type disease; the registrational study enrolled RAS wild-type patients exclusively. An activating KRAS mutation downstream of HER2 maintains MAPK signalling irrespective of receptor blockade, so the alteration that matched this agent also predicts its failure.
Fam-trastuzumab deruxtecan HER2-directed ADC	Not confirmed	HER2 IHC absent	A different basis from the antibody regimens above. An antibody-drug conjugate kills by delivering a cytotoxic payload after internalisation rather than by blocking HER2 signalling, so downstream KRAS activation is not an absolute barrier — RAS-mutant patients were enrolled in the registrational colorectal study, with lower response rates. The decisive gap here is confirmation: the tumour-agnostic HER2 indication requires IHC 3+, and this case has amplification by sequencing only with no HER2 immunohistochemistry on file. Reconsider if IHC returns 3+.
Cetuximab / panitumumab anti-EGFR, as monotherapy	Primary · intrinsic	KRAS p.G12C	Activating KRAS mutation confers intrinsic resistance to anti-EGFR antibodies used alone. Note this does not extend to their use with a KRAS G12C inhibitor, where the anti-EGFR antibody blocks the adaptive feedback that limits single-agent activity — which is why the reserve regimens on this report pair the two.

Resistance here is primary (present before exposure), not acquired. No agent on this roadmap has been given and escaped; acquired mechanisms to checkpoint blockade — B2M and JAK1/JAK2 loss — are the subject of an active monitoring action rather than an observed finding.

Section 3 Key findings

MSH2

MSH2 A636P (Lynch founder) · VAF 46.0%

FDA-matched

DNA mismatch repair · loss-of-function missense · germline-range VAF

Drives MSI-H/dMMR - first-line checkpoint blockade in §1

APC

APC R1450* (truncating) · VAF 78.0%

Context

WNT signaling · loss-of-function missense

Driver - therapeutic implications in §2

BRCA2

BRCA2 S1982Rfs*22 (founder) · VAF 45.0%

Investigational

Homologous recombination repair · frameshift, loss of function

Driver - therapeutic implications in §2

KRAS

KRAS G12C · VAF 34.0%

Resistance signal

MAPK signaling · activating missense

Driver - therapeutic implications in §2

PIK3CA

PIK3CA H1047R · VAF 29.0%

Tier II

PI3K/AKT signaling · activating missense

Variant of uncertain actionability - see Section 5 for context.

ERBB2

ERBB2 amplification

Resistance signal

HER2 amplification · not actionable with co-occurring KRAS G12C

HER2-directed therapy excluded - see §2A

^† No curated evidence in CIViC for this variant. Absence of curated evidence does not imply clinical insignificance - it may reflect a curation gap rather than a benign or likely-benign call.

Key terms

Tier I - Variant of strong clinical significance - FDA-approved therapy or professional-guideline evidence in this tumour type.

Tier II - Variant of potential clinical significance - therapy evidence in another tumour type, or strong biological / clinical-trial support.

Tier III - Variant of uncertain clinical significance.

Tier IV - Benign or likely benign. An asterisk (Tier IV*) marks a curation gap - no curated evidence in CIViC - which is not a confirmed benign call.

VAF - Variant allele fraction - the percentage of sequencing reads carrying the variant; a rough proxy for the fraction of tumour cells affected.

Hot-spot - A recurrently mutated codon catalogued as a cancer hotspot.

Section 3A Pathways affected See Appendix B · pathway topology

Four pathways are disrupted in this tumour, and they are not equally consequential. DNA mismatch repair is the one that matters most: loss of MSH2 function removes the cell’s ability to correct replication errors, producing the microsatellite instability and the 41 mut/Mb mutational burden that make this tumour immunogenic — and the frameshifts in RNF43, TGFBR2 and ARID1A reported as variants of uncertain significance are themselves the fingerprint of that defect. RTK→MAPK/PI3K signalling is activated at two independent points: KRAS G12C locks the GTPase in its active state, and PIK3CA H1047R independently drives PI3K–AKT, so blocking either arm alone leaves the other intact. ERBB2 amplification sits upstream of both but is therapeutically inert here, because HER2-directed therapy requires RAS wild-type. WNT/β-catenin is constitutively active through truncating APC loss — the canonical initiating event in colorectal cancer, and not currently druggable. Homologous recombination repair is compromised by the BRCA2 frameshift, which raises the genomic-instability score; that signature is not validated in colorectal cancer and is confounded by the mismatch-repair defect, so it is interpreted with caution.

DNA mismatch repair

Loss of MutSα (MSH2:MSH6) removes correction of replication errors, producing microsatellite instability and genome-wide hypermutation. The dominant defect in this tumour.

MSH2

RTK → MAPK / PI3K signalling

Proliferative and survival signalling, activated here at two independent points — so blocking one arm leaves the other intact.

KRAS PIK3CA ERBB2

WNT / β-catenin

Truncating APC loss drives constitutive β-catenin signalling. The canonical initiating event in colorectal cancer; no approved targeted therapy.

APC

Homologous recombination repair

BRCA2 frameshift compromises double-strand break repair and raises the genomic-instability score — a metric not validated in colorectal cancer and confounded by dMMR.

BRCA2

Section 3B Hallmarks of cancer impacted

Cellular plasticity and differentiation state

active APC

Genome instability and mutation

active BRCA2 MSH2

Deregulating cellular energetics

active ERBB2 PIK3CA

Sustaining proliferative signalling

active KRAS ERBB2 PIK3CA …

Sustaining proliferative signalling is the most convergently active process in this tumour, driven by four concurrent alterations: APC R1450* truncation releases β-catenin from its destruction complex, locking Wnt transcriptional programmes into a constitutively active state; KRAS G12C renders RAS GTPase self-inactivation defective, keeping the MAPK proliferation cascade running without upstream input; PIK3CA H1047R constitutively activates PI3K-AKT to suppress apoptosis and accelerate cell-cycle entry; and ERBB2 amplification floods the cell with receptor tyrosine kinase signalling that reinforces both the MAPK and PI3K arms - together these alterations make the tumour functionally independent of external growth signals and resistant to single-pathway blockade. Genome instability is also active, with BRCA2 frameshift loss disabling homologous recombination so that double-strand DNA breaks accumulate and are repaired by error-prone pathways, while MSH2 missense disruption impairs mismatch repair, generating a hypermutator phenotype that accelerates clonal evolution and may confer sensitivity to immune checkpoint blockade. The ERBB2 and PIK3CA alterations additionally rewire cellular metabolism - channelling glucose flux toward biosynthetic pathways that support the rapid biomass accumulation demanded by the proliferative programme - reinforcing the tumour's capacity for sustained growth even under nutrient-limited conditions.

Key terms

Hallmark - One of the biological capabilities a cancer acquires (Hanahan & Weinberg). Bars show how strongly this tumour's altered genes map to each hallmark.

Axis - A connected signalling route (e.g. the CDK4/6-CDKN2A-RB1 axis) along which several alterations converge on one therapeutic target.

