# Radiotherapy versus observation following surgical resection of WHO grade 2 atypical meningioma (ROAM/EORTC-1308): an international, multicentre, open-label, phase 3, randomised controlled trial

- **DOI:** [10.1016/s0140-6736(26)01804-0](https://doi.org/10.1016/s0140-6736(26)01804-0)
- **Rating:** Sound (3.7 out of 5)
- **Verdict:** Randomised evidence supports adjuvant radiotherapy after complete atypical meningioma resection, but imprecision limits certainty about the benefit’s magnitude.
- **Framework version:** v5.13
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-11-20
- **Evaluation page:** https://www.nabu.science/paper/10.1016%2Fs0140-6736(26)01804-0

## Key claims

1. Disease-free survival was improved in the radiotherapy group compared with the observation group (hazard ratio 0·51 [95% CI 0·27–0·97]; p=0·0396).
   - Support: Supported
   - Type: Primary finding
   - Basis: randomised ITT evidence with unmasked local imaging assessment, fewer events than planned, and an imprecise effect estimate
   - Location: Abstract, Findings; Results
2. Adjuvant fractionated radiotherapy halves the risk of recurrence, regardless of methylation class, and with minimal late toxicity.
   - Support: Tentative
   - Type: Secondary
   - Basis: post-hoc as-treated molecular adjustment without demonstrated subgroup interactions cannot establish benefit regardless of methylation class
   - Location: Discussion, final paragraph
3. There were no between-group differences in health-related quality of life or neurocognitive function and there was minimal radiation toxicity.
   - Support: Tentative
   - Type: Secondary
   - Basis: substantial secondary-outcome missingness and unblinded patient reporting prevent firm conclusions about absence of detriment
   - Location: Research in context, Added value of this study
4. Grade 2 or 3 radiation-related serious adverse events occurred in five (8%) of 66 participants in the radiotherapy group, with no treatment-related deaths.
   - Support: Supported
   - Type: Secondary
   - Basis: systematic harms reporting supports the counts despite as-treated, open-label ascertainment
   - Location: Abstract, Findings; Results; Table 2
5. [Early radiotherapy-related adverse events] were expected and all were mild (grade 1 or 2).
   - Support: Unsupported
   - Type: Secondary
   - Basis: the reported early grade 3 radiation-related seizure contradicts universal grade 1-or-2 severity
   - Location: Discussion, adverse effects

## Strengths

- **Randomised evidence addresses a treatment uncertainty** The comparison with observation directly informs postoperative management after complete resection of atypical meningioma, extending evidence previously drawn from retrospective cohorts and non-randomised phase 2 studies.
  - Dimension: contribution
  - Location: Introduction; Research in context; Discussion, prior trials
- **Prospective central review supports treatment consistency** Participating centres completed benchmark planning cases, and individual radiotherapy plans required central approval before treatment. Web-based randomisation and ITT analysis further support the primary comparison.
  - Dimension: methodological rigour
  - Location: Methods, Randomisation and masking, Procedures, Statistical analysis
- **Clinical interpretation explains reasons to defer** The Discussion notes that 70% of observed participants remained recurrence-free and identifies clinical circumstances in which deferring radiotherapy remains reasonable.
  - Dimension: positioning
  - Location: Discussion, clinical interpretation and treatment trade-offs

## Limitations

- **Molecular independence exceeds the exploratory evidence** The conclusion asserts benefit regardless of methylation class, although molecular analyses were post-hoc, as-treated, and restricted to available tissue. Adjusted treatment estimates do not establish benefit in every molecular subgroup.
  - Dimension: positioning
  - Location: Methods, molecular statistical analysis; Results, molecular analyses; Discussion, final paragraph
- **Incomplete unblinded outcomes limit reassurance claims** Only 38% contributed five-year HRQoL data, and approximately 45% of 24-month cognitive assessments were missing. Unblinded patient reporting adds uncertainty, making the summary-box assertion of no differences too definitive.
  - Dimension: methodological rigour
  - Location: Methods, masking and Outcomes; Results, secondary outcomes; Research in context, Added value
- **Safety narrative contradicts reported serious events** The Discussion calls all early symptoms grade 1 or 2 despite reporting an early grade 3 radiation-related seizure. The Results also describe six serious events among nine observation-group participants, an impossible count as written.
  - Dimension: methodological rigour
  - Location: Discussion, adverse effects and serious adverse events; Results, serious adverse events

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - "Although almost half the patients reported new symptoms, these were expected and all were mild (grade 1 or 2)." (Discussion, adverse effects) vs "a grade 3 seizure and a grade 2 headache, both of which the patients recovered from." (Discussion, serious adverse events). The universal grade 1-or-2 description of early symptoms is contradicted by the explicitly reported early grade 3 radiation-related serious seizure. The claim is not supported by the paper's own data.
  - "events were mild, and there were no grade 3 or" (Discussion, paragraph on adverse events) vs "Grade 2 or 3 radiation-related serious adverse events occurred in five (8%) of 66 participants" (Abstract, Findings). The broad no-grade-3 wording conflicts with the separately reported grade 3 serious event, but does not invalidate the abstract's five-participant harms count. The result itself is unaffected.
  - Minor reporting notes (3):
    - "There were six serious adverse events reported by nine (10%) of 91 participants in the observation group (appendix p 34)." (Results, page 1479). Nine distinct affected participants require at least nine events, so six total events cannot account for nine affected participants. The effect on the result is unclear.
    - "There were 84 early (within 90 days of radiotherapy) radiation-related adverse reactions reported by 30 (45%) participants in the radiotherapy group" (Results, page 1478) vs "In our study, 30 (46%) of 66 participants in the radiotherapy group had early adverse events related to radiotherapy" (Discussion, page 1480). Thirty divided by 66 equals 45.45%, which rounds to 45% rather than 46% at whole-percentage precision. The result itself is unaffected.
    - "Time to second-line treatment was similar between the two groups" (Results, second-line treatment paragraph) vs "radiotherapy group 62·56 months [95% CI 59·21–65·90]" (Results, second-line treatment paragraph). The printed observation upper confidence limit of 56·92 months is below the radiotherapy lower limit of 59·21 months, an approximate comparison that cannot reconstruct the unspecified between-group test. The effect on the result is unclear.
- **Reporting and transparency:** Clean. Ethics approval, trial registration, competing interests, and controlled-access participant-data sharing are declared. No concern grounded in the supplied reporting establishes a research-conduct issue.
  - Conflict of interest: declared · Ethics approval: declared · Preregistration: declared · Data availability: declared
- **Reference integrity:** Clean
  - 38 references in manuscript
  - 36 of 37 checkable references found in an index
  - 1 are books, websites or datasets — counted, but not index-checkable
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Erratum published 21 Nov 2026**, listed in Crossref: https://doi.org/10.1016/s0140-6736(26)02308-1
- **Erratum published 10 Oct 2026**, listed in Crossref: https://doi.org/10.1016/s0140-6736(26)01998-7
- **Last checked:** 2026-11-20
