Oncology · Medical Literature
Selpercatinib versus multi-kinase inhibitors for advanced medullary thyroid cancer: A network meta-analysis of RET-targeted therapies.
Abstract
Several RET-targeted agents, including highly selective RET inhibitors (SRIs) and multi-kinase inhibitors (MKIs), have been approved for the treatment of advanced medullary thyroid cancer (MTC). Despite these agents targeting the same pathway, direct comparative data is lacking. This study aims to evaluate the relative efficacy and safety of these agents via network meta-analysis (NMA). We systematically searched PubMed, Embase, Cochrane Library, Web of Science, and CNKI (up to March 2026) for RCTs evaluating RET-targeted therapies for advanced MTC. The primary outcome was progression-free survival (PFS); secondary outcomes included objective response rate (ORR) and ≥Grade 3 adverse events (AEs). A frequentist random-effects model was employed, and treatments were ranked using P-scores. The study was registered in PROSPERO (CRD420261342793). Five RCTs (n = 1,076) were included. Selpercatinib demonstrated the most significant PFS benefit (HR = 0.10, 95% CI: 0.05-0.18; P-score = 0.999), outperforming cabozantinib (HR = 0.28) and vandetanib (HR = 0.46). It also ranked first for ORR (OR = 122.6, 95% CI: 34.5-435.6). Regarding safety, selpercatinib showed no significant difference in ≥Grade 3 AEs compared to placebo (OR = 1.34), whereas anlotinib exhibited the highest toxicity (OR = 12.00). Although selpercatinib was associated with hepatotoxicity (OR = 4.20), it avoided the off-target toxicities typical of MKIs, such as hypertension and diarrhea. This network meta-analysis demonstrates that selpercatinib exhibits superior efficacy (PFS and ORR) and a more favorable safety profile compared to MKIs for advanced MTC. These results support its role as a recommended first-line option for advanced MTC. However, due to the heterogeneity of included populations (mixed RET mutation status and therapy lines), clinicians should interpret these findings with caution regarding the generalizability to strictly defined RET-mutant populations. Clinical decisions should be individualized.Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261342793, identifier CRD420261342793.
Clinical relevance
Among patients with advanced medullary thyroid cancer (MTC), how do the selective RET inhibitor selpercatinib and various multi-kinase inhibitors (MKIs) compare in terms of progression-free survival, objective response rate, and grade ≥3 adverse events, as synthesized across available RCT evidence via network meta-analysis? __SNAPSHOT__ {"studyDesign":"Systematic review with network meta-analysis (NMA) of randomized controlled trials; frequentist random-effects model; treatment ranking by P-scores; registered in PROSPERO (CRD420261342793); adherent to PRISMA-NMA reporting guidelines","setting":"Multi-database systematic search (PubMed, Embase, Cochrane CENTRAL, Web of Science, Scopus, CNKI, ClinicalTrials.gov) from inception to March 2026; underlying RCTs conducted across international multi-center settings","sampleSize":"n = 1,076 across five RCTs (ZETA [vandetanib], EXAM [cabozantinib], ALTER01031 [anlotinib], LIBRETTO-531 [selpercatinib], EORTC-1209 [nintedanib]); individual trial sample sizes not reported in the available abstract","population":"Adults with advanced or metastatic medullary thyroid cancer enrolled in phase 3 RCTs evaluating RET-targeted therapies; baseline characteristics were generally well-balanced across trials, though the LIBRETTO-531 trial specifically enrolled RET-mutated, treatment-naïve patients while other trials enrolled broader or previously treated populations","intervention":"RET-targeted agents classified as: (1) Selective RET inhibitors (SRIs): selpercatinib, pralsetinib; (2) Multi-kinase inhibitors (MKIs): vandetanib, cabozantinib, anlotinib, nintedanib, sorafenib","comparator":"Placebo or best supportive care (BSC) as the common network node; LIBRETTO-531 used cabozantinib or vandetanib as active comparator and was indirectly linked to the placebo node via bridging estimation, forming a star-shaped network geometry","primaryOutcome":"Progression-free survival (PFS); secondary outcomes: objective response rate (ORR) and incidence of ≥Grade 3 adverse events (AEs)","followUp":"Not reported at the NMA level; individual trial follow-up varied (LIBRETTO-531 had a median follow-up of 12 months in the source phase 3 trial)","fundingSource":"Not reported in the available abstract or article metadata"}
Key findings
Five RCTs (n = 1,076) were included in the network. For the primary endpoint of PFS, selpercatinib demonstrated the highest P-score ranking (P-score = 0.999) with an HR of 0.10 (95% CI: 0.05–0.18) versus placebo, representing the most favorable PFS estimate across all agents evaluated. In indirect comparisons, selpercatinib outperformed cabozantinib (HR = 0.28 vs. placebo) and vandetanib (HR = 0.46 vs. placebo). NNT for PFS benefit is not calculable from the reported HRs alone without median PFS data. For the secondary endpoint of ORR, selpercatinib also ranked first with an odds ratio of 122.6 (95% CI: 34.5–435.6) versus placebo. Regarding ≥Grade 3 adverse events, selpercatinib showed no statistically significant increase compared to placebo (OR = 1.34; significance not explicitly stated in the abstract), while anlotinib exhibited the highest toxicity burden (OR = 12.00 vs. placebo). A specific safety signal for hepatotoxicity was identified with selpercatinib (OR = 4.20 vs. placebo), though it appeared to avoid the off-target toxicities characteristic of MKIs, such as hypertension and diarrhea. No overall survival data were reported in this NMA. Absolute risk differences and NNH values are not calculable from the available data.