Section 4 Immunotherapy biomarkers Tumour-agnostic IO decision drivers

These four markers are the reason immunotherapy leads this roadmap. A mismatch-repair-deficient tumour accumulates thousands of mutations it cannot correct; each is a potential neoantigen, and the resulting immune visibility is what checkpoint blockade exploits. MSI-High is both the mechanism and the approval criterion — it carries a tumour-agnostic indication and, in colorectal cancer specifically, first-line evidence against chemotherapy. TMB of 41 mut/Mb is far above the 10 mut/Mb tumour-agnostic threshold and is a consequence of the same defect rather than an independent finding; the two should be read together, not summed. PD-L1 CPS 8 is supportive but not determinative — PD-L1 expression does not gate eligibility in MSI-H colorectal cancer, and a low score would not have argued against treatment. HRD is the outlier: it is elevated, but genomic-instability metrics are validated in ovarian and not colorectal disease, and mismatch-repair deficiency inflates them, so it is not treated as an actionable immunotherapy or PARP signal here. The practical consequence is that this patient’s best-evidenced option was available from diagnosis and was not used, because MSI status was not established until first-line therapy had already failed.

TMB 41.0 mut/Mb High

MSI / MMR MSI-H MSI-High

HRD Positive gLOH 19.4% Not validated in CRC

PD-L1 CPS 8 Positive, non-gating 22C3 · 20 May 2026

Key terms

TMB - Tumour mutational burden, in mutations per megabase (mut/Mb). TMB-high (>= 10 mut/Mb) predicts immunotherapy benefit (FDA pembrolizumab, KEYNOTE-158).

MSI / MMR - Microsatellite instability / mismatch-repair status. MSI-High (MSI-H) or deficient MMR (dMMR) predicts immunotherapy benefit; MSS = microsatellite stable.

HRD - Homologous-recombination deficiency - a genomic-instability score; >= 42 is positive (Myriad MyChoice CDx). Components: LOH (loss of heterozygosity), LST (large-scale state transitions), TAI (telomeric allelic imbalance).

PD-L1 - Programmed death-ligand 1 immunohistochemistry, reported as TPS (tumour proportion score, %), CPS (combined positive score), or IC.

ReferencesKey literatureEvidence underpinning the roadmap · full list in Appendix D/3

These publications were reviewed and cited by the molecular tumour board for this case.

Pembrolizumab in Microsatellite-Instability–High Advanced Colorectal Cancer

N Engl J Med · 2020 · PMID 33264544 ↗

KEYNOTE-177 — first-line pembrolizumab versus chemotherapy in MSI-H/dMMR metastatic colorectal cancer.

PD-1 Blockade in Tumors with Mismatch-Repair Deficiency

N Engl J Med · 2015 · PMID 26028255 ↗

The foundational study establishing mismatch-repair status as a predictor of response to PD-1 blockade.

Pembrolizumab in microsatellite-instability-high and mismatch-repair-deficient advanced solid tumors: updated results of the KEYNOTE-158 trial

Nat Cancer · 2025 · PMID 39979665 ↗

KEYNOTE-158 updated results — 4.5 years of follow-up in MSI-H/dMMR solid tumours; ORR 33.8%, median overall survival 19.8 months.

Overall Survival Analysis of the Phase III CodeBreaK 300 Study of Sotorasib Plus Panitumumab Versus Investigator's Choice in Chemorefractory KRAS G12C Colorectal Cancer

J Clin Oncol · 2025 · PMID 40215429 ↗

CodeBreaK 300 overall-survival analysis — sotorasib plus panitumumab in chemorefractory KRAS G12C colorectal cancer.

Section 5 Longitudinal analysis How the tumour's clonal populations have evolved between serial samples

Samples compared

Two sequential molecular profiles, 77 days apart.

Prior 2026-03-02 Guardant Health Guardant360 CDx On: FOLFOX + Bevacizumab

77 days

Current 2026-05-18 Foundation Medicine FoundationOne CDx

Read the trajectories below in the context of documented therapy: FOLFOX + Bevacizumab.

Treatment & molecular timeline

Therapy exposure, variant allele fractions, serum tumour marker and clinical events on one time axis. VAF is read on the left ruler (0–80%); the x-axis is scaled to elapsed time between samples.

Cross-assay caveat. Adjacent samples differ in specimen type (liquid biopsy → tissue); absolute VAF is not directly comparable across this interval - read the trajectory qualitatively.

APC ▲ targetable BRCA2 ▲ targetable KRAS ▲ targetable ⚠ resistance PIK3CA TP53

Clonal evolution

Each track plots the founding (prior, hollow) and current (filled) variant allele fraction. A rising fraction marks an expanding clone; a variant with no prior point is a newly emergent subclone.

0 20 40 60% VAF

Unassayed in prior panel

Present in the May tissue panel and not evaluable in the March plasma assay — the mismatch-repair genes are outside that assay's reportable target capture, and both allele fractions are germline-range. No trajectory is computed: absence from a region a panel does not cover is not evidence of absence from the tumour.

MSH2 MSH2 A636P (Lynch founder)

Unmapped

46.0%

+46.0% +46.0%

BRCA2 BRCA2 S1982Rfs*22 (founder)

Genome instability

45.0%

+45.0% +45.0%

Newly emergent subclones

Copy-number calling on cell-free DNA requires a higher tumour fraction than single-nucleotide variant detection, so a negative plasma result is weaker evidence of absence than for a point mutation.

ERBB2 ERBB2 amplification

Proliferative signaling

0.0%

+0.0% +0.0%

Expanding clones - rising VAF

KRAS KRAS G12C

Proliferative signaling

2.6

34.0%

+31.4 Expanding

PIK3CA PIK3CA H1047R

Proliferative signaling

1.7

29.0%

+27.3 Expanding

Founding clone — stable clone size

TP53 p.R175H was measured at 6.1% allele fraction in the March plasma — which, against that sample’s 21% tumour fraction, corresponds to roughly 58% of tumour cells — and is not detected in the May specimen, so it has no track here. The two samples are different compartments — plasma is shed predominantly by the hepatic deposits, while the May panel was run on a repeat biopsy of the primary — so this is a difference in what was sampled and must not be read as clearance of that clone. Serial ctDNA to re-measure it is an active monitoring action.

APC APC R1450* (truncating)

Evading growth suppressors

15.8

78.0%

+62.2 Persisted

Hallmark dominance over time

Which cancer hallmarks are gaining ground as the clones expand.

Evading growth suppressors

APC

▲ Rising

Net VAF gains across APC - this axis is gaining ground as the clones expand.

Sustaining proliferative signalling

KRAS · PIK3CA · ERBB2

▲ Rising

Net VAF gains across KRAS, PIK3CA - this axis is gaining ground as the clones expand.

Unmapped pathway

MSH2

■ Present throughout

Not a change this interval — the mismatch-repair genes are outside the March panel, and the 46% allele fraction is germline-range. This is the tumour’s defining defect, not a new event.

Genome instability and mutation

BRCA2 · TP53

■ Sampling difference

Neither is a true interval change: BRCA2 is germline-range and outside the March panel, and TP53 is unsampled in the May tissue rather than cleared.

Interpretation

Between the two timepoints the dominant change is the expansion of a single subclone. KRAS p.G12C rose from roughly a quarter of tumour cells to near-clonal, and PIK3CA p.H1047R moved with it at a similar fraction — consistent with one population carrying both, selected under FOLFOX. APC p.R1450* was clonal at both timepoints and is the backbone that identifies these as the same tumour; its apparent VAF rise (15.8% to 78%) reflects the difference in tumour content between a plasma sample and a tissue specimen, not clonal growth.

