Background: Selpercatinib, a potent/selective RET inhibitor, is approved for the treatment of RET fusion-positive non-small cell lung cancers. While the drug is known to have substantial intracranial activity (intracranial ORR 82%) in patients with existing brain metastases, (1) central nervous system (CNS) outcomes in patients without brain metastases and (2) CNS progression patterns in patients with brain metastases have not been explored. Methods: Patients with advanced RET fusion-positive lung cancers were prospectively treated with selpercatinib on the registrational LIBRETTO-001 trial (NCT03157128) or the LIBRETTO-201 multi-center expanded access program (EAP, NCT03906331). Key overall and CNS eligibility criteria were previously presented. Patients with and without pre-selpercatinib brain metastases who underwent serial CNS and extracranial imaging were eligible for analysis. The data cutoff was November 24, 2020. Cumulative incidence rates (CIRs) were calculated using a competing risk model with systemic progression of disease (PD) or death as competing risks; patients with simultaneous CNS and systemic PD were treated as having had CNS PD. Results: Sixty-two patients (48 LIBRETTO-001, 14 EAP) were identified. Median age was 64. Thirty-two (52%) were female and 47 (76%) were never smokers. The most common 5’ fusion partner was KIF5B (68%). The median number of prior therapies was 2 (range 1-11); 34% received prior multikinase inhibitor therapy. The median time on treatment was 21.8 months. Thirty-one (50%) patients had no baseline brain metastases. In these patients, the CIR of CNS metastasis was 0% at 6 months and 12 months; none of these patients developed CNS metastasis during selpercatinib treatment. The 9 patients that progressed did so extracranially. Of the 31 patients with baseline brain metastases, 12 (39%) had prior CNS radiation and 3 (10%) had prior CNS surgery. At the time of data cut-off, 23 patients had some evidence of progression, including 8 in both the CNS and systemically, 6 only in the CNS, and 9 only systemically. Overall, 17 of the 31 patients with baseline brain metastasis did not develop evidence of CNS progression as of the data cut. Among patients with baseline brain metastasis, the CIR for evidence of CNS PD was 6.7% at 6 months and 27.4% at 12 months. Conclusions: In patients with RET fusion-positive lung cancers without baseline brain metastases, new CNS metastases were not observed during selpercatinib therapy. Among patients with baseline brain metastasis, a substantial number did not experience progression in the CNS on treatment.