← Adverse-Event Coding & Causality
Clinical / Pharmacovigilance · Drug-safety physician (PV)

Adverse-Event Coding & Causality — agent trajectory for expert review

claude-opus-4-8 · 2026-06-06T09:15:30.595731+00:00 · 7 tool calls · 27.2s · 10028 tokens
run id: ae_causality_20260606_021404_03_review-case-icsr-003-a-patient-on-a-her2-antibod

Query

Review case ICSR-003 (a patient on a HER2 antibody-drug conjugate with abnormal liver labs). Code the events to PT/SOC, assess seriousness, and carefully assess drug causality given the patient's concomitant medications and history.

Trajectory

Agent · turn 1

I'll review this case systematically. Let me start by pulling the case details and the relevant drug label.

⚙ calls get_case(case_id=ICSR-003)
⚙ calls get_drug_label(drug_class=HER2 antibody-drug conjugate (ADC))
↳ get_case result
show data
{
  "case_id": "ICSR-003",
  "study_drug": "Trastuzumab deruxtecan (T-DXd)",
  "drug_class": "HER2 antibody-drug conjugate",
  "patient": {
    "age": 58,
    "sex": "F",
    "comorbidities": [
      "HER2+ metastatic breast cancer",
      "known gallstones"
    ],
    "concomitant_meds": [
      "atorvastatin",
      "amoxicillin/clavulanate (started 5 days ago for a dental abscess)",
      "acetaminophen"
    ]
  },
  "narrative": "58-year-old woman on a HER2 antibody-drug conjugate was found on routine labs to have elevated liver enzymes (ALT 4x ULN, AST 3x ULN) with a mildly raised bilirubin. She had started amoxicillin/clavulanate five days earlier for a dental abscess and has known gallstones; a right-upper-quadrant ultrasound showed cholelithiasis without duct dilatation. She was asymptomatic, not hospitalized; the antibiotic was stopped and liver enzymes trended down over the following week without changing the study drug.",
  "events": [
    "elevated liver enzymes (ALT/AST)",
    "raised bilirubin"
  ]
}
↳ get_drug_label result
show data
{
  "drug_class": "HER2 antibody-drug conjugate",
  "examples": [
    "Trastuzumab deruxtecan (T-DXd)"
  ],
  "labeled": [
    "Interstitial lung disease",
    "Pneumonitis",
    "Neutropenia",
    "Anaemia",
    "Nausea",
    "Vomiting",
    "Fatigue",
    "Alanine aminotransferase increased"
  ],
  "boxed_warning": "Interstitial lung disease / pneumonitis (can be fatal) and embryo-fetal toxicity.",
  "note": "ILD/pneumonitis is a labeled, potentially fatal class risk; myelosuppression and GI toxicity are common. Isolated transaminase rise has many alternative causes."
}
Agent · turn 2

Let me code the events properly using the MedDRA dictionary.

⚙ calls meddra_lookup(term=elevated ALT)
⚙ calls meddra_lookup(term=elevated AST)
⚙ calls meddra_lookup(term=raised bilirubin)
↳ meddra_lookup result
show data
{
  "query": "elevated ALT",
  "count": 1,
  "matches": [
    {
      "pt": "Alanine aminotransferase increased",
      "soc": "Investigations",
      "synonyms": [
        "elevated liver enzymes",
        "high alt",
        "alt elevation",
        "transaminitis",
        "elevated alt",
        "raised liver enzymes",
        "elevated transaminases"
      ]
    }
  ]
}
↳ meddra_lookup result
show data
{
  "query": "elevated AST",
  "count": 1,
  "matches": [
    {
      "pt": "Aspartate aminotransferase increased",
      "soc": "Investigations",
      "synonyms": [
        "high ast",
        "ast elevation",
        "elevated ast"
      ]
    }
  ]
}
↳ meddra_lookup result
show data
{
  "query": "raised bilirubin",
  "count": 0,
  "matches": []
}
Agent · turn 3

Bilirubin didn't match. Let me try alternate lay/clinical terms.