The MSH2, BRCA2 and ERBB2 findings are labelled emergent above because they were absent from the March assay, but that is an artefact of panel coverage rather than genuine emergence: the plasma panel does not report the mismatch-repair genes, and the MSH2 and BRCA2 allele fractions (46% and 45%) are in the germline range — these were almost certainly present from the outset. TP53 p.R175H is the reverse case: 6.1% allele fraction in March plasma (approximately 58% of tumour cells at that sample’s 21% tumour fraction) and not detected in the May specimen. Plasma is shed predominantly by the hepatic deposits while the May panel was run on a repeat biopsy of the primary, so this is compartment discordance and differential shedding, not clonal clearance. It is tracked by serial ctDNA rather than treated as resolved.

What this does not change is the therapeutic order. ERBB2 amplification is not a target in this tumour: HER2-directed therapy in colorectal cancer requires RAS wild-type disease, and the same KRAS p.G12C that now dominates the sampled population sustains MAPK signalling downstream of the receptor irrespective of HER2 blockade. The dominance of the KRAS/PIK3CA subclone makes that bypass more entrenched, not less. Nor does clonal evolution displace immunotherapy: mismatch-repair deficiency is a property of the tumour rather than of a subclone, and MSI-High with a burden of 41 mut/Mb is already established on the May tissue — no further MSI testing is required to act, only the pending confirmatory immunohistochemistry. Checkpoint blockade therefore remains first-line and the KRAS G12C combinations remain in reserve for progression, with serial ctDNA to follow both the expanding subclone and the unsampled TP53 population.

Show raw variant trajectory & per-sample calls

2026-03-02 · Guardant Health · Guardant360 CDx

APC R1450* (truncating) TP53 R175H KRAS G12C PIK3CA H1047R

2026-05-18 · Foundation Medicine · FoundationOne CDx

MSH2 A636P (Lynch founder) KRAS G12C PIK3CA H1047R APC R1450* (truncating) BRCA2 S1982Rfs*22 (founder) ERBB2 amplification

Gene	Alteration	VAF current	VAF prior	Δ VAF	Status
MSH2	MSH2 A636P (Lynch founder)	46.0%	-	-	New
KRAS	KRAS G12C	34.0%	2.6%	+31.4%	Progressed ↑
PIK3CA	PIK3CA H1047R	29.0%	1.7%	+27.3%	Progressed ↑
APC	APC R1450* (truncating)	78.0%	15.8%	+62.2%	Persisted
BRCA2	BRCA2 S1982Rfs*22 (founder)	45.0%	-	-	New
ERBB2	ERBB2 amplification	-	-	-	New
TP53	TP53 R175H	-	6.1%	n/a	Not detected (tissue)

Appendix A

Full List of Molecular Alterations Based on AMP/ASCO/CAP tier framework · cancerhotspots v2

Gene	Variant	Type	Allele / FC	Pathway role	Tier	Resistance
MSH2	MSH2 A636P (Lynch founder)	Mut	46%	DNA mismatch repair	Tier III†	N/A
KRAS	KRAS G12C	Driver	34%	MAPK signaling	Tier I	N/A
PIK3CA	PIK3CA H1047R	Driver	29%	PI3K/AKT signaling	Tier II	N/A
APC	APC R1450* (truncating)	Driver	78%	WNT signaling	Tier III†	N/A
BRCA2	BRCA2 S1982Rfs*22 (founder)	Driver	45%	Homologous recombination repair	Tier III†	N/A
ERBB2	ERBB2 amplification	Amp	Unavailable	ERBB2 signaling	Tier II	N/A

^† No curated evidence in CIViC for this variant. Absence of curated evidence does not imply clinical insignificance - it may reflect a curation gap rather than a benign or likely-benign call.

Appendix B

Pathway Topology Maps

These are canonical topology maps for the pathways carrying an alteration in this case, drawn from a curated interaction database, with the patient’s altered genes highlighted. They show how each pathway is wired, not what was measured in this tumour — node positions and edges are identical for any patient with an alteration in that pathway, and no map encodes allele fraction, clone size or treatment response. Read them as reference anatomy for Section 3A, not as patient-specific findings.

What is and is not covered, stated plainly. The map labelled ERBB2 signaling contains neither ERBB2 nor KRAS as a node — it is the shared RAS/MAPK and PI3K downstream cascade, so the signalling axis behind this case is represented, just not under a gene name you would search for. PI3K/AKT signaling carries PIK3CA; DNA repair carries BRCA2 and is homologous recombination, not mismatch repair; WNT signaling carries APC inside a composite complex node. There is no mismatch-repair map, which is the one gap that matters — MMR deficiency is the dominant finding in this case and the basis of the first-line recommendation, and no topology for it exists in the source database. The ERBB2 map is also the most elaborate here despite HER2-directed therapy being excluded (Section 2A); that reflects database coverage, not clinical weight.

0 altered nodes · 9 of 10 proteins shown · 3 modules collapsed · SIGNOR via OmniPath · also visible in: PI3K/AKT signaling

Collapsed modules (3)

BRAF → ERK1/2 relay (1 protein)	BRAF → MEK1/2 → ERK1/2 (activates)
GRB2 → BRAF relay (2 proteins)	GRB2 → SOS1 → HRAS → BRAF (other)
PIP3-PDK1 relay	PIP3 → PDPK1 → AKT (activates)

Full pathway: see Appendix B2.

1 altered node · 17 of 46 proteins shown · 5 modules collapsed · SIGNOR via OmniPath · also visible in: ERBB2 signaling, WNT signaling

Collapsed modules (5)

AMPK → MITOCHONDRIAL_BIOGENESIS relay (1 protein)	AMPK → PPARGC1A → MITOCHONDRIAL_BIOGENESIS (activates)
BRAF → ERK1/2 relay (1 protein)	BRAF → MEK1/2 → ERK1/2 (activates)
ERK1/2 → PI3K relay (1 protein)	ERK1/2 → GAB1 → PI3K (binds)
GSK3B → GLYCOGEN_SYNTHESIS relay (1 protein)	GSK3B → GYS1 → GLYCOGEN_SYNTHESIS (activates)
RAS adapter relay	RTKS → GRB2 → SOS1 → HRAS (other)

Full pathway: see Appendix B2.

0 altered nodes · 34 of 34 proteins shown · 2 modules collapsed · SIGNOR via OmniPath · also visible in: PI3K/AKT signaling

Collapsed modules (2)

DAAM1 → ROCK1 relay (1 protein)	DAAM1 → RHOA → ROCK1 (binds)
WNT → GSK3B/AXIN/APC relay (1 protein)	WNT → LPR5/6 → GSK3B/AXIN/APC (binds)

Full pathway: see Appendix B2.

1 altered node · 9 of 20 proteins shown · 4 modules collapsed · SIGNOR via OmniPath

Collapsed modules (4)

ATM → RIF1 relay (1 protein)	ATM → TP53BP1 → RIF1 (binds)
DNA_DAMAGE → DNA_REPAIR relay (1 protein)	DNA_DAMAGE → MLH1/PMS2 → DNA_REPAIR (activates)
DNA_DAMAGE → DNA_REPAIR relay (2 proteins)	DNA_DAMAGE → SLX4 → ERCC4/ERCC1 → DNA_REPAIR (other)
MRE11 → DNA_REPAIR relay (1 protein)	MRE11 → MRE11/RAD50/NBS1 → DNA_REPAIR (other)

Full pathway: see Appendix B2.

