RECIST 1.1 Response Category Interpreter
RECIST 1.1 Response Category Interpreter
Assign a RECIST 1.1 overall response from the sum of target diameters, the non-target disease and new lesions. The rule that changes answers: progression is measured from the smallest sum recorded on study, not from baseline — so a patient can be progressing while still well below where they started.
RECIST 1.1 overall response
Sums and lesion status to a categoryBaseline sum 120 mm; smallest sum on study 60 mm; current sum 74 mm; non-target disease persisting; no new lesions
RECIST 1.1 overall response, as published
Complete response: all non-nodal target lesions gone, and every pathological node reduced to a short axis under 10 mm
Partial response: the sum of diameters falls by at least 30% from BASELINE
Progressive disease: the sum rises by at least 20% over the SMALLEST SUM ON STUDY and by at least 5 mm in absolute terms
Stable disease: neither
Non-target lesions
Complete response: all gone and any tumour marker normalised · Non-CR/non-PD: one or more persists · Progressive disease: unequivocal progression
New lesions — any new unequivocal malignant lesion is progression
- the nadir, not the baseline
- the single rule on this page that changes answers. Progression is referenced to the smallest sum recorded on study, which includes the baseline sum only if that is the smallest. A patient whose sum fell from 120 mm to 60 mm and has come back to 74 mm is progressing, because that is 23% and 14 mm above the nadir — and they are still 38% below baseline, where the partial-response test would still be satisfied
- and it is both, not either
- at least 20% AND at least 5 mm. A nadir sum of 20 mm rising to 24 mm is exactly 20% but only 4 mm, and is NOT progression — the absolute term is what RECIST 1.1 added to stop small sums being called on measurement noise. A nadir of 100 mm rising to 106 mm is 6 mm but only 6%, and is not progression either
- the sum
- longest diameter for each non-nodal target lesion, SHORT axis for each nodal one, added together. Up to five target lesions and no more than two per organ. RECIST justifies cutting the limit from ten lesions to five with data rather than by convention: “data warehouse analysis shows no loss of information if lesion number reduced from 10 to 5”
- the nodal complete-response rule
- “Any pathological lymph nodes (whether target or non-target) must have reduction in short axis to under 10 mm.” It applies to nodes that were never chosen as targets, so a single 11 mm node blocks a complete response however completely everything else has gone. And because nodal targets keep contributing their short axis, the sum in a complete response is usually not zero
- confirmation and intervals
- in non-randomised trials where response is the primary endpoint, a complete or partial response must be confirmed at a later time point, generally four weeks later. Stable disease as a best response needs the criteria met at a minimum interval, generally not less than six to eight weeks
Worked example
Baseline sum 120 mm; smallest sum on study 60 mm; current sum 74 mm; non-target disease persisting; no new lesions
No new lesions and no unequivocal non-target progression, so the answer turns on the target sum
Change from the NADIR: 74 ÷ 60 = 1.233, a 23.3% increase. At or above 20%, so the first condition is met
Absolute change from the nadir: 74 − 60 = 14 mm. At or above 5 mm, so the second condition is met too
Both conditions met → progressive disease
Now read the same patient against baseline, which is the mistake: 74 ÷ 120 = 0.617, a 38.3% decrease, which satisfies the partial-response criterion. A page that referenced progression to baseline would call this a continuing partial response
Change the current sum to 72 mm and it is exactly 20% up from the nadir with a 12 mm rise — still progression, at the boundary
Change the nadir to 20 mm and the current sum to 24 mm: 20.0% up, but only 4 mm. Not progression, because the 5 mm absolute term is not met. That single rule is why RECIST 1.1 differs from RECIST 1.0 here
The four target-lesion categories, with their reference point
| Category | Criterion | Measured against |
|---|---|---|
| Complete response | All non-nodal target lesions gone; every pathological node reduced to a short axis under 10 mm | Not a percentage — an absolute state |
| Partial response | Sum of diameters at least 30% lower | Baseline sum |
| Stable disease | Neither of the above nor progression | Baseline for shrinkage, nadir for growth |
| Progressive disease | Sum at least 20% higher and at least 5 mm higher in absolute terms, or any new lesion, or unequivocal non-target progression | Smallest sum on study (the nadir) |
Where RECIST 1.1 differs from RECIST 1.0
| Rule | RECIST 1.0 | RECIST 1.1 |
|---|---|---|
| Target lesions in total | 10 | 5 |
| Target lesions per organ | 5 | 2 |
| Lymph nodes | Measured like any other lesion | Short axis; measurable as a target only at 15 mm or more; must fall under 10 mm for a complete response |
| Progression on target lesions | 20% increase | 20% increase and an absolute increase of at least 5 mm |
A set of rules, not a formula, and the rule that catches people
RECIST is a rulebook rather than an equation, and the arithmetic in it is trivial: add up some diameters and compute two percentages. Almost everything that goes wrong in applying it goes wrong in the rules around that arithmetic — which lesions count, what they are measured along, and above all what each percentage is measured against.