⚙ calls meddra_lookup(term=bilirubin)
⚙ calls meddra_lookup(term=hyperbilirubinaemia)
↳ meddra_lookup result
show data
{
  "query": "bilirubin",
  "count": 1,
  "matches": [
    {
      "pt": "Blood bilirubin increased",
      "soc": "Investigations",
      "synonyms": [
        "elevated bilirubin",
        "high bilirubin",
        "hyperbilirubinaemia",
        "hyperbilirubinemia"
      ]
    }
  ]
}
↳ meddra_lookup result
show data
{
  "query": "hyperbilirubinaemia",
  "count": 1,
  "matches": [
    {
      "pt": "Blood bilirubin increased",
      "soc": "Investigations",
      "synonyms": [
        "elevated bilirubin",
        "high bilirubin",
        "hyperbilirubinaemia",
        "hyperbilirubinemia"
      ]
    }
  ]
}
Agent · turn 4

I now have everything needed for a complete assessment.

ICSR-003 — Case Review

Study drug: Trastuzumab deruxtecan (T-DXd) | Class: HER2 antibody-drug conjugate (ADC)
Patient: 58F, HER2+ metastatic breast cancer, known gallstones.


1. Event Coding (PT / SOC)

Reported term MedDRA PT SOC
Elevated liver enzymes (ALT 4× ULN) Alanine aminotransferase increased Investigations
Elevated AST (3× ULN) Aspartate aminotransferase increased Investigations
Mildly raised bilirubin Blood bilirubin increased Investigations

Note: These are coded as lab investigation PTs. The pattern (ALT/AST predominant, mildly raised bilirubin) does not meet Hy's Law (no documented bilirubin >2× ULN with transaminases >3× ULN in the absence of cholestasis/alternative cause), and a clinical diagnosis term (e.g., "Drug-induced liver injury") is not warranted on the available data. Coding to the discrete lab abnormalities is the appropriate, non-over-interpreting choice.


2. Seriousness Assessment

Not serious.

Testing against the standard ICH E2A seriousness criteria:
- Death — No
- Life-threatening — No
- Hospitalization / prolongation — No (explicitly not hospitalized)
- Persistent/significant disability — No
- Congenital anomaly — No
- Other medically important condition — Considered, but argues against: patient asymptomatic, modest enzyme elevations, no synthetic dysfunction, no Hy's Law, resolved spontaneously within a week without intervention or study-drug modification.

Non-serious adverse event.


3. Causality Assessment (drug = T-DXd)

Assessed causality to study drug: Unlikely / Not related.

The label confirms Alanine aminotransferase increased is a labeled (expected) reaction for this class, so T-DXd is biologically plausible. However, the totality of evidence points strongly to a confounder rather than the study drug:

Evidence against T-DXd causality:
- Strong competing cause — amoxicillin/clavulanate. Co-amoxiclav is one of the most common causes of drug-induced liver injury, with onset typically days to a few weeks after starting. The timing here (started 5 days before the abnormality) fits classically.
- Positive dechallenge of the confounder, not the study drug. The antibiotic was stopped and enzymes trended down, while T-DXd was continued unchanged. Improvement despite ongoing study-drug exposure is powerful evidence against T-DXd as the cause.
- Second competing cause — biliary disease. Known gallstones with cholelithiasis on ultrasound could contribute to the mildly raised bilirubin (though no duct dilatation was seen).
- Possible additional contributors: atorvastatin (statin transaminitis) and acetaminophen are also on board.
- No de-/re-challenge of T-DXd was performed (and none was needed), so no positive temporal evidence ties the event to the ADC.

Conclusion: Despite ALT elevation being label-expected for HER2 ADCs, the clinical course — temporal link to co-amoxiclav, resolution on stopping the antibiotic while the study drug continued unchanged, plus underlying gallstones — makes the study drug an unlikely cause. Most probable cause: amoxicillin/clavulanate-related hepatotoxicity, with possible biliary contribution.