Source · SIGNOR via OmniPath (CC-BY-SA) · Pruning: Northstar pathway_pruner

Appendix D

Evidence Sources

D/1 · FDA label matches

Drug–biomarker matches resolved against the FDA precision-oncology label set. Agents marked Resistance — excluded matched on biomarker but were ruled out on this patient’s molecular context; see Appendix C.

Label	Source type	Citation	Reference ID
FDA precision-oncology label match for Regorafenib	Fda Precision Oncology	FDA-LABEL	fda-regorafenib
FDA precision-oncology label match for Tucatinib	Fda Precision Oncology Resistance — excluded	FDA-LABEL	fda-tucatinib
FDA precision-oncology label match for Adagrasib	Fda Precision Oncology	FDA-LABEL	fda-adagrasib
FDA precision-oncology label match for Sotorasib	Fda Precision Oncology	FDA-LABEL	fda-sotorasib
FDA precision-oncology label match for Fam-Trastuzumab Deruxtecan-nxki	Fda Precision Oncology Resistance — excluded	FDA-LABEL	fda-fam-trastuzumab-deruxtecan-nxki
FDA precision-oncology label match for Dostarlimab-gxly	Fda Precision Oncology	FDA-LABEL	fda-dostarlimab-gxly
FDA precision-oncology label match for Pembrolizumab	Fda Precision Oncology	FDA-LABEL	fda-pembrolizumab

D/2 · Guideline sources

Tumour-type standard-of-care regimens, selected by cancer type, line and fitness rather than by biomarker.

Label	Source type	Citation	Reference ID
Standard-of-care chemotherapy guideline reference for Folinic acid + Fluorouracil + Irinotecan	Soc Chemo Guideline	NCCN Colon Cancer, Category 1;	chemo-folinic-acid-fluorouracil-irinotecan

D/3 · Literature

Peer-reviewed publications cited in this report, including those contributed by the molecular tumour board during review. Every identifier below was resolved against PubMed at the time of issue.

Label	Source type	Citation	Reference ID
Pembrolizumab in Microsatellite-Instability–High Advanced Colorectal Cancer	Literature	PMID:33264544	pmid:33264544
PD-1 Blockade in Tumors with Mismatch-Repair Deficiency	Literature	PMID:26028255	pmid:26028255
Pembrolizumab in microsatellite-instability-high and mismatch-repair-deficient advanced solid tumors: updated results of the KEYNOTE-158 trial	Literature	PMID:39979665	pmid:39979665
Overall Survival Analysis of the Phase III CodeBreaK 300 Study of Sotorasib Plus Panitumumab Versus Investigator's Choice in Chemorefractory KRAS G12C Colorectal Cancer	Literature	PMID:40215429	url:69cb9f0a1a67870d061d39461d39e59248eb8500f46cc50c8e319ee44f004788

FDA precision-oncology filter - full matched list

Drug	Matched gene	Cancer type	Status	Indication summary	Combination
Regorafenib	KRAS	Colorectal cancer	Level 1	Colorectal cancer that has spread to other parts of the body. It is used in patients who have not gotten better with other treatments.	N/A
Tucatinib	ERBB2	Colorectal cancer	Potentially applicable IHC confirmation required	treat adults whose cancer is HER2 positive , including. Colorectal cancer that cannot be removed by surgery or has spread to other parts of the body and has the wild-type RAS gene. It is used with trastuzumab in patients who have received treatment that included a fluoropyrimidine , oxaliplatin , and irinotecan hydrochloride and it did not work or is no longer working.¹ ¹This use is approved under FDA’s Accelerated Approval Program . As a condition of approval, a confirmatory trial(s) must show that tucatinib provides a clinical benefit in these patients.	trastuzumab
Adagrasib	KRAS	Colorectal cancer	Level 1	treat adults with cancer that has spread and has an abnormal KRAS gene, including. Colorectal cancer . It is used with cetuximab in patients who have received chemotherapy that included a fluoropyrimidine , oxaliplatin , and irinotecan hydrochloride . These uses are approved under FDA’s Accelerated Approval Program . As a condition of approval, a confirmatory trial(s) must show that adagrasib provides a clinical benefit in these patients.	cetuximab
Sotorasib	KRAS	Colorectal cancer	Level 1	treat adults whose cancer has an abnormal KRAS gene. It is used in. colorectal cancer that has spread to other parts of the body. It is used with panitumumab in patients who have received chemotherapy that included a fluoropyrimidine , oxaliplatin , and irinotecan hydrochloride .	panitumumab
Fam-Trastuzumab Deruxtecan-nxki	ERBB2	Solid tumors	Potentially applicable IHC confirmation required	Solid tumors that are HER2 positive and cannot be removed by surgery or have spread to other parts of the body. It is used in patients who have received systemic therapy and cannot be treated with other therapies.¹ ¹This use is approved under FDA’s Accelerated Approval Program . As a condition of approval, confirmatory trial(s) must show that fam-trastuzumab deruxtecan provides a clinical benefit in these patients.	N/A
Dostarlimab-gxly	Mismatch Repair	Solid tumors	Level 1	solid tumors that are dMMR, have spread or come back, and got worse during or after other treatment and cannot be treated with other therapies¹ ¹This use is approved under FDA’s Accelerated Approval Program . As a condition of approval, confirmatory trial(s) must show that dostarlimab-gxly provides a clinical benefit in these patients.	N/A
Pembrolizumab	Microsatellite Instability	Microsatellite instability -high (MSI-H) or mismatch repair deficient (dMMR) cancer	Level 1	Microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) cancer that has spread to other parts of the body or cannot be removed by surgery, after prior treatment	N/A

Sources: FDA precision-oncology filter table · cancer.gov / fda.gov approvals.

Appendix E

Clinical trials

Strongly matched

Total matched

Recruiting now

Within 50 mi

The clinical trials below may require further characterisation of the patient's tumour to ensure eligibility. Please refer to the full eligibility criteria contained within the source hyperlink to confirm a patient match before sharing with the patient.

Biomarker-matched, excluded on RAS status (1)

American Society of Clinical Oncology · RECRUITING · Updated 29 May 2026

Board caveat — ERBB2 arm not recommended

The ERBB2 cohort in this registry treats with pertuzumab and trastuzumab — antibody-mediated HER2 signalling blockade. This patient’s KRAS p.G12C sits downstream of the receptor and sustains MAPK signalling irrespective of HER2 inhibition, which is why HER2-directed therapy in colorectal cancer is restricted to RAS wild-type disease (see Section 2A). The board therefore does not recommend enrolment on the ERBB2 arm. The registry remains open on other cohorts and is retained here for that reason.

The purpose of the study is to learn from the real world practice of prescribing targeted therapies to patients with advanced cancer whose tumor harbors a genomic variant known to be a drug target or to predict sensitivity to a drug. NOTE: Due to character limits, the arms section does NOT include all TAPUR Study relevant biomarkers. For additional information, contact TAPUR@asco.org, or if a patient, your nearest participating TAPUR site (see participating centers)…

Matched on: ERBB2 — board caveat, see below

Source: clinicaltrials.gov

Eligibility match: CIViC genomic-marker match

Nearest site: Windham Hospital (WH) · 47 mi (United States)

All sites: 181

Broader / cohort · tumour-type or text match (7)

Merus B.V. · RECRUITING · Updated 29 Jul 2026

This is a Phase 1/2 open-label, multi-center, multi-national study with an initial dose escalation part to determine the recommended Phase II dose (RP2D) of MCLA-158 single agent in patients with mCRC. The dose escalation part has been completed and the RP2D will be further evaluated in an expansion part of the study. Cohorts of selected solid tumor indications for which there is evidence of EGFR dependency and potential sensitivity to EGFR inhibition will be evaluated including head and neck…

Matched on: ERBB2, KRAS, ERBB2 amplification

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Massachusetts General Hospital - Dana Farber · 50 mi (United States)

All sites: 54

Revolution Medicines, Inc. · RECRUITING · Updated 15 May 2026

Evaluate the safety and tolerability of RMC-6236 in adults with specific RAS mutant advanced solid tumors.

