# Tumor Response Assessment: RECIST, PERCIST, and Beyond

## Overview

Standardized response criteria are essential for consistent evaluation of treatment efficacy across clinical trials and routine practice. Anatomic criteria such as RECIST measure changes in tumor size, while metabolic criteria such as PERCIST, Deauville, Hopkins, and Lugano assess functional and metabolic changes. A key advantage of metabolic response assessment is that it often precedes anatomic response, enabling earlier treatment modification when therapy is ineffective.

## RECIST 1.1 (Response Evaluation Criteria in Solid Tumors)

### Principles

RECIST 1.1 is an anatomic measurement-based system that uses CT or MRI to evaluate tumor response. Up to five target lesions are selected, with a maximum of two per organ, and each is measured in its longest diameter. Lymph nodes are measured by short axis, with nodes 15 mm or greater considered measurable and nodes below 10 mm considered normal.

### Response Categories

Complete response is defined as the disappearance of all target lesions, with lymph nodes reduced to below 10 mm in short axis. Partial response requires at least a 30% decrease in the sum of longest diameters from baseline. Progressive disease is declared when there is at least a 20% increase in the sum of longest diameters with an absolute increase of at least 5 mm, or when new lesions appear. Stable disease applies when neither partial response nor progressive disease criteria are met.

### Limitations

RECIST does not account for metabolic changes such as the distinction between viable and necrotic tumor. It is poorly suited for therapies that cause tumor necrosis without significant size change, including targeted therapies and immunotherapy. Pseudoprogression on immunotherapy can be misclassified as progressive disease. Additionally, there is meaningful variability in lesion measurement between readers.

<image>RECIST 1.1 response categories illustrated with serial CT measurements showing complete response, partial response, stable disease, and progressive disease</image>

## PERCIST 1.0 (PET Response Criteria in Solid Tumors)

### Principles

PERCIST is a metabolic response framework based on FDG PET/CT. It uses SUL (SUV corrected for lean body mass) rather than SUVmax, which reduces variability related to body composition. The key measurement is SULpeak, defined as the average SUL within a 1 cm3 spherical region of interest centered on the hottest voxel of the tumor. Consistent imaging parameters across timepoints, including scanner type, dose, uptake time, and reconstruction algorithm, are required for valid comparison.

### Response Categories

Complete metabolic response is defined as complete resolution of FDG uptake in all lesions to background blood pool level. Partial metabolic response requires at least a 30% decrease in SULpeak with an absolute decrease of at least 0.8 SUL units. Progressive metabolic disease is declared when SULpeak increases by at least 30% with an absolute increase of at least 0.8 SUL units, or when new FDG-avid lesions appear. Stable metabolic disease applies when neither partial metabolic response nor progressive metabolic disease criteria are met.

| Criteria | Measurement | CR | PR | SD | PD |
|---|---|---|---|---|---|
| RECIST 1.1 | Sum of longest diameters (CT) | Disappearance of all lesions | ≥30% decrease | Neither PR nor PD | ≥20% increase (+5 mm) or new lesions |
| PERCIST 1.0 | SULpeak (PET) | Resolution to blood pool | ≥30% decrease (≥0.8 SUL) | Neither PMR nor PMD | ≥30% increase (≥0.8 SUL) or new lesions |
| Deauville (Lymphoma) | Visual 5-point scale (PET) | Score 1–3 | Score 4–5, reduced | Score 4–5, unchanged | Score 4–5, increased or new |
| Choi (GIST) | Size + density (CT) | Disappearance | ≥10% size or ≥15% density decrease | Neither PR nor PD | ≥10% size increase, no density decrease |
| iRECIST | Sum diameters + confirmation | Disappearance confirmed | ≥30% decrease | Neither response nor progression | Confirmed ≥20% increase at 4–8 wk |

### Advantages Over RECIST

PERCIST detects response earlier than anatomic changes become apparent. It is better suited for assessing targeted therapies and immunotherapies, and it evaluates entire tumor biology rather than size alone. PERCIST can also assess response in disease that is non-measurable by RECIST criteria.

## Deauville Five-Point Scale (Lugano Classification for Lymphoma)

### Scoring System

The Deauville scale uses a visual five-point scoring system. Score 1 indicates no uptake above background. Score 2 indicates uptake at or below the mediastinal blood pool. Score 3 indicates uptake above the mediastinal blood pool but at or below liver uptake. Score 4 indicates uptake moderately greater than liver. Score 5 indicates uptake markedly greater than liver or the presence of new lesions. Score X designates new areas of uptake that are unlikely related to lymphoma.

### Application

The Deauville scale is the standard for both interim and end-of-treatment response assessment in Hodgkin lymphoma and aggressive non-Hodgkin lymphoma such as DLBCL. On interim PET performed after two cycles, scores 1 through 3 indicate adequate response while scores 4 and 5 indicate inadequate response. On end-of-treatment PET, scores 1 through 3 indicate complete metabolic response while scores 4 and 5 indicate residual disease. These results guide treatment modification, allowing de-escalation for good responders and escalation for poor responders.