The reference point is the thing to get right. A partial response is a fall of at least 30% from the baseline sum. Progression is a rise of at least 20% from the smallest sum recorded on study — the nadir — which is the baseline only in a patient who never responded at all. That asymmetry is deliberate and it has a consequence that surprises people every time: a patient whose target sum fell from 120 mm to 60 mm and has come back to 74 mm has progressive disease, because they are 23% and 14 mm above their nadir, even though they remain 38% below where they started and would still satisfy the partial-response test against baseline. Referencing progression to baseline instead systematically hides progression in exactly the patients who responded best.
The second rule worth stating out loud is that progression on target lesions is a conjunction. RECIST 1.1 requires at least a 20% increase and an absolute increase of at least 5 mm, and the absolute term was added precisely because a 20% rise on a small sum can be measurement variability rather than growth. A nadir of 20 mm rising to 24 mm is exactly 20% and only 4 mm, and is not progression. A nadir of 100 mm rising to 106 mm is 6 mm and only 6%, and is not progression either. Both halves, every time.
Then there are the two things that override the sums entirely. Any new unequivocal malignant lesion is progression, including in a patient whose target disease has disappeared. And unequivocal progression of non-target disease — disease that is recorded but never measured — is progression whatever the measured lesions are doing, which is the one place where unmeasured disease outranks measured disease. RECIST sets a high bar for it, asking for an overall level of substantial worsening rather than a modest change at one site, because the judgement is subjective. Non-target disease does one more thing: while it persists it blocks a complete response and leaves a partial response, even when every target lesion has gone. A figure from a cohort is not this patient’s outcome and a response category is not a diagnosis: a stratum in which 42 per cent were alive at fifteen years tells you about that stratum, not which 42 per cent. This page computes a published quantity and states the criteria behind it. It renders no dose, no prescription and no treatment decision — that is the treating team’s. Every coefficient, conversion factor and threshold here is attributed to the source it was read in and, where it is a prognostic figure, to its derivation cohort; where the treating protocol differs, the protocol takes precedence.
Frequently asked questions
Is RECIST progression measured from baseline or from the nadir?
From the nadir — “the smallest sum on study”, which includes the baseline sum only if that happens to be the smallest. A patient who responded and then regrew can be progressing while still far below baseline, and this is the single most consequential rule in the criteria. Partial response, by contrast, is measured from baseline.
Does a 20% increase always mean progressive disease?
No. RECIST 1.1 requires at least a 20% increase from the nadir AND an absolute increase of at least 5 mm in the sum of diameters. A nadir of 20 mm rising to 24 mm is exactly 20% but only 4 mm, and is not progression. The absolute term was added in version 1.1 to stop small sums being called progressive on measurement noise.
How many target lesions does RECIST 1.1 allow?
Up to five in total, with no more than two per organ. RECIST 1.0 allowed ten and five. The working group reduced it on the basis of a data warehouse analysis showing no loss of information in going from ten to five.
How are lymph nodes handled?
By short axis throughout. A node is measurable and eligible as a target lesion only at a short axis of 15 mm or more; a node at 10 mm up to 15 mm is pathological but non-measurable and is recorded as non-target; under 10 mm it is non-pathological. For a complete response, every pathological node, target or not, must reduce to a short axis under 10 mm.
Can the sum of diameters be greater than zero in a complete response?
Yes, and usually is when nodal target lesions were selected. A node keeps contributing its short axis to the sum even after it has fallen below 10 mm and stopped being pathological, so a complete response often has a small residual sum. Judging complete response by whether the sum reached zero is a mistake.
Do complete and partial responses need confirming?
In non-randomised trials where response is the primary endpoint, yes — generally at a time point four weeks later, which RECIST says is to ensure the response is not the result of measurement error. Stable disease claimed as a best response needs the criteria met at a minimum interval, generally not less than six to eight weeks.
Related calculators
References
- Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45(2):228–47.
- RECIST Working Group / EORTC. RECIST 1.1. recist.eortc.org.
- Nishino M, Jagannathan JP, Ramaiya NH, Van den Abbeele AD. Revised RECIST guideline version 1.1: what oncologists want to know and what radiologists need to know. AJR Am J Roentgenol. 2010;195(2):281–9.
- Okuno S. RECIST: new CRA orientation. Alliance for Clinical Trials in Oncology, November 2016.
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/