Matched on: KRAS

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Dana Farber Cancer Institute · 50 mi (United States)

All sites: 21

Institut de Recherches Internationales Servier · RECRUITING · Updated 17 Jun 2026

BDTX-4933-101 is a first-in-human, open-label, Phase 1/2 dose escalation, dose optimization and expansion study designed to evaluate the safety and tolerability of S241656 as monotherapy and in combination with other anti-cancer therapies in participants with selected advanced malignancies. The study population for the Dose Escalation part of the study comprises adults with recurrent advanced/metastatic non-small cell lung cancer (NSCLC), Gastrointestinal (GI) cancers, and other solid tumors…

Matched on: KRAS, KRAS G12C

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Dana-Farber Cancer Institute · 50 mi (United States)

All sites: 27

Parabilis Medicines, Inc. · RECRUITING · Updated 23 Jul 2026

The goal of this clinical trial is to determine if FOG-001 is safe and effective in participants with locally advanced or metastatic solid tumors or in participants with familial adenomatous polyposis (FAP).

Matched on: APC

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Massachusetts General Hospital · 50 mi (United States)

All sites: 33

Revolution Medicines, Inc. · RECRUITING · Updated 3 Aug 2026

This study is to evaluate the safety, tolerability, and PK profiles of Elironrasib and Daraxonrasib as monotherapies and combination therapy in patients with KRAS G12C-mutated solid tumors.

Matched on: KRAS, KRAS G12C

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Dana Farber Cancer Institute · 50 mi (United States)

All sites: 55

Frontier Medicines Corporation · RECRUITING · Updated 18 Mar 2026

The goal of this clinical trial is to evaluate FMC-376 in participants with advanced solid tumors with KRAS G12C mutations. This clinical trial will be conducted in 3 parts: Phase 1A (Dose Escalation), Phase 1B (Dose Expansion), and Phase 2 (Cohort Expansion). Multiple dose levels in participants with advanced solid tumors will be evaluated.

Matched on: KRAS, KRAS G12C

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Dana-Farber Cancer Institute · 50 mi (United States)

All sites: 26

Boehringer Ingelheim · RECRUITING · Updated 6 Aug 2026

This study is open to adults aged 18 years and older with different types of HER2+ cancer that has spread and cannot be removed by surgery. People can take part in this study if their tumours show HER2 aberrations and previous treatment was not successful. The purpose of this study is to find a suitable dose of zongertinib that people with different types of HER2+ cancer that has spread can tolerate best when taken together with trastuzumab deruxtecan (T-DXd), with trastuzumab emtansine…

Matched on: ERBB2, ERBB2 amplification

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Dana-Farber Cancer Institute · 50 mi (United States)

All sites: 105

▶ International trials (2 trials - sites outside your selected country)

These trials have their nearest enrolling site outside the patient's country. They meet the same eligibility criteria but are shown separately to help prioritise locally accessible options.

Chipscreen Biosciences, Ltd. · RECRUITING · Updated 19 Mar 2026

A randomised, open-label, multicenter phase III study to evaluate the efficacy and safety of tucidinostat in combination with sintilimab and bevacizumab versus fruquintinib monotherapy in MSS/pMMR colorectal cancer patients.

Matched on: KRAS

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Rui-Hua Xu · 7899 mi (China)

All sites: 1

Akeso · RECRUITING · Updated 22 Apr 2026

This is a randomized, open-label, controlled, multicenter phase 3 study. All patients are resectable microsatellite instability-high (MSI-H) or mismatch repair-deficient (dMMR) colon cancer. The purpose of this study is to evaluate the efficacy and safety of neoadjuvant/adjuvant treatment of AK104 (Cadonilimab) versus adjuvant chemotherapy in patients with resectable MSI-H/dMMR colon cancer.

Matched on: MSH2

Source: clinicaltrials.gov

Eligibility match: Text match based on eligibility section

Nearest site: Sun Yat-sen University Cancer Center · 7899 mi (China)

All sites: 1

ClinicalTrials.gov v2 · genomic eligibility from JAX-CKB and CIViC

Confidence levels: High - curated genomic-eligibility evidence from JAX-CKB or CIViC with a direct variant/therapy match and active recruiting status. Medium - evidence from free-text regex matching or off-label extrapolation. Low - indirect biomarker or disease-type match only; manual review recommended. Distance is haversine from the patient's geocoded postal code to the nearest enrolling site.

Appendix F

Dosing · PGx · DDI

Immunotherapy biomarkers

Biomarker	Value	Interpretation	Status
TMB	41.0 mut/Mb	High	Assessed
MSI / MMR	MSI-H	MSI-High	Assessed
HRD	Positive (gLOH 19.4%)	Elevated — not validated in colorectal cancer; confounded by dMMR	Assessed
PD-L1	CPS 8 (22C3)	Positive by CPS, but not a gating biomarker — PD-L1 expression does not determine checkpoint-inhibitor eligibility in MSI-H colorectal cancer, where MSI status is the approval criterion. Supportive context only.	Assessed

Dosing - FDA prescribing summary

Dosing information is retrieved live from the FDA drug label (openFDA) at report generation time and reflects the current prescribing information for each matched agent at time of the report's publication. Data are extracted programmatically; always consult the full prescribing information before prescribing.

