Evaluation of cyclin-dependent kinase 4 immunolabelling and a germline p.G201S variant as predictors of clinical outcome in canine mammary tumours: a preliminary study.

Cyclin-dependent kinase 4 (CDK4), a regulator of the G1/S transition, is frequently dysregulated in cancer. Although extensively studied in human breast cancer, its significance in canine mammary tumours remains poorly understood. This preliminary study investigated the prognostic value of CDK4 immunohistochemistry (IHC) and of sequence variation in the CDK4 kinase-domain segment surrounding residue 201. Thirty-three prospectively enrolled female dogs with tumours affecting a single mammary chain underwent unilateral radical mastectomy. An 89-bp CDK4 segment was sequenced and paired with CDK4 and oestrogen receptor (ER) IHC and clinicopathological follow-up. Thirty dogs (11 deaths) entered the primary analyses (three were suboptimal). Missense variant c.601G > A (p.G201S) was identified in seven of 30 dogs (23%; allele frequency 15%) but not associated with any clinicopathological variable or survival (hazard ratio [HR] 1.54, 95% confidence interval [CI] 0.39-6.00). CDK4 immunolabelling was associated with ER immunolabelling (odds ratio 8.75, 95% CI 1.21-63.4; P = 0.058). CDK4 immunolabelling was not associated with survival (HR 2.73, 95% CI 0.78-9.48; P = 0.11); after adjustment for advanced stage, it reached only borderline significance (HR 3.92, 95% CI 1.01-15.2; P = 0.048), but this did not survive a worst-case sensitivity analysis. The 12-month time-dependent receiver operating characteristic area was 0.79 (95% CI 0.56-0.96) and the negative predictive value (NPV) 93%. p.G201S most likely represents a common germline polymorphism rather than a driver event. CDK4 immunolabelling tracked the ER-positive phenotype but was not an independent prognostic marker; its high short-term NPV suggests a rule-out tool requiring validation in a larger cohort.