Bone-RADS validated: high sensitivity for malignancy in adult lytic long-bone lesions

Design

Retrospective diagnostic validation, four independent raters

Population

423 adult extremity long-bone lytic lesions, mean age 62

Primary outcome

Malignant diagnosis by biopsy or 2-year radiographic stability

Effect

OR 9.7 per point (95% CI 6.2 to 14.9); sensitivity 97.1%, specificity 67.4%

Bone-RADS is an American College of Radiology system that scores lytic bone lesions using five radiographic features and any history of primary cancer. This retrospective study externally validated it on 423 adult extremity long-bone lesions identified from radiograph reports in a single year, scored independently by four reviewers. Lesions without biopsy were counted benign if stable for two years.

Each one-point increase in score was associated with an odds ratio of 9.7 (95% CI 6.2 to 14.9) for malignancy. Margin score, endosteal erosion, pathological fracture and a soft-tissue mass each carried independent weight. Sensitivity was 97.1%, specificity 67.4%, positive predictive value 85.2% and negative predictive value 92.4%.

Agreement between reviewers on the full four-tier score was moderate (weighted kappa 0.52), but the more practical split into possibly benign versus possibly malignant was more reproducible.

For orthopaedic surgeons who see incidental lytic lesions referred from fracture clinic or primary care, the score offers a structured first triage. It does not replace a musculoskeletal radiologist or biopsy planning at a sarcoma unit.

• Use the five Bone-RADS features to structure the description of any lytic long-bone lesion.
• Treat a soft-tissue mass, endosteal erosion or pathological fracture as features that raise concern.
• Refer any lesion scoring as possibly malignant to a sarcoma or bone tumour unit before biopsy.
• A low score reassures but does not exclude malignancy: about 1 in 13 low-risk calls was wrong.
• Record any history of primary cancer, which is part of the score.

Why it matters

A shared scoring language for lytic lesions could cut both missed sarcomas and unnecessary biopsies of benign lesions.

Don't overread it

This was a single-centre retrospective validation with a high proportion of malignant lesions, so performance in general practice may differ.

The statistics, in plain English

A sensitivity of 97% means very few cancers were missed, while a specificity of 67% means about a third of benign lesions were flagged for further work-up. Predictive values depend on how common malignancy is in the population tested, so they will be lower in a general fracture clinic than in this referral sample.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.