Summary Line

58F on T-DXd with ALT increased / AST increased / Blood bilirubin increased (all SOC Investigations); non-serious; causality to study drug Unlikely — confounded by recent amoxicillin/clavulanate (classic timing + positive dechallenge while T-DXd continued) and known cholelithiasis.

Recommended follow-up (data-quality): capture baseline LFTs, ULN values, exact bilirubin level (to formally exclude Hy's Law), hepatitis serologies/alcohol history if available, and confirm enzyme normalization — to firm up the differential, though it does not change the current assessment.

Final output

I now have everything needed for a complete assessment.

ICSR-003 — Case Review

Study drug: Trastuzumab deruxtecan (T-DXd) | Class: HER2 antibody-drug conjugate (ADC)
Patient: 58F, HER2+ metastatic breast cancer, known gallstones.


1. Event Coding (PT / SOC)

Reported term MedDRA PT SOC
Elevated liver enzymes (ALT 4× ULN) Alanine aminotransferase increased Investigations
Elevated AST (3× ULN) Aspartate aminotransferase increased Investigations
Mildly raised bilirubin Blood bilirubin increased Investigations

Note: These are coded as lab investigation PTs. The pattern (ALT/AST predominant, mildly raised bilirubin) does not meet Hy's Law (no documented bilirubin >2× ULN with transaminases >3× ULN in the absence of cholestasis/alternative cause), and a clinical diagnosis term (e.g., "Drug-induced liver injury") is not warranted on the available data. Coding to the discrete lab abnormalities is the appropriate, non-over-interpreting choice.


2. Seriousness Assessment

Not serious.

Testing against the standard ICH E2A seriousness criteria:
- Death — No
- Life-threatening — No
- Hospitalization / prolongation — No (explicitly not hospitalized)
- Persistent/significant disability — No
- Congenital anomaly — No
- Other medically important condition — Considered, but argues against: patient asymptomatic, modest enzyme elevations, no synthetic dysfunction, no Hy's Law, resolved spontaneously within a week without intervention or study-drug modification.

Non-serious adverse event.


3. Causality Assessment (drug = T-DXd)

Assessed causality to study drug: Unlikely / Not related.

The label confirms Alanine aminotransferase increased is a labeled (expected) reaction for this class, so T-DXd is biologically plausible. However, the totality of evidence points strongly to a confounder rather than the study drug:

Evidence against T-DXd causality:
- Strong competing cause — amoxicillin/clavulanate. Co-amoxiclav is one of the most common causes of drug-induced liver injury, with onset typically days to a few weeks after starting. The timing here (started 5 days before the abnormality) fits classically.
- Positive dechallenge of the confounder, not the study drug. The antibiotic was stopped and enzymes trended down, while T-DXd was continued unchanged. Improvement despite ongoing study-drug exposure is powerful evidence against T-DXd as the cause.
- Second competing cause — biliary disease. Known gallstones with cholelithiasis on ultrasound could contribute to the mildly raised bilirubin (though no duct dilatation was seen).
- Possible additional contributors: atorvastatin (statin transaminitis) and acetaminophen are also on board.
- No de-/re-challenge of T-DXd was performed (and none was needed), so no positive temporal evidence ties the event to the ADC.

Conclusion: Despite ALT elevation being label-expected for HER2 ADCs, the clinical course — temporal link to co-amoxiclav, resolution on stopping the antibiotic while the study drug continued unchanged, plus underlying gallstones — makes the study drug an unlikely cause. Most probable cause: amoxicillin/clavulanate-related hepatotoxicity, with possible biliary contribution.


Summary Line

58F on T-DXd with ALT increased / AST increased / Blood bilirubin increased (all SOC Investigations); non-serious; causality to study drug Unlikely — confounded by recent amoxicillin/clavulanate (classic timing + positive dechallenge while T-DXd continued) and known cholelithiasis.

Recommended follow-up (data-quality): capture baseline LFTs, ULN values, exact bilirubin level (to formally exclude Hy's Law), hepatitis serologies/alcohol history if available, and confirm enzyme normalization — to firm up the differential, though it does not change the current assessment.

