GuardianVets reports 34% of after-hours calls convert to primary care appointmentsMethadone infusions matched fentanyl for pain control in 42 dogs after spinal surgery18-element blood panel separated canine prostatic adenocarcinoma from controls in 115-dog studyToday's Veterinary Practice adds podcast interviews with specialistsNC State runs the only veterinary gerontology program in the worldVeterinary practices are testing AI on eight administrative tasksAssociates split on whether offering budget treatment plans compromises careIDEXX acquired CoVetAI, an ambient-listening software for veterinary practicesGuardianVets reports 34% of after-hours calls convert to primary care appointmentsMethadone infusions matched fentanyl for pain control in 42 dogs after spinal surgery18-element blood panel separated canine prostatic adenocarcinoma from controls in 115-dog studyToday's Veterinary Practice adds podcast interviews with specialistsNC State runs the only veterinary gerontology program in the worldVeterinary practices are testing AI on eight administrative tasksAssociates split on whether offering budget treatment plans compromises careIDEXX acquired CoVetAI, an ambient-listening software for veterinary practices
The Exam RoomBy The VeterinaryPracticeNow Desk4 min readSeptember 22, 2026
18-element blood panel separated canine prostatic adenocarcinoma from controls in 115-dog study
A case-control study built a Metallobalance index from 18 serum elements with 100% training-set sensitivity for canine PAC, but validation comes next.
A case-control study published in Frontiers in Veterinary Science measured serum concentrations of 18 trace elements and macrominerals in 115 dogs and built a linear discriminant score, the Metallobalance index, that showed high apparent training-set discrimination between prostatic adenocarcinoma and controls.
What the study found
Dogs with histopathologically confirmed prostatic adenocarcinoma (n = 24) had serum multi-element profiles that differed from a control group that the study describes as comprising "65 clinically healthy male dogs, 25 dogs with benign prostatic hyperplasia, and one dog with a prostatic cyst." An 18-element linear discriminant analysis model generated a scalar Metallobalance index (MBI) score. The study reports that "serum concentrations of 18 elements (Li, Na, Mg, P, S, K, Ca, V, Fe, Co, Cu, Zn, As, Se, Rb, Sr, Mo, and Cs)" were measured "by inductively coupled plasma-mass spectrometry (ICP-MS)." In the training set, the MBI was associated with high apparent discrimination between PAC and controls. The area under the receiver operating characteristic curve was 0.983. According to the study, "at a cutoff of 0.8030, sensitivity, specificity, and accuracy were 100.0, 92.3, and 93.9%, respectively." In dogs with PAC, the study reports that preoperative MBI was higher in dogs with lymph node metastasis than in dogs without lymph node metastasis. The MBI did not differ by surgical margin status (R0 versus R1 resection) and did not clearly change 2 weeks postoperatively.
How the study was built
This was a retrospective case-control study conducted at a single Japanese veterinary hospital. The PAC group comprised 24 dogs with histopathologically confirmed prostatic adenocarcinoma that underwent surgical excision between October 2012 and August 2021. The control group comprised 65 clinically healthy male dogs enrolled for wellness examinations between June 2017 and August 2023, along with 25 dogs with benign prostatic hyperplasia and one dog with a prostatic cyst, for a total of 91 controls. Serum samples were analyzed by ICP-MS for the 18 elements listed above. The study derived preliminary reference intervals for each element in the healthy cohort, then used all 18 elements in a linear discriminant analysis to generate the MBI.
Case-control designs can identify associations but cannot establish causation or temporal sequence. The reported performance metrics are training-set estimates. The model was built and tested on the same dogs, so the numbers overestimate real-world clinical performance. Independent prospective validation in a separate cohort is required before the test can be used diagnostically.
The number, and its error bars
At the MBI cutoff of 0.8030, the study reports that sensitivity was 100.0%, specificity 92.3%, and accuracy 93.9%. The area under the receiver operating characteristic curve was 0.983. The study does not report confidence intervals for these estimates. In a post hoc analysis restricted to controls aged 7 years or older, the apparent area under the curve was 0.984. In a comparison of BPH versus PAC using the original MBI equation without refitting, the apparent AUC was 0.998. The authors state that these are unvalidated training-set estimates and should not be interpreted as clinical performance.
What the authors say it cannot tell you
The authors write that "independent validation in a larger prospective cohort is required before clinical application." The study did not uniformly perform urinalysis in the healthy cohort, so subclinical urinary-tract disease could not be excluded as a confounder. The authors report that "findings did not support short-term postoperative monitoring or a validated prognostic role."
Who paid for it
The source does not state funding sources or conflicts of interest.
What this means for your practice
VeterinaryPracticeNow's read, not the study's. Everything in this section is our editorial interpretation for practice owners. The study does not claim any of it.
If this validates, it becomes the first noninvasive blood screen for canine prostatic adenocarcinoma, a disease that typically presents late, metastasizes early, and has no serum biomarker equivalent to human PSA. The test uses ICP-MS, a platform available at reference laboratories that offer trace-element and toxicology testing, which matters for adoption speed if the validation studies land.
The training-set accuracy is high, but the diagnostic tree waits on independent validation. A perfect-sensitivity claim built and tested on the same cohort is not the same number you will see when the test runs on the next ambiguous prostate case. Those numbers compress when the validation cohort shows up.
The exploratory lymph-node finding is worth tracking. If a preoperative blood panel can flag metastasis before opening the abdomen, it changes surgical planning and owner conversations on day one. The margin-status and postoperative-change analyses did not show signal, so this is not a monitoring tool yet.
