A three-year trial in Hong Kong found that personalized care for adults diagnosed with non-type 1 diabetes by age 40 improved several measures of diabetes distress during the first year. However, it did not significantly reduce the risk of new or worsening diabetes complications or death compared with usual clinic care.
The PRISM trial included 884 Chinese adults. Participants were assigned to either JADE-PRISM care, which combined genetic and biomarker testing with intensive specialist support, or a comparison group that received a JADE risk report and usual clinic care under local guidelines.
Complications were not significantly lower
The trial’s main outcome combined new or worsening heart disease, stroke, kidney disease, eye disease or nerve damage, along with death.
Within three years, this outcome occurred in 116 of 441 people in the JADE-PRISM group, or 26.3%, compared with 125 of 443 people in the JADE-only group, or 28.2%. The difference was not statistically significant. The reported odds ratio was 0.908, with a 95% confidence interval of 0.675 to 1.221.
The result remained nonsignificant after adjustments for baseline differences, in a per-protocol analysis and after excluding 68 people with autoimmune or monogenic diabetes. The trial had been designed to detect a 40% reduction in the combined complication outcome, but did not find that reduction within three years.
Distress improved more during the first year
JADE-PRISM provided one year of intensified team care at a specialist diabetes center. The program included visits every three months, questionnaire reviews, treatment algorithms aimed at strict targets, psychosocial support, and medication and monitoring reminders.
Testing included GAD antibodies, fasting C-peptide, 34 monogenic diabetes genes and three polygenic risk scores. Participants then received usual clinic care with yearly diabetes-team reviews during years two and three.
After one year, people in the JADE-PRISM group reported larger improvements in distress related to their treatment regimen, social support and physicians. They also scored higher on diabetes empowerment and reported better performance on several self-care habits, including overcoming barriers, eating to plan and checking blood sugar. Self-reported medication adherence did not differ between the groups.
By the third year, most differences in distress and related measures had faded. Physician-related distress remained significantly lower in the JADE-PRISM group, which also reported better perceived control over diabetes.
Some risk measures and treatment targets improved
At three years, average changes favored JADE-PRISM for HbA1c, systolic blood pressure and body weight. HbA1c changed by −0.2% in the personalized-care group compared with +0.1% in the comparison group. Systolic blood pressure changed by −6.6 mmHg versus −1.7 mmHg, and body weight by −1.8 kilograms versus −0.6 kilograms.
Nearly one-quarter of the JADE-PRISM group, 23.8%, reached at least three strict treatment targets at year three, compared with 12.2% of the JADE-only group. The targets included an HbA1c below 6.2% and blood pressure below 120/75 mmHg.
Medication use also differed. GLP-1 receptor agonists were used by 30.9% of JADE-PRISM participants and 14.9% of JADE-only participants. SGLT2 inhibitor use was 62.2% and 49%, respectively. Severe low blood sugar requiring hospital admission occurred once in the JADE-PRISM group and five times in the comparison group.
What the testing identified—and what it cannot show
The testing identified GAD antibodies in 46 participants, or 5.2%; monogenic diabetes in 23, or 2.6%; and low C-peptide in 82, or 9.3%. Within these subgroups, blood sugar measures did not differ by study group.
The findings do not establish that genetic testing itself reduced distress or improved risk factors. Testing was bundled with specialist care, more frequent visits, strict treatment goals, nurse support and other elements of the program, so the trial cannot determine which component helped.
The study was open-label: participants and clinicians knew which care they received, although data analysts were blinded to participant identity. That design could have influenced self-reported measures such as distress and empowerment. The trial also took place at one endocrinologist-led center in Hong Kong, among Chinese participants in a highly subsidized healthcare setting.
Results from the PRISM trial were published in Diabetes Care in 2026. The findings suggest that a broad, personalized support program may improve how some adults experience and manage young-onset diabetes in the short term, while showing no significant reduction in complications over the three-year follow-up.
AI tools were used to assist with the preparation of this article under our editorial standards.
Leave a Reply