Reference answer key (deterministic) — open after grading

Deterministic reference key for ICSR-003 (Trastuzumab deruxtecan (T-DXd)). Causality categories follow WHO-UMC; where the narrative supports more than one defensible category the rationale says so. A strong analyst recovers all events, the correct PT/SOC from the dictionary, the right seriousness criterion, and well-reasoned causality grounded in temporality, dechallenge/rechallenge, alternative etiology, and the drug label.

Event (as narrated)Expected PTExpected SOCSeriousness (ICH E2A)Causality (WHO-UMC) + rationale
elevated liver enzymes (ALT/AST)Alanine aminotransferase increasedInvestigationsNon-serious — Does not meet a seriousness criterion (asymptomatic, no hospitalization, resolved)Unlikely — Strong alternative etiology: amoxicillin/clavulanate (a classic cause of hepatocellular/cholestatic injury) started 5 days before onset, plus known gallstones. Positive dechallenge occurred on stopping the ANTIBIOTIC while the study drug was continued unchanged, pointing away from the ADC. 'Possible' is defensible because the ADC labels transaminase rise, but the confounder and the dechallenge make the ADC unlikely.
raised bilirubinBlood bilirubin increasedInvestigationsNon-serious — Does not meet a seriousness criterionUnlikely — Same confounders (antibiotic, gallstones); mild and resolved with the antibiotic stopped while the study drug continued.
OVERALL CASENon-serious — No ICH-E2A criterion met (non-serious, asymptomatic, resolved)Case-level ICH-E2A determination

Expert grading

Event extraction completeness 1 (poor) 5 (excellent)
5: Identifies every reportable adverse event in the narrative — including a serious event that is unrelated to the drug and secondary/lab events — without lumping distinct events or inventing ones.
1: Misses reportable events (e.g. overlooks the neutropenia behind a febrile-neutropenia admission, or drops the unrelated fracture), or merges separate events into one.
MedDRA coding accuracy 1 (poor) 5 (excellent)
5: Maps each lay event to the correct Preferred Term and its System Organ Class using the dictionary lookup (e.g. 'low white count' -> Neutropenia / Blood and lymphatic system disorders; 'shortness of breath with infiltrates' -> Pneumonitis / Respiratory). PT and SOC are consistent with the dictionary, not guessed from memory.
1: Wrong PT or mismatched SOC, codes to a symptom when a diagnosis PT exists (or vice versa), or fabricates a code never returned by meddra_lookup.
Seriousness determination (ICH E2A) 1 (poor) 5 (excellent)
5: Correctly classifies each event as serious/non-serious and names the right ICH E2A criterion (death, life-threatening, hospitalization, disability, congenital anomaly, medically important) — e.g. flags the hospitalized CRS/pneumonitis/febrile-neutropenia events as serious and the asymptomatic resolved lab abnormality as non-serious.
1: Calls a clearly serious (hospitalized/life-threatening) event non-serious or vice versa, or cites the wrong/no criterion, or conflates severity grade with seriousness.
Causality assessment quality (WHO-UMC) 1 (poor) 5 (excellent)
5: Assigns a defensible WHO-UMC category with sound reasoning: weighs temporality (onset vs dosing), dechallenge/rechallenge, alternative etiologies (confounding meds, comorbidity, trauma), and whether the event is labeled — e.g. probable/certain for step-up-dose CRS, unlikely for the antibiotic-confounded transaminitis, unrelated for the traumatic fracture.
1: Reflexively blames or exonerates the drug, ignores a strong alternative etiology or the temporal mismatch, conflates seriousness with causality, or misuses the WHO-UMC categories.
Evidence faithfulness 1 (poor) 5 (excellent)
5: Grounds every PT/SOC in an actual meddra_lookup result and every causality factor in the narrative or the drug label; no invented events, codes, lab values, or label claims; states uncertainty where the narrative is genuinely ambiguous.
1: Invents events or MedDRA codes, asserts label content not returned by get_drug_label, or contradicts the narrative (wrong timing, fabricated dechallenge).
Overall verdict
Your name / ID