Practice implications
This NMA provides indirect comparative evidence—currently the highest level of evidence available in the absence of multi-arm head-to-head RCTs across all agents—suggesting that selpercatinib is superior to MKIs for both PFS and ORR in advanced MTC, with a more favorable grade ≥3 toxicity profile relative to MKIs, particularly anlotinib. These findings are consistent with and largely driven by the LIBRETTO-531 phase 3 trial. However, critical interpretive cautions apply: (1) The network is star-shaped (all arms connected only through a placebo node), meaning all cross-drug comparisons are entirely indirect and subject to transitivity assumptions that may not hold—patient populations, prior treatment status, RET mutation prevalence, disease stage, and trial eras differ substantially across the five source trials; (2) LIBRETTO-531 enrolled exclusively RET-mutated, treatment-naïve patients, whereas other included trials enrolled molecularly unselected or previously treated patients, introducing significant clinical heterogeneity that may inflate the apparent benefit of selpercatinib; (3) Overall survival data were not analyzed; (4) Pralsetinib, another SRI, was listed as an eligible intervention but does not appear to have contributed a qualifying RCT to the network, limiting conclusions about SRI class effects; (5) The authors explicitly note insufficient statistical power in this small network (<20%). Findings are most applicable to RET-mutated, treatment-naïve patients with advanced MTC; patients with RET wild-type MTC or prior kinase inhibitor exposure are not well represented. The hepatotoxicity signal with selpercatinib (OR = 4.20) warrants prospective liver function monitoring and should inform consent discussions. Practice change toward selpercatinib as preferred first-line therapy for RET-mutated advanced MTC is already supported by the direct LIBRETTO-531 data and existing guideline updates; this NMA adds contextual evidence supporting the broader class superiority of selective RET inhibition over MKIs across the available RCT landscape. NOTE: Only the abstract was available for full review; full NMA methodology, heterogeneity statistics, inconsistency testing, and publication bias assessments could not be independently evaluated. __LIMITATIONS__ Star-shaped network geometry: all indirect comparisons are routed through a single placebo node, precluding any direct multi-drug comparisons and amplifying uncertainty in pairwise estimates,Profound clinical heterogeneity: enrolled populations differ significantly in RET mutation status, prior treatment history, disease stage, and calendar era across the five source RCTs, potentially violating the transitivity assumption required for valid NMA inference,Small network with acknowledged limited statistical power (<20% by authors' own assessment), which substantially increases the risk of imprecise or unstable effect estimates,Bridging estimation used to connect LIBRETTO-531 (which used an active comparator, not placebo) to the placebo node introduces additional methodological assumptions and potential bias,Overall survival not analyzed as an outcome, limiting conclusions about the most clinically meaningful endpoint in oncology,Pralsetinib did not contribute a qualifying RCT to the efficacy network, preventing direct SRI vs. SRI comparison and limiting class-level conclusions,Open-label design of LIBRETTO-531 (and likely other included RCTs) introduces performance and detection bias in subjective endpoints,Publication bias and selective outcome reporting cannot be fully excluded given the small number of included trials,Generalizability is limited: results are most applicable to RET-mutated, treatment-naïve patients and may not extend to RET wild-type MTC, previously treated patients, or pediatric populations,Only the abstract was available for review; full heterogeneity statistics (I², Cochran Q), inconsistency tests, and sensitivity analyses could not be assessed __BOTTOM_LINE_CATEGORY__ confirmatory __BOTTOM_LINE__ This network meta-analysis of five RCTs (n = 1,076) provides indirect comparative evidence that selpercatinib achieves superior PFS (HR 0.10, 95% CI 0.05–0.18) and ORR versus MKIs for advanced MTC, with a more tolerable grade ≥3 adverse event profile, albeit with a hepatotoxicity signal requiring monitoring. The findings are consistent with the direct phase 3 LIBRETTO-531 data and support selpercatinib as the preferred first-line agent for RET-mutated advanced MTC, though clinicians should recognize that all cross-drug comparisons in this NMA are indirect, the network is small, and population heterogeneity across included trials is substantial. This study is best characterized as confirmatory and hypothesis-generating for the broader class advantage of selective RET inhibition, rather than providing new independent evidence sufficient to change current practice.