What Does Measurable Disease Mean?

"Measurable disease" sounds like a judgment about how bad things are. It is not. It is a technical yes-or-no about whether your tumors are big enough and clear enough on scans to be tracked with a ruler.

The ruler in question is RECIST, short for Response Evaluation Criteria in Solid Tumors. Version 1.1 is the one in use. The FDA's guidance on cancer drug endpoints names revised RECIST 1.1 as an appropriate standardized criteria set, while insisting that whichever criteria a trial uses must be written into the protocol in advance.

A person can be very sick and still have no measurable disease. A person can feel fine and have plenty of it. The label describes the scan, not the patient.

RECIST 1.1 sets a floor. A lesion that is not a lymph node has to be larger than 10 mm across its longest axis on CT to count as measurable.

The scan technique matters too. The RECIST Committee strongly recommends a CT slice thickness of 5 mm. If slices are thicker than 5 mm, the minimum size for a measurable lesion doubles the slice thickness. On a 10 mm slice scan, a lesion would need to be 20 mm.

This is why a repeat scan is sometimes ordered on a different protocol. It is not distrust of the first radiologist. It is a technical requirement.

Nodes are measured on their short axis, the narrower width, not the long one. Three thresholds matter:

• Under 10 mm short axis — treated as normal
• 10 mm or more short axis — considered abnormal
• 15 mm or more short axis — measurable, and eligible to be tracked

So a node at 12 mm is abnormal but not measurable. It gets watched, not counted.

Some real disease simply cannot be tracked this way. The RECIST 1.1 criteria list non-measurable disease as including:

• Lesions smaller than the size thresholds
• Leptomeningeal disease, meaning cancer in the membranes around the brain and spinal cord
• Ascites, meaning fluid collecting in the abdomen
• Bone lesions without an identifiable soft tissue part

That last one surprises people. Blastic bone metastases are not measurable at all. A lytic or mixed lytic-blastic deposit can be measured, but only when it has a soft tissue component big enough to reach the size threshold on CT or MRI. Bone scan and plain films do not count as adequate imaging for this. That is a common reason someone with obvious cancer is told they have no measurable disease.

The radiologist does not measure everything. RECIST 1.1 caps the tracked set at 5 target lesions in total, with no more than 2 in any single organ. No more than 2 of those may be lymph nodes. Everything else is followed as non-target disease, meaning it is watched for clear worsening but not measured.

The five chosen lesions become the record. Their diameters are added into one number, the sum of diameters. Every future scan is compared against that sum.

Response is graded against that single number.

Complete response. All non-nodal lesions gone, and any target lymph nodes back under 10 mm short axis. Note the wording: nodes do not have to vanish, they have to shrink below the normal threshold.

Partial response. The sum drops by 30% or more from baseline.

Progressive disease. The sum rises by 20% or more compared with the nadir, which is the smallest sum recorded at any point since treatment started. There is a second requirement people miss: the increase must also be at least 5 mm in absolute terms. Any new lesion also counts as progression on its own.

Stable disease. Anything in between.

That 5 mm rule prevents a false alarm. If the sum falls to 20 mm, a 20% rise is only 4 mm, which is within measurement noise. RECIST refuses to call that progression.

Responses are usually confirmed on repeat imaging at least 4 weeks later.

Many trials require measurable disease at entry. The reason is in the FDA guidance. Objective response rate is the proportion of patients whose tumors shrank by a predefined amount for a minimum time, calculated as complete responses plus partial responses. Stable disease is not counted in it.

You cannot compute a response rate from a patient whose disease cannot be measured. That is the whole obstacle.

The FDA guidance also explains why the scans are read the way they are. When response rate or progression-free survival is the primary endpoint, tumor assessments generally should be verified by central reviewers blinded to which treatment each patient received.

If you are told you do not qualify for a trial on these grounds, ask two things. Whether a different trial accepts non-measurable disease. And whether a lesion that fell just under the threshold might cross it on the next scan.

Progression-free survival is defined by the FDA as the time from randomization until objective tumor progression or death, whichever comes first. It is a trial measurement. It is not a prediction for one person.

RECIST also has known blind spots. It was built for chemotherapy, where shrinkage is the expected signal. Immunotherapy can produce temporary swelling before improvement, which is why separate immune-related criteria were developed.

• Do I have measurable disease under RECIST, and which lesions were chosen as targets?
• What is my current sum of diameters, and what was the baseline?
• What is the nadir being used for comparison?
• Is any of my disease non-measurable, and how will that be followed?
• Was the CT done at 5 mm slices or thicker?

Imaging Tests explains how CT, MRI, and PET differ. Cancer Staging covers how stage is assigned. Pathology Reports covers the tissue side of the record.

• U.S. Food and Drug Administration — Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics
• FDA guidance text — Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics
• RECIST 1.1 — Update and Clarification: From the RECIST Committee (PubMed Central, NIH)
• Immune-related tumour response assessment criteria: a comprehensive review (PubMed Central, NIH)