Alternative Response Criteria
title: “Alternative Response Criteria” type: guidance category: recist/guidance tags: [recist, choi, irecist, mrecist, immunotherapy, gist, hcc] sources:
- file: “recist_eisenhauer_2009.pdf” page: “228-247”
- file: “recist_schwartz_2016.pdf” page: “132-137”
- file: “ajr_vandenabbeele_2012.pdf” page: “entire” date_created: 2025-04-15 date_modified: 2025-04-15 confidence: high anki_deck: “Radiology::RECIST” weak_area: false
Overview
RECIST 1.1 is the reference standard for solid tumors treated with cytotoxic systemic therapies. Certain tumor types and treatment modalities require modified criteria because:
- Tumor size changes before density changes (GIST on imatinib)
- Treatment causes pseudoprogression — apparent worsening that represents immune infiltration, not true progression (immunotherapy)
- Only the viable tumor component should be measured (HCC on locoregional therapy)
- Metabolic response precedes morphological change (lymphoma)
RECIST 1.1 — Reference Standard
| Feature | Detail |
|---|---|
| Measurement | Unidimensional (longest diameter) |
| Basis | Anatomical/morphological change only |
| Primary use | Solid tumors on cytotoxic chemotherapy |
| New lesion detection | FDG-PET (supplementary) |
| PD definition | ≥20% increase in sum + ≥5mm absolute |
Choi Criteria — GIST on Imatinib
Publication: Choi et al., 2007 (updated) Indication: Gastrointestinal stromal tumors (GIST) treated with imatinib
Key Innovation
GIST tumors responding to imatinib often undergo cystic degeneration — the tumor decreases in density (HU) before it shrinks in size. RECIST would call this SD; Choi captures it as PR.
Choi Response Definitions
| Response | Criterion |
|---|---|
| CR | Complete disappearance of all lesions; no new lesions |
| PR | ≥10% decrease in longest diameter OR ≥15% decrease in CT attenuation (HU); no new lesions; no obvious progression of non-measurable disease |
| SD | Does not meet criteria for CR, PR, or PD |
| PD | ≥10% increase in longest diameter AND no decrease in attenuation; OR new lesions; OR new intratumoral nodules or increase in size of existing nodules |
Why Density Matters for GIST
Imatinib causes tumor necrosis and cystic change — the tumor becomes less dense (hypodense on CT) even if its diameter hasn’t shrunk. This density change on contrast-enhanced CT is a valid surrogate for tumor response because:
- Viable GIST cells are contrast-enhancing
- Decreased enhancement = decreased viable tumor
- RECIST’s exclusive focus on size misses this metabolic response
When to Use Choi vs. RECIST for GIST
Clinical trials and clinical practice for GIST on imatinib should use Choi criteria. RECIST 1.1 will systematically underestimate response rates in this population. Regulatory agencies (FDA) accept Choi criteria for GIST.
iRECIST — Immunotherapy
Publication: Seymour et al., 2017 (RECIST Working Group) Indication: Solid tumors treated with immune checkpoint inhibitors (anti-PD-1, anti-PD-L1, anti-CTLA-4)
The Problem: Pseudoprogression
Immunotherapy activates T-cells against tumors, causing inflammatory infiltration into the tumor. This makes the tumor appear larger or new lesions appear on CT before it actually shrinks. This is pseudoprogression — not true progression.
- True progression: tumor regrows after responding or never responds
- Pseudoprogression: initial apparent growth → eventual response
RECIST 1.1 would declare PD on the initial apparent growth, potentially discontinuing effective therapy prematurely.
