BACKGROUND: Acute ischemic stroke with isolated posterior cerebral artery occlusion (iPCAO) lacks management evidence from randomized trials. We aimed to evaluate whether the association between endovascular treatment (EVT) and outcomes in iPCAO acute ischemic stroke is modified by initial stroke severity (baseline National Institutes of Health Stroke Scale [NIHSS]) and arterial occlusion site. METHODS: Based on the multicenter, retrospective, case-control study of consecutive iPCAO acute ischemic stroke patients (PLATO study [Posterior Cerebral Artery Occlusion Stroke]), we assessed the heterogeneity of EVT outcomes compared with medical management (MM) for iPCAO, according to baseline NIHSS score (≤6 versus >6) and occlusion site (P1 versus P2), using multivariable regression modeling with interaction terms. The primary outcome was the favorable shift of 3-month modified Rankin Scale (mRS). Secondary outcomes included excellent outcome (mRS score 0–1), functional independence (mRS score 0–2), symptomatic intracranial hemorrhage, and mortality. RESULTS: From 1344 patients assessed for eligibility, 1059 were included (median age, 74 years; 43.7% women; 41.3% had intravenous thrombolysis): 364 receiving EVT and 695 receiving MM. Baseline stroke severity did not modify the association of EVT with 3-month mRS distribution ( P interaction =0.312) but did with functional independence ( P interaction =0.010), with a similar trend on excellent outcome ( P interaction =0.069). EVT was associated with more favorable outcomes than MM in patients with baseline NIHSS score >6 (mRS score 0–1, 30.6% versus 17.7%; adjusted odds ratio [aOR], 2.01 [95% CI, 1.22–3.31]; mRS score 0 to 2, 46.1% versus 31.9%; aOR, 1.64 [95% CI, 1.08–2.51]) but not in those with NIHSS score ≤6 (mRS score 0–1, 43.8% versus 46.3%; aOR, 0.90 [95% CI, 0.49–1.64]; mRS score 0–2, 65.3% versus 74.3%; aOR, 0.55 [95% CI, 0.30–1.0]). EVT was associated with more symptomatic intracranial hemorrhage regardless of baseline NIHSS score ( P interaction =0.467), while the mortality increase was more pronounced in patients with NIHSS score ≤6 ( P interaction =0.044; NIHSS score ≤6: aOR, 7.95 [95% CI, 3.11–20.28]; NIHSS score >6: aOR, 1.98 [95% CI, 1.08–3.65]). Arterial occlusion site did not modify the association of EVT with outcomes compared with MM. CONCLUSIONS: Baseline clinical stroke severity, rather than the occlusion site, may be an important modifier of the association between EVT and outcomes in iPCAO. Only severely affected patients with iPCAO (NIHSS score >6) had more favorable disability outcomes with EVT than MM, despite increased mortality and symptomatic intracranial hemorrhage.