Drug	Route	Dosage forms & strengths	Recommended dose / regimen	Key dose adjustments	Approved indication (summary)
Erbitux (cetuximab) ImClone LLC BLA125084 ↗	Intravenous	Injection: 100 mg/50 mL (2 mg/mL) or 200 mg/100 mL (2 mg/mL) as a clear, colorless solution in a single-dose vial. Injection: 100 mg/50 mL (2 mg/mL) or 200 mg/100 mL (2 mg/mL) in a single-dose vial.	400 mg/m 2 administered as a 120-minute intravenous infusion one week prior to initiating a course of radiation therapy.	N/A	Erbitux ® is an epidermal growth factor receptor (Egfr) antagonist indicated for treatment of: Head and Neck Cancer Locally or regionally advanced squamous cell carcinoma of the head and neck in combination with radiation therapy. ( 1.1, 14.1 ) Recurrent locoregional disease or metastatic squamous cell carcinoma of the head and neck in combination with platinum-based therapy with fluorouracil. ( 1.1, 14.1 ) Recurrent or metastatic squamous cell carcinoma of the head and neck progressing after platinum-based therapy. ( 1.1, 14.1 ) Colorectal Cancer K-Ras wild-type, Egfr-expressing, metastatic colorectal cancer as determined by an Fda-approved test in combination with Folfiri for first-line treatment, in combination with irinotecan in patients who are refractory to irinotecan-based chemotherapy, as a single-agent in patients who have failed oxaliplatin- and irinotecan-based chemotherapy or who are intolerant to irinotecan. ( 1.2, 5.7, 12.1, 14.2 ) Limitations of Use: Erbitux is not indicated for treatment of Ras-mutant colorectal cancer or when the results of the Ras mutation tests are unknown. Braf V600E Mutation-Positive Metastatic Colorectal Cancer (Crc) in combination with encorafenib, for the treatment of adult patients with metastatic colorectal cancer (Crc) with a Braf V600E mutation, as detected by an Fda-approved test, after prior therapy. 1.1 Squamous Cell Carcinoma of the Head and Neck (Scchn) Erbitux ® is indicated: in combination with radiation therapy for the initial treatment of locally or regionally advanced squamous cell carcinoma of the head and neck (Scchn). in combination with platinum-based therapy with fluorouracil for the first-line treatment of patients with recurrent locoregional disease or metastatic Scchn. as a single-agent for the treatment of patients with recurrent or metastatic Scchn for whom prior platinum-based therapy has failed. Label date: 2026-04-16
Keytruda (pembrolizumab) Merck Sharp & Dohme LLC BLA125514 ↗	Intravenous	Injection: 100 mg/4 mL (25 mg/mL) solution in a single-dose vial.	200 mg every 3 weeks or 400 mg every 6 weeks by intravenous infusion over 30 minutes, until disease progression or unacceptable toxicity.	Reference IV product for the systemic oncology indication. The subcutaneous co-formulation (pembrolizumab and berahyaluronidase alfa, BLA 761467) is a different application with different dosing and is not the product referenced by this roadmap.
Krazati (adagrasib) Mirati Therapeutics, Inc NDA216340 ↗	Oral	Tablets: 200 mg, oval shaped, white to off-white, immediate release film coated tablets with "200" on one side and stylized "M" on the opposite side. Tablets: 200 mg.	600 mg orally twice daily	dose reductions for adverse reactions for use of KRAZATI as a single agent or in combination with cetuximab are outlined in Table 1. If adverse reactions occur, a maximum of two dose reductions are permitted. Permanently discontinue KRAZATI in patients who are unable to tolerate 600 mg once daily. Table 1: Recomme · Dose Reduction Dosage First dose reduction 400 mg twice daily Second dose reduction 600 mg once daily Refer to the cetuximab prescribing information for dose modifications for adverse reactions associated with cetuximab.	Krazati is an inhibitor of the Ras GTPase family indicated for: Non-small cell lung cancer (Nsclc)* • As a single agent, for the treatment of adult patients with Kras G12C-mutated locally advanced or metastatic Nsclc, as determined by an Fda-approved test, who have received at least one prior systemic therapy. Colorectal cancer (Crc)* • In combination with cetuximab, for the treatment of adult patients with Kras G12C-mutated locally advanced or metastatic Crc, as determined by an Fda-approved test, who have received prior treatment with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy. *These indications are approved under accelerated approval based on objective response rate (Orr) and duration of response (Dor). Continued approval for these indications may be contingent upon verification and description of a clinical benefit in confirmatory trials. ( 1.1, 1.2 ) 1.1 Kras G12C-Mutated Locally Advanced or Metastatic Non-Small Cell Lung Cancer Krazati, as a single-agent, is indicated for the treatment of adult patients with Kras G12C-mutated locally advanced or metastatic non-small cell lung cancer (Nsclc), as determined by an Fda-approved test, who have received at least one prior systemic therapy. This indication is approved under accelerated approval based on objective response rate (Orr) and duration of response (Dor). Continued approval for this indication may be contingent upon verification and description of a clinical benefit in a confirmatory trial. 1.2 Kras G12C-Mutated Locally Advanced or Metastatic Colorectal Cancer Krazati in combination with cetuximab is indicated for the treatment of adult patients with Kras G12C-mutated locally advanced or metastatic colorectal cancer (Crc), as determined by an Fda-approved test, who have received prior treatment with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy. This indication is approved under accelerated approval based on Orr and Dor. Continued approval for this indication may be contingent upon verification and description of a clinical benefit in a confirmatory trial. Label date: 2026-03-25
Lumakras (sotorasib) Amgen Inc NDA214665 ↗	Oral	Tablets: 320 mg, beige, oval-shaped, immediate release, film-coated, debossed with "Amg" on one side and "320" on the opposite side. Tablets: 240 mg, yellow, oval-shaped, immediate release, film-coated, debossed with "Amg" on one side and "240" on the opposite side. Tablets: 120 mg, yellow, oblong-shaped, immediate release, film-coated, debossed with "Amg" on one side and "120" on the opposite side. Tablets: 320 mg, 240 mg, 120 mg.	960 mg (three 320 mg tablets or four 240 mg tablets or eight 120 mg tablets) orally once daily until disease progression or unacceptable toxicity…	dose reduction levels are summarized in Table 1. If adverse reactions occur, a maximum of two dose reductions are permitted. Discontinue LUMAKRAS if patients are unable to tolerate the minimum dose of 240 mg once daily. When LUMAKRAS is administered in combination with panitumumab, and LUMAKRAS is temporarily wit · permanently discontinued, temporarily withhold or permanently discontinue panitumumab, respectively. Refer to the full prescribing information of panitumumab for dose modifications for adverse reactions associated with the use of panitumumab.	Lumakras is an inhibitor of the Ras GTPase family indicated for: Kras G12C-mutated Locally Advanced or Metastatic Non-Small Cell Lung Cancer (Nsclc) As a single agent, for the treatment of adult patients with Kras G12C-mutated locally advanced or metastatic Nsclc, as determined by an Fda-approved test, who have received at least one prior systemic therapy. This indication is approved under accelerated approval based on overall response rate (Orr) and duration of response (Dor). Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s). Kras G12C-mutated Metastatic Colorectal Cancer (mCRC) In combination with panitumumab, for the treatment of adult patients with Kras G12C-mutated mCRC as determined by an Fda approved-test, who have received prior fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy. 1.1 Kras G12C-mutated Locally Advanced or Metastatic Non-Small Cell Lung Cancer (Nsclc) Lumakras as a single agent is indicated for the treatment of adult patients with Kras G12C -mutated locally advanced or metastatic non-small cell lung cancer (Nsclc), as determined by an Fda-approved test, who have received at least one prior systemic therapy. This indication is approved under accelerated approval based on overall response rate (Orr) and duration of response (Dor). Continued approval for this indication may be contingent upon verification and description of clinical benefit in a confirmatory trial(s). 1.2 Kras G12C-mutated Metastatic Colorectal Cancer (mCRC) Lumakras, in combination with panitumumab, is indicated for the treatment of adult patients with Kras G12C -mutated metastatic colorectal cancer (mCRC), as determined by an Fda-approved test, who have received prior fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy. Label date: 2025-01-22
Stivarga (regorafenib) Bayer HealthCare Pharmaceuticals Inc. NDA203085 ↗	Oral	Stivarga is a 40 mg, light pink, oval-shaped, film-coated tablet, debossed with 'Bayer' on one side and '40' on the other side.	160 mg STIVARGA (four 40 mg tablets) taken orally once daily for the first 21 days of each 28-day cycle…	Dose Modifications If dose modifications are required, reduce the dose in 40 mg (one tablet) increments; the lowest recommended daily dose of STIVARGA is 80 mg daily.	Stivarga is a kinase inhibitor indicated for the treatment of adult patients with: • Metastatic colorectal cancer (Crc) who have been previously treated with fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy, an anti-Vegf therapy, and, if Ras wild-type, an anti-Egfr therapy. • Locally advanced, unresectable or metastatic gastrointestinal stromal tumor (Gist) who have been previously treated with imatinib mesylate and sunitinib malate. • Hepatocellular carcinoma (Hcc) who have been previously treated with sorafenib 1.1 Colorectal Cancer Stivarga is indicated for the treatment of adult patients with metastatic colorectal cancer (Crc) who have been previously treated with fluoropyrimidine-, oxaliplatin- and irinotecan-based chemotherapy, an anti-Vegf therapy, and, if Ras wild-type, an anti-Egfr therapy. 1.2 Gastrointestinal Stromal Tumors Stivarga is indicated for the treatment of adult patients with locally advanced, unresectable or metastatic gastrointestinal stromal tumor (Gist) who have been previously treated with imatinib mesylate and sunitinib malate. 1.3 Hepatocellular Carcinoma Stivarga is indicated for the treatment of adult patients with hepatocellular carcinoma (Hcc) who have been previously treated with sorafenib. Label date: 2026-02-09
Vectibix (panitumumab) Amgen, Inc BLA125147 ↗	Intravenous	Injection: 100 mg/5 mL (20 mg/mL) colorless solution in single-dose vial. Injection: 400 mg/20 mL (20 mg/mL) colorless solution in single-dose vial. Injection: 100 mg/5 mL (20 mg/mL) and 400 mg/20 mL (20 mg/mL) in single-dose vials.	6 mg/kg, administered as an intravenous infusion every 14 days until disease progression or unacceptable toxicity [see Dosage and Administration (2…	Dose Modifications Dose Modifications for Vectibix in Combination with Sotorasib When Vectibix is administered in combination with sotorasib, if treatment with sotorasib is temporarily withheld or permanently discontinued, temporarily withhold or permanently discontinue Vectibix, respectively Reduce infusion rate by 50% in patients experiencing a mild or…	Vectibix is an epidermal growth factor receptor (Egfr) antagonist indicated for the treatment of: Adult patients with wild-type Ras (defined as wild-type in both Kras and Nras as determined by an Fda-approved test) Metastatic Colorectal Cancer (mCRC): In combination with Folfox for first-line treatment. ( 1, 14.2 ) As monotherapy following disease progression after prior treatment with fluoropyrimidine, oxaliplatin, and irinotecan-containing chemotherapy. ( 1, 14.1 ) Kras G12C-mutated Metastatic Colorectal Cancer (mCRC) In combination with sotorasib, for the treatment of adult patients with Kras G12C- mutated mCRC, as determined by an Fda-approved test, who have received prior treatment with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy. *Limitations of Use: Vectibix is not indicated for the treatment of patients with Ras -mutant mCRC unless used in combination with sotorasib in Kras G12C-mutated mCRC. Vectibix is not indicated for the treatment of patients with mCRC for whom Ras mutation status is unknown ( 1, 2.1, 5.2, 12.1, 14.3 ). Metastatic Colorectal Cancer (mCRC) Ras Wild-Type mCRC Vectibix is indicated for the treatment of adult patients with wild-type Ras (defined as wild-type in both Kras and Nras as determined by an Fda-approved test) metastatic colorectal cancer (mCRC): As first-line therapy in combination with Folfox. As monotherapy following disease progression after prior treatment with fluoropyrimidine-, oxaliplatin-, and irinotecan-containing chemotherapy. Kras G12C -mutated mCRC Vectibix, in combination with sotorasib, is indicated for the treatment of adult patients with Kras G12C- mutated mCRC, as determined by an Fda-approved test, who have received prior treatment with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy. Limitations of Use: Vectibix is not indicated for the treatment of patients with Ras -mutant mCRC unless used in combination with sotorasib in Kras G12C-mutated mCRC. Vectibix is not indicated for the treatment of patients with mCRC for whom Ras mutation status is unknown. Label date: 2026-06-23
Dostarlimab-Gxly	N/A	-	FDA label not available - consult prescribing information directly.	N/A	N/A