<image>Deauville five-point scale reference images showing scores 1 through 5 with corresponding mediastinal blood pool and liver reference regions in lymphoma PET/CT assessment</image>

## Lugano Classification (2014)

### Integrated Response Criteria for Lymphoma

The Lugano classification combines PET-based assessment for FDG-avid lymphomas with CT-based assessment for non-FDG-avid histologies. For FDG-avid lymphomas, complete response requires a Deauville score of 1 through 3 with or without a residual mass. Partial response is assigned with a Deauville score of 4 or 5 when uptake is reduced from baseline and residual masses are present. Stable disease corresponds to a Deauville score of 4 or 5 with no significant change. Progressive disease is declared with a Deauville score of 4 or 5 when uptake has increased or new lesions have appeared. For non-FDG-avid lymphomas, CT-based response uses the sum of product of diameters for up to six target lesions.

## Hopkins Criteria (Head and Neck Cancer)

The Hopkins criteria provide a five-point scale for evaluating post-chemoradiation PET/CT in head and neck squamous cell carcinoma. The internal jugular vein and liver serve as internal reference standards. Scores 1 through 3 support observation, while scores 4 and 5 warrant biopsy or intervention. The scale carries a high negative predictive value exceeding 95% for complete response.

## Choi Criteria (GIST)

### Rationale

Gastrointestinal stromal tumors treated with imatinib may demonstrate a marked response through decreased attenuation on contrast-enhanced CT without significant size reduction. As a result, RECIST underestimates response in GIST.

### Response Criteria

Under the Choi criteria, partial response is defined as at least a 10% decrease in tumor size or at least a 15% decrease in tumor density measured in Hounsfield units on contrast-enhanced CT. Progressive disease requires at least a 10% increase in size without a decrease in density, or the appearance of new lesions or new intratumoral nodules. FDG PET response is even earlier, often detectable within one to four weeks of starting imatinib.

## Immune-Related Response Criteria

### iRECIST

iRECIST is a modified version of RECIST designed for immunotherapy to account for pseudoprogression. It introduces the concept of unconfirmed progressive disease, which requires confirmation at four to eight weeks. If the subsequent scan shows continued progression, the finding is reclassified as confirmed progressive disease. If the follow-up scan demonstrates stability or decrease, the response is recategorized accordingly.

### iPERCIST

iPERCIST applies the same confirmatory approach to metabolic response assessment. New or increasing FDG-avid lesions require confirmation before progression is declared.

<image>Pseudoprogression on immunotherapy: initial PET/CT shows new FDG-avid lesions (iUPD), followed by confirmatory scan at 8 weeks showing complete resolution, reclassified as metabolic response</image>

## Quantitative PET Metrics

### SUV-Based Metrics

SUVmax is the maximum voxel value within a lesion and is the most widely used metric due to its simplicity and reproducibility. SUVmean is the average value within a region of interest and is less affected by noise but depends on how the region is defined. SUVpeak is the average within a 1 cm3 sphere centered on the hottest voxel and is more robust than SUVmax. SULpeak is SUVpeak corrected for lean body mass and is the preferred metric in PERCIST.

### Volumetric Metrics

Metabolic tumor volume is the total volume of metabolically active tumor above a defined threshold, commonly SUV 2.5 or 41% of SUVmax. Total lesion glycolysis is calculated by multiplying MTV by the SUVmean within the volume. Both MTV and TLG carry prognostic value in many malignancies, including lymphoma, lung cancer, esophageal cancer, and cervical cancer. Emerging evidence supports their use as predictive biomarkers for treatment outcome.

## Harmonization Challenges

SUV measurements vary with scanner type, reconstruction algorithm, acquisition parameters, and patient preparation. The EARL accreditation program, operated by EANM Research Ltd, standardizes PET quantification across sites. This harmonization is particularly important for multicenter trials and serial response assessment performed at different institutions.

<image>Impact of different reconstruction algorithms and scanner types on SUVmax measurements of the same lesion, illustrating the need for harmonization in multicenter response assessment</image>

## Clinical Pearls

The same scanner, protocol, and reconstruction should be used for serial response assessment whenever possible to minimize measurement variability.

RECIST and PERCIST can give discordant results, and metabolic response may predict outcomes better than anatomic response in many tumor types.

The Deauville scale for lymphoma relies on visual comparison to mediastinum and liver, with numeric SUV values playing a secondary role.

In GIST, a rapid metabolic response on FDG PET within one to two weeks of imatinib predicts drug sensitivity and can guide early therapy changes.

Pseudoprogression occurs in fewer than 10% of immunotherapy patients but is clinically important. Confirmatory imaging should always be considered before abandoning effective immunotherapy.

MTV and TLG are increasingly recognized as superior prognostic markers compared with SUVmax alone.

## References

- Eisenhauer, E. A., et al. "RECIST 1.1." *European Journal of Cancer*, 2009.
- Wahl, R. L., et al. "PERCIST 1.0." *Journal of Nuclear Medicine*, 2009.
- Cheson, B. D., et al. "Lugano Classification for Lymphoma Response." *Journal of Clinical Oncology*, 2014.
- Seymour, L., et al. "iRECIST for Immunotherapy Response." *Lancet Oncology*, 2017.
- Boellaard, R., et al. "FDG PET/CT: EANM Procedure Guidelines." *European Journal of Nuclear Medicine*, 2015.