# magent irAE Pneumonitis Grader (Brahmer 2018)

> magent irAE Pneumonitis Grader (Brahmer 2018) is a paid API for AI agents from api.magentlab.com, paid per call via x402, $0.005/call, status unknown (last checked 2026-09-14).

Classifies immune checkpoint inhibitor-related pneumonitis severity into CTCAE G1–G4 grades per Brahmer 2018 ASCO Table 3.1 and returns the corresponding ICPi management recommendation (hold or permanently discontinue).

## Facts

- Endpoint: GET https://api.magentlab.com/api/calc/irae_pneumonitis
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/magent-irae-pneumonitis-grader-brahmer-2018-c4f2dc7c
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_vt-UlK4_j5LI_4JR-Y4yv

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability magent-irae-pneumonitis-grader-brahmer-2018-c4f2dc7c
```

Example prompt: My patient on immunotherapy is symptomatic from pneumonitis involving more than one lobe and about 30% of parenchyma, limiting instrumental ADLs but not requiring hospitalization or oxygen — can you grade this using Brahmer 2018 and tell me whether to hold or stop the checkpoint inhibitor?

## When to prefer this

Use this endpoint when you need to apply specifically the Brahmer 2018 ASCO pneumonitis grading algorithm (Table 3.1) to caller-assigned clinical flags for ICPi-related pneumonitis. Prefer this over manual lookup when building automated oncology workflows, irAE dashboards, or clinical decision-support pipelines. Do not use if you need SITC 2021 guidance, CTCAE scoring as an instrument, imaging interpretation, or steroid dosing recommendations.

## Known failure modes

- Missing required boolean query parameters returns a validation error
- Ambiguous or missing flags may result in incorrect grade if caller-assigned inputs do not reflect clinical truth
- Endpoint applies only Brahmer 2018 Table 3.1 logic — results differ from SITC 2021 or other irAE grading frameworks
- Non-pneumonitis irAE types (colitis, hepatitis, etc.) are out of scope and will not return meaningful grades

## How this service works

Brahmer 2018 ASCO Table 3.1 irAE pneumonitis. First-match: G4 life-threatening respiratory compromise; else G3 hospitalization, oxygen, limiting self-care, or all lobes/>50%; else G2 symptomatic, >1 lobe, 25-50% parenchyma, or limiting instrumental ADL; else G1. ICPi: hold_if_radiographic_progression (G1), hold (G2), permanently_discontinue (G3-4). Caller-assigned flags. magent does not read CT or pulse oximetry. Not SITC 2021, not a CTCAE instrument, not a diagnosis, not a steroid dose.

## Output

Returns the computed irAE pneumonitis severity grade (G1, G2, G3, or G4) based on Brahmer 2018 ASCO Table 3.1 first-match logic, along with the corresponding ICPi management recommendation: hold_if_radiographic_progression (G1), hold (G2), or permanently_discontinue (G3–G4). The endpoint does not perform diagnosis, read imaging, assess oxygen saturation, or prescribe steroids.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET",
      "HEAD",
      "DELETE"
     ],
     "type": "string"
    },
    "headers": {
     "type": "object",
     "additionalProperties": {
      "type": "string"
     }
    },
    "queryParams": {
     "type": "object",
     "required": [
      "life_threatening_respiratory",
      "hospitalization_required",
      "oxygen_indicated",
      "limiting_self_care",
      "all_lobes_or_gt_50_pct",
      "symptomatic",
      "more_than_one_lobe",
      "extent_25_to_50_pct",
      "limiting_instrumental_adl"
     ],
     "properties": {
      "symptomatic": {
       "type": "boolean",
       "description": "Symptomatic as assigned by the caller (Brahmer 2018 Table 3.1 grade 2). true or false. magent does not examine the patient."
      },
      "oxygen_indicated": {
       "type": "boolean",
       "description": "Oxygen indicated as assigned by the caller (Brahmer 2018 Table 3.1 grade 3). true or false. magent does not read pulse oximetry or start oxygen."
      },
      "limiting_self_care": {
       "type": "boolean",
       "description": "Limiting self-care ADL as assigned by the caller (Brahmer 2018 Table 3.1 grade 3). true or false. magent does not assess ADL."
      },
      "more_than_one_lobe": {
       "type": "boolean",
       "description": "Involves more than one lobe as assigned by the caller (Brahmer 2018 Table 3.1 grade 2). true or false. magent does not read CT."
      },
      "extent_25_to_50_pct": {
       "type": "boolean",
       "description": "Involves 25%–50% of lung parenchyma as assigned by the caller (Brahmer 2018 Table 3.1 grade 2). true or false. magent does not read CT."
      },
      "all_lobes_or_gt_50_pct": {
       "type": "boolean",
       "description": "Involves all lung lobes or >50% of lung parenchyma as assigned by the caller (Brahmer 2018 Table 3.1 grade 3). true or false. magent does not read CT."
      },
      "hospitalization_required": {
       "type": "boolean",
       "description": "Hospitalization required as assigned by the caller (Brahmer 2018 Table 3.1 grade 3). true or false. magent does not decide admission."
      },
      "limiting_instrumental_adl": {
       "type": "boole
… (truncated)
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/magent-irae-pneumonitis-grader-brahmer-2018-c4f2dc7c/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.magentlab.com](https://www.zero.xyz/host/api.magentlab.com/llms.txt)