Route key: Oraltaken by mouth · IntravenousIV infusion · SubcutaneousSC injection · Inhalationinhaled

Source: FDA Drug Label API (openFDA · api.fda.gov/drug/label.json), retrieved at report generation time. Dose extraction is automated from free-text label sections; minor formatting variations may occur. This table is an aide-mémoire only - verify against the full Prescribing Information (PI) before clinical use. Combination regimens: doses shown are per-agent; refer to the combination PI for scheduling and sequencing guidance.

Pharmacogenomics

No pharmacogenomic findings reported for this case.

Drug interactions

Over-the-counter interactions

Drug A	OTC Drug	Severity	Description	Sources
Sotorasib	omeprazole	high	Acid-Reducing Agents: Avoid coadministration with proton pump inhibitors (PPIs) and H2 receptor antagonists.	fda_label
Sotorasib	pantoprazole	high	Acid-Reducing Agents: Avoid coadministration with proton pump inhibitors (PPIs) and H2 receptor antagonists.	fda_label
Sotorasib	esomeprazole	high	Acid-Reducing Agents: Avoid coadministration with proton pump inhibitors (PPIs) and H2 receptor antagonists.	fda_label
Sotorasib	calcium carbonate	high	If an acid-reducing agent cannot be avoided, administer LUMAKRAS 4 hours before or 10 hours after a local antacid.	fda_label
Sotorasib	magnesium hydroxide	high	If an acid-reducing agent cannot be avoided, administer LUMAKRAS 4 hours before or 10 hours after a local antacid.	fda_label
Sotorasib	aluminum hydroxide	high	If an acid-reducing agent cannot be avoided, administer LUMAKRAS 4 hours before or 10 hours after a local antacid.	fda_label
Regorafenib	st john's wort	high	• Strong CYP3A4 inducers: Avoid strong CYP3A4 inducers. • Strong CYP3A4 inhibitors: Avoid strong CYP3A4 inhibitors. • BCRP substrates: Monitor patients closely for symptoms of increased exposure to BCRP substrates.	fda_label

Drugs checked: Adagrasib, Dostarlimab-gxly, Fam-Trastuzumab Deruxtecan-nxki, Pembrolizumab, Sotorasib, Tucatinib, Regorafenib. Interaction checks cover the proposed drug set; verify against the final treatment plan. Consult a clinical pharmacist before prescribing.

Appendix G

Methodology - About This Document

Document status

This document is a clinical consultation issued by the named signing clinician. Treatment decisions are made by the treating oncology team in consultation with the patient.

Purpose

Clinician-facing molecular interpretation report summarising actionable molecular alterations, pathway context, hallmark impacts, and matched clinical trials for the case identified above.

Data sources

CIViC (ClinicalTrials.gov Interpreting Variants in Cancer) · cancerhotspots.org v2 · SIGNOR via OmniPath (pathway topology) · ClinicalTrials.gov v2 API · JAX-CKB (genomic eligibility curation).

Tier framework

AMP/ASCO/CAP 2017 (Li et al., J Mol Diagn 19:4). Tier I: strong clinical significance (FDA-approved or guideline-listed for this cancer type). Tier II: potential clinical significance. Tier III: unknown clinical significance.

Validators applied

Synthesis vocabulary constrained to gene symbols, drug names, tier labels, hallmark identifiers, germline gene names, immunotherapy biomarkers, PGx genes, and trial IDs present in the structured evidence layer. FDA approval claims are cross-checked against the curated FDA precision oncology filter for the patient's disease context.

