Validation and revision of the RANO Leptomeningeal Metastasis Group scorecard for response assessment.

Background: The Response Assessment in Neuro-Oncology (RANO) group has proposed a scorecard to evaluate response assessed by magnetic resonance imaging (MRI) during treatment of leptomeningeal metastasis (LM). Methods: To validate the LM-RANO scorecard, cerebrospinal MRI of 22 patients with LM from solid tumors were rated by 10 neuro-oncologists and 9 neuroradiologists at baseline and during follow-up after treatment. The original scorecard and its instructions were sent to the raters. The inter-observer agreement for the evaluation of single items was assessed using the Krippendorff alpha coefficient. The agreement between numerical parameters including scores at baseline and at follow-up after treatment and total scores was determined using the intraclass coefficient of correlation (ICC). Results: Instructions of the scorecard were not clear for the majority of raters. The alpha concordance coefficients obtained for the rating of single items at baseline and at follow-up were low (Krippendorf alpha coefficient < 0.61 for all raters). The best concordances were noted for spinal nodules (0.60). The worst concordances were obtained for brain linear leptomeningeal enhancement and cranial nerve enhancement. The concordance was better among neuroradiologists than among neuro-oncologists. A poor agreement was also noted when evaluating changes between baseline and follow-up and for total scores (ICC < 0.65 for the best score for all raters). Conclusions: Assessing response of LM by MRI remains challenging. The definition of a measurable lesion and the determination of response were the most important challenges. A central review is therefore recommended for clinical trials in LM. Based on this study we propose a new scorecard that will require a similar assessment as conducted here.