What Does RECIST Mean in Cancer?

RECIST stands for Response Evaluation Criteria in Solid Tumors. It is a written rulebook that turns a stack of CT images into one of four words: complete response, partial response, stable disease, or progressive disease.

The version in use is RECIST 1.1. It exists because "the tumor looks a bit smaller" means different things to different readers. If a drug trial in Boston and a trial in Lyon both say 40 percent of patients responded, that number is only comparable if both used the same ruler.

RECIST applies to solid tumors. It does not apply to leukemia, and it is a poor fit for some cancers such as mesothelioma and bone-only disease.

Not every spot on your scan is counted. At the baseline scan, before treatment starts, the radiologist picks the lesions that will be tracked.

Target lesions. Up to 5 in total, and no more than 2 in any single organ. They must be measurable, meaning large enough to measure reliably. Each is measured by its longest diameter in one direction only.

Lymph nodes get their own rule. A node counts as a measurable target lesion only if its short axis is 15 mm or more. Nodes with a short axis under 10 mm are treated as normal and are not recorded at all.

Non-target lesions. Everything else known about, including smaller spots, fluid around the lung, and bone disease. These are watched and described but not measured with a ruler.

The target lesion diameters are added up. That total is the baseline sum of diameters. Everything that follows compares back to a sum.

Complete response (CR). All target lesions have disappeared, and any target lymph node has shrunk to a short axis under 10 mm.

Partial response (PR). The sum of diameters has dropped by 30 percent or more from baseline.

Progressive disease (PD). The sum has risen by 20 percent or more, compared with the smallest sum recorded at any point in the study, and that rise is at least 5 mm in absolute terms. A new lesion anywhere also means progressive disease, no matter what the sum is doing.

Stable disease (SD). Neither PR nor PD. The tumor has not shrunk by 30 percent and has not grown by 20 percent.

Two details in there matter more than they look.

First, progression is judged against the smallest sum on study, not against baseline. If your tumors shrank a lot and then crept back up, the comparison point is the low water mark, not where you started.

Second, the 5 mm absolute rule stops tiny tumors from generating false alarms. A lesion going from 11 mm to 14 mm is a 27 percent rise, which sounds like progression. It is only 3 mm, so it does not qualify.

Stable disease covers an enormous range. A tumor that shrank 29 percent and a tumor that grew 19 percent both land in the same box. That is why your oncologist may sound pleased about a scan that the report calls stable, or unimpressed by one that is technically a partial response.

RECIST also cannot see everything that matters. It does not measure how you feel, whether pain has eased, whether you are eating again, or whether a tumor is pressing on something dangerous. A cancer can meet the definition of stable disease while causing a serious problem in one spot.

Immune drugs broke the old assumptions. Some tumors get bigger on the first scan after starting immunotherapy, because immune cells have flooded into them, and then shrink. That is pseudoprogression, meaning apparent growth that is not real cancer growth. Under strict RECIST 1.1, those patients would be called progressors and taken off a drug that was actually working.

iRECIST is the modification built for this. It adds two categories:

• iUPD, immune unconfirmed progressive disease. Growth or a new lesion has been seen once. Treatment can continue.
• iCPD, immune confirmed progressive disease. The next scan, done 4 to 8 weeks later, shows a further increase: at least 5 mm more in the target lesion sum, or any further increase in non-target disease or in a new lesion.

If the repeat scan comes back better or unchanged, the iUPD label is dropped. If the repeat happens sooner than 4 weeks, it does not count for confirming iCPD.

• Which lesions were chosen as targets, and are they the same ones as last time?
• What is the current sum of diameters, and what was the smallest sum so far?
• Is the reference point for this comparison the baseline scan or a later, smaller one?
• Was this scan done with the same technique as the last one?
• If the report says progressive disease, was it the sum, or a new lesion?
• If I am on immunotherapy, is this iUPD, and when is the confirming scan?

Consistency of technique is not a small point. RECIST assumes each assessment uses the same imaging method and technique. A CT with contrast compared against one without can shift measurements on its own.

Call 911 or go to an emergency department now for any of these:

• New weakness in an arm or leg, trouble speaking, or a seizure
• Shortness of breath at rest, or coughing up blood
• Belly pain with vomiting and no bowel movement
• New back pain with numbness, or loss of bladder or bowel control

Call your cancer team the same day, and follow the urgent number they gave you, for:

• A headache that wakes you or is worst on waking, especially with vomiting
• Temperature 100.4 F (38 C) or higher while on cancer treatment

Imaging Tests, Pathology Reports, and Getting a Second Opinion.

• RECIST Working Group (EORTC) — RECIST 1.1
• RECIST Working Group (EORTC) — iRECIST
• PubMed Central (European Journal of Cancer) — RECIST 1.1: Update and Clarification from the RECIST Committee
• PubMed Central (Abdominal Radiology) — Pearls and Pitfalls of RECIST v1.1 Non-Target Lesion Assessment
• PubMed Central (AJR) — New RECIST Guidelines for Advanced Non-Small Cell Lung Cancer: Comparison with Original RECIST