Contact

Astron Health · support@astron.health
Padman Vamadevan · padman@astron.health

Reference data

CIViC	Snapshot 2024-12
cancerhotspots.org	cancerhotspots v2
PharmCAT	PharmCAT v2.15
ClinicalTrials.gov	v2 API - queried at generation
SIGNOR / OmniPath	Queried at generation time (CC-BY-SA)

Appendix G · Caveats

Notes and caveats

Therapeutic group placement

The Section 2 grouped therapeutic display places drugs as follows: FDA precision-oncology matches without a confirmatory biomarker requirement are placed in the standard-of-care group; matches with pending IHC confirmation are placed in the biomarker-supported group. Investigational therapy placement is based on genomic eligibility criteria; contraindicated placement is based on curated resistance data.

Variant detection

Variant calls are derived from the source molecular report. Northstar normalises labels to HGVS-style notation but does not re-call variants. Pathogenicity classifications are evidence-based and may differ from clinical germline interpretation.

Copy number

Copy-number estimates depend on tumour purity and ploidy assumptions in the upstream assay. Amplification calls (≥ 6 copies) are considered high-level; focal deletions (0 copies) are considered homozygous. Mosaic events may be underestimated.

Fusions

Fusion detection sensitivity varies by assay design. RNA-based assays detect expressed fusions; DNA-only panels infer from structural breakpoints and may miss intronic events or produce false positives in regions of low coverage.

Tumour content

Variant allele fractions reflect the fraction of sequenced molecules carrying the alteration; they are diluted by normal stromal contamination. The tumour-content estimate in the patient meta grid is sourced directly from the upstream report and has not been independently verified.

Germline filtering

Germline variants are filtered using population-frequency databases. A small fraction of hereditary variants may escape filtering if they are absent from the current databases. Genetic counselling is advised for variants in established hereditary cancer genes (BRCA1/2, MLH1, etc.).

Germline flagging methodology

The germline banner above Section 1 flags variants that are (1) in an NCCN-recognised cancer predisposition gene, (2) classified Pathogenic / Likely pathogenic by ClinVar or located at a known germline hot-spot residue, and (3) at a VAF consistent with germline heterozygosity given the tumour purity (default window 38–62 % at purity ≥ 50 %). The threshold is intentionally conservative - false positives are resolved cheaply by confirmatory germline testing. ClinVar snapshot date: queried live at report generation time via NCBI e-utilities. Cancer predisposition gene list: NCCN-recognised list (current at report generation).

Immunotherapy biomarker provenance

Section 4 surfaces four tumour-agnostic IO decision drivers. TMB is computed from nonsynonymous + synonymous coding mutations passing standard QC, divided by the callable panel size; the clinical classification threshold (default ≥ 10 mut/Mb = TMB-High) follows FDA approval of pembrolizumab (KEYNOTE-158) with per-Oncotree-code overrides where published evidence supports a different cut-off. MSI is classified from the upstream caller score (MSIsensor-pro / mSINGS convention): MSI-H ≥ 20 %, MSS otherwise. HRD is computed via the scarHRD genomic-scar composite (LOH + LST + TAI) when segmented copy-number and B-allele frequency data are available; see the HRD scoring methodology note below. Rendered as "Not assessed" when segmented copy-number and BAF data are unavailable from the upstream assay. PD-L1 expression is reported from immunohistochemistry on a separate specimen; entered manually by the ordering team. Interpretation thresholds are companion-diagnostic-specific and tumour-type-specific. Results may differ between clones - interpret in the context of the assay used. In oncogene-driven tumours (EGFR / ALK / ROS1 / RET / MET exon 14 / BRAF V600E NSCLC), targeted therapy generally takes precedence over PD-L1-gated IO monotherapy regardless of expression level; this is flagged with an asterisk (*) on the class chip when applicable. Rendered as "Not assessed" when no result has been submitted.

HRD scoring methodology

HRD score is computed via scarHRD (Sztupinszki et al., NPJ Breast Cancer 2018; GPL-3 R package), which sums three genomic-scar components derived from segmented copy-number and B-allele frequency data:

LOH - Loss of heterozygosity (Abkevich et al. 2012)
LST - Large-scale state transitions (Popova et al. 2012)
TAI - Telomeric allelic imbalance (Birkbak et al. 2012)

The composite HRD score = LOH + LST + TAI (0–100+ integer scale). Threshold ≥ 42 = HRD-positive, following the Myriad MyChoice CDx convention used in PARP inhibitor trial enrolment (SOLO-1, PAOLA-1, PRIMA, PROfound).

Important caveats: The HRD score is a measure of accumulated genomic instability (genomic scarring). It persists even after restoration of HR function (e.g. following BRCA reversion mutation), because the scars from past HR deficiency are irreversible. Interpret in full clinical context; a high score does not guarantee ongoing HR deficiency at the time of treatment. This method is appropriate for panel-based CGP assays that produce segmented CN and BAF data. It does not use mutational signature SBS3 (requires WGS) or structural variant calls (required by CHORD).

Pharmacogenomics caller (Drug metabolism section)

The Drug metabolism considerations section is generated by PharmCAT (pharmcat.org, MPL-2.0 licence) operating on a germline-quality VCF restricted to the PharmCAT expected positions via the PharmCAT preprocessor script. Genes in the panel but absent from the input VCF or with insufficient coverage are reported as "Not assessed". PharmCAT version: reported per-run in the source column. These calls reflect inherited star-allele metaboliser phenotypes and should not be conflated with the somatic targeted therapy recommendations in the Therapeutic implications section. Consult CPIC guidelines (cpicpgx.org) for dosing recommendations by phenotype.

Coverage QC

Regions of low coverage (< 100× mean) may have reduced sensitivity for somatic variant calling. Coverage statistics are reported by the source assay; Northstar does not perform independent QC beyond what is provided in the source report.

Database currency

CIViC snapshot: 2024-12. cancerhotspots.org: cancerhotspots v2. SIGNOR via OmniPath: queried at report generation time. ClinicalTrials.gov v2: trial records snapshotted per TrialProvenance.snapshot_date. JAX-CKB: eligibility curation snapshot recorded per trial.

Trial eligibility

Matched trials reflect genomic eligibility only. Clinical eligibility (performance status, prior therapy, organ function, exclusion criteria) must be assessed by the treating clinician. Trial availability and recruitment status change frequently; verify at ClinicalTrials.gov before referral.

Synthesis constraints

Mechanistic descriptions in the clinician synthesis (Section 1) should be verified against primary literature before use in a clinical decision.

Tier framework

Tiers follow AMP/ASCO/CAP 2017 (Li et al., J Mol Diagn 19:4). Tier I: strong clinical significance (FDA-approved or guideline-listed for this cancer type). Tier II: potential clinical significance. Tier III: unknown clinical significance.

Geographic distance

Site distances are computed using the haversine formula from the patient's geocoded postal code to the nearest enrolling site's coordinates. Geocoding precision depends on the specificity of the patient location provided; distances should be treated as approximate.

Limitations

This report does not capture spatial or clonal heterogeneity, tumour evolution under therapy, drug–drug interactions, or pharmacokinetic / pharmacodynamic considerations. Interpretation of rare variants or variants in understudied cancer types should be confirmed by an expert molecular tumour board.

Appendix B2

Full Pathway Topology

Unpruned pathway maps for audit and reference. The strategic-summary maps in Appendix B collapse linear non-anchor chains into named modules; this appendix shows every node and edge as supplied by the upstream graph.

Source · SIGNOR via OmniPath (CC-BY-SA)