iRECIST Framework
| iRECIST Term | Meaning |
|---|---|
| iUPD (immune unconfirmed PD) | Initial progression meeting RECIST 1.1 criteria — requires confirmation |
| iCPD (immune confirmed PD) | Confirmed progression after repeat scan 4–8 weeks later |
| iPR | PR meeting RECIST criteria but in iUPD patient |
| iCR | CR meeting RECIST criteria but in iUPD patient |
iRECIST Rules
- When RECIST 1.1 criteria for PD are met → call iUPD (not PD)
- Repeat imaging at 4–8 weeks to confirm
- If confirmed (further increase or new lesions persist) → iCPD → treatment discontinuation considered
- If scans show improvement → iPR or iCR → continue treatment; prior iUPD is overwritten
iRECIST is Collected Alongside RECIST 1.1
iRECIST does NOT replace RECIST 1.1 in clinical trials. Both are recorded. Regulatory decisions still use RECIST 1.1. iRECIST is an exploratory framework to capture the pseudoprogression phenomenon without disrupting standard assessment.
Other Immunotherapy Criteria
- irRC (immune-related response criteria, 2009): First framework for immunotherapy pseudoprogression
- irRECIST (2013): Simplified version of irRC using unidimensional measurements
mRECIST — Hepatocellular Carcinoma (HCC)
Publication: Lencioni & Llovet, 2010 Indication: HCC treated with locoregional therapies (TACE, radioembolization, ablation)
The Problem: Necrotic Tumors
HCC treated with TACE or radioembolization often develops a large necrotic center. The entire tumor — including necrosis — would be measured by RECIST. But necrotic tissue is not viable tumor and should not be tracked.
Only the arterial phase enhancing (viable) component represents living tumor.
mRECIST Measurement
| Feature | RECIST 1.1 | mRECIST |
|---|---|---|
| What to measure | Total tumor diameter | Viable tumor diameter only |
| Detection method | Size change on all phases | Arterial phase hyperenhancement |
| Target lesion requirement | Any measurable lesion | Must show arterial enhancement at baseline |
| CR definition | All lesions gone | Complete disappearance of arterial enhancement |
| PR definition | ≥30% decrease in total diameter | ≥30% decrease in enhancing diameter |
mRECIST Response for HCC
| Response | Criterion |
|---|---|
| CR | Disappearance of all arterial enhancement in target lesions |
| PR | ≥30% decrease in sum of diameters of viable tumor (arterial enhancing) |
| PD | ≥20% increase in viable tumor diameter OR new arterial-enhancing lesion |
| SD | Neither PR nor PD |
Why Arterial Enhancement?
HCC derives its blood supply from the hepatic artery (unlike normal liver which is portal venous). Arterial phase CT/MRI shows HCC as a hyperenhancing mass. As tumor dies after treatment, it loses this arterial enhancement — even if the total tumor size (including necrotic rim) is unchanged. Measuring only the viable enhancing portion captures true treatment response.
Lugano Classification — Lymphoma
Publication: Cheson et al., 2014 (Lugano conference) Indication: Hodgkin and non-Hodgkin lymphoma
Lymphoma uses FDG-PET as the primary response tool (not supplementary as in RECIST). See separate lymphoma criteria for full detail.
Comparison Matrix
| Criteria | Tumor Type | Modality | Key Feature |
|---|---|---|---|
| RECIST 1.1 | Solid tumors (general) | CT/MRI | Size-based; FDG-PET for new lesions |
| Choi | GIST (on imatinib) | CT (contrast-enhanced) | Incorporates density change |
| iRECIST | Solid tumors (immunotherapy) | CT/MRI + PET | Unconfirmed PD for pseudoprogression |
| mRECIST | HCC (locoregional therapy) | CT/MRI (arterial phase) | Targets viable arterial enhancement only |
| Lugano/Cheson | Lymphoma | FDG-PET (primary) | Metabolic response; Deauville scoring |
When to Use Alternative Criteria
- GIST + imatinib: Use Choi (or RECIST + Choi dual-reporting)
- Any solid tumor + immunotherapy: Collect iRECIST alongside RECIST 1.1
- HCC + TACE/radioembolization: Use mRECIST for arterial phase evaluation
- Lymphoma: Use Lugano/Cheson criteria (not RECIST)
- All other solid tumors: RECIST 1.1 remains the reference standard