Primary Versus Secondary Non-Urothelial Tumors Involving the Bladder: A 10-Year Analysis of Clinicopathologic Profiles and Adverse Feature Burden.

Non-urothelial bladder tumors and secondary bladder involvement from extravesical primaries are uncommon but clinically challenging. We compared clinicopathologic patterns between primary non-urothelial tumors and secondaries, and explored correlates of adverse pathologic features to inform diagnostic triage and surgical planning. We performed a single-center retrospective cohort (2014-2024) of consecutive bladder lesions meeting WHO 2022 criteria and AJCC 8th staging. Eligible cases were primary non-urothelial malignancies (squamous cell carcinoma (SCC), adenocarcinoma (ADK), small-cell/neuroendocrine (NEC), sarcomatoid) or secondary bladder involvement (colorectal, prostate, cervix, ovary, uterus, breast). Outcomes included advanced pT (≥pT3), lympho-vascular invasion (LVI), perineural invasion (PNI), nodal metastasis, margin status, and composite adverse events. Of 235 analyzable cases, 59 were primary and 176 were secondary. Age and sex distributions were similar. Secondaries had a higher adverse burden: advanced pT 56.8% vs. 23.7%, LVI 47.2% vs. 27.1%, PNI 40.3% vs. 22.0%, node-positive 11.9% vs. 0%, and any adverse 65.3% vs. 33.9% (all significant). Histology composition differed (p < 10-6): secondaries were ADK-dominant (59.1%), whereas primaries were enriched for SCC (38.5%), sarcomatoid (28.8%), and NEC (21.2%). Among secondaries, prostate origin showed the most ominous profile (advanced pT 97.5%, PNI 77.5%, positive margins 64.7%); colorectal cases combined high advanced pT (70.2%) with lower margin positivity (27.6%). Adverse-feature count correlated with pT (ρ = 0.586). Secondary bladder involvement carries substantially higher adverse-pathology rates than primary non-urothelial tumors, with origin-specific risk gradients (prostate > colorectal ≳ cervix). Rigorous origin adjudication and a margin-focused, anatomy-adapted surgical strategy may improve outcomes; prospective outcome-linked validation is warranted.