What NCCN category 2A actually means
2A is widely read as "weak evidence, weak recommendation." It is not. The panel agreement behind 2A is the same as behind category 1 — and 2A is what you get by default when no category is printed at all.
Background reading, not clinical guidance. This describes published research; it is not a recommendation for any individual case, and the cited studies may have been superseded.
NCCN grades each recommendation on a scale that combines two separate things: how strong the underlying evidence is, and how much the panel agreed. Reading the label as a single measure of confidence is where the misunderstanding starts.
The part people get wrong
Category 1 and category 2A require the same level of panel agreement — NCCN's "uniform consensus," defined as at least 85% support. They differ only in the evidence beneath them: category 1 rests on high-level evidence such as randomised phase 3 trials, 2A on lower-level evidence.
So 2A does not mean the panel was lukewarm. It means the panel was just as united, on a thinner evidence base.
The categories that do encode disagreement are further down: 2B is lower-level evidence with only 50–85% panel support, and category 3 means real dispute, with at least a quarter of the panel objecting.
2A is also the default
Every NCCN guideline carries a standing line: *all recommendations are category 2A unless otherwise indicated*. A recommendation with no visible category is not unclassified — it is 2A.
How much of the guidelines this covers
An independent academic analysis — not an NCCN publication — reviewed the 2019 guidelines for the ten most common cancers in the United States and counted 1,818 staging, treatment and surveillance recommendations. Of those, 7% were category 1, 87% were 2A, 6% were 2B, and none were category 3.
The same authors had counted 1,023 recommendations in 2010 and found essentially the same distribution. The guidelines grew by 77% over the decade; the proportion resting on high-level evidence barely moved.
Why this matters for reading a plan
Most of oncology practice runs on expert consensus rather than phase 3 trials, and the guidelines say so openly if you read the categories correctly. That is not a scandal — much of medicine could never be randomised — but it does change what "guideline-recommended" is evidence of.
OpenOnco surfaces the category alongside each cited recommendation for this reason. A clinician should be able to see whether they are looking at a randomised result or a room full of experts who agreed.
Comments
Threads are public and indexed — please keep personal medical data out of them. Found a clinical error? Open an issue instead.
Comments are not switched on yet — the backend still needs to be configured.