NRG Oncology investigators analyzed clinical and digital histopathology data from five phase III prostate cancer trials (NRG/RTOG 9202, 9408, 9413, 9910, and 0126) to develop and validate multimodal artificial intelligence models (MMAI) that could outperform the National Comprehensive Cancer Network® (NCCN®) Clinical Practice Guidelines in Oncology in the prediction of distant metastasis and other outcomes. The researchers found that MMAI models could, in fact, stratify patients into risk groups that more accurately reflected their prognosis in comparison to NCCN risk groups as defined in the Guidelines. These results were presented by Jonathan D. Tward, MD, PhD, and colleagues during the Plenary Session of the 2022 American Society for Radiation Oncology (ASTRO) Annual Meeting (Abstract 2).
Jonathan D. Tward, MD, PhD
"Patients with localized prostate cancer are known to exhibit highly variable prognoses. Based on the NCCN risk stratification system, treatment guidelines result in overtreatment for many [patients]. Conversely, it also understates the risk in some [patients] who would benefit from treatment intensification. We are thrilled to have developed a risk-stratification tool that offers this patient population a more accurate prognosis. As a result, patients and their physicians can tailor the intensity of their treatment to meet the patients’ goals better and improve outcomes," stated lead study author Dr. Tward, of the University of Utah.
Performance of the MMAI Model
In the analysis, the previously locked MMAI 5-year metastasis model was applied across 5,569 patients, and deciles with similar distant metastasis rates were binned into three tiers: favorable (1–6th decile), moderate (7–9th), and unfavorable (10th) risk. Patients unable to be classified into NCCN risk groups were excluded. The Fine-Gray competing risk method was used to estimate the cumulative incidence of metastasis. The three-tier MMAI model was compared to standard prognostic factors such as prostate-specific antigen (PSA), grade group, and three-tier NCCN.
“The MMAI model identified sixfold more patients with a lower risk of distant metastasis than NCCN risk groups. Conversely, patients in the MMAI unfavorable-risk group had a substantially greater risk of distant metastasis than the NCCN high-risk group. Men in the lower-risk group could routinely consider radiation therapy alone sufficient and avoid the added toxicity of using androgen-deprivation therapy combined with radiation. The degree of prognostic improvement seen in the three-tier MMAI stratification is striking, and strongly points toward the MMAI models becoming a new standard supported by level 1 evidence for personalized risk assessment in newly diagnosed [patients] contemplating their treatment options," added Dr. Tward.
At a median follow-up for censored patients of 11.4 years, the median PSA was 8.4 (interquartile range = 5.8–12), 14 (interquartile range = 8.8–25), and 32 (interquartile range = 16–63) ng/mL for MMAI favorable, moderate, and unfavorable risks. Of the 3,829 patients with grade 1/2 disease, 71%, 25%, and 4% were stratified as MMAI favorable, moderate, and unfavorable risk, respectively. Similarly, the 867 grade group 3 patients were stratified as 63%, 30%, and 7%, respectively, and the 716 grade group 4/5 patients were stratified as 7%, 50%, and 43%, respectively. NCCN classified 584 (10%) patients as low-, 3,060 (55%), as intermediate-, and 1,925 (35%) as high-risk, with estimated 5-year distant metastasis rates of 1% (95% confidence interval [CI] 0%–2%), 3% (95% CI = 3%–4%), and 10% (95% CI = 9%–12%) respectively. In contrast, the MMAI model grouped 3,342 (60%) patients as favorable, 1,671 (30%) as moderate, and 556 (10%) as unfavorable risk, with estimated 5-year distant metastasis rates of 1% (95% CI = 1%–2%), 6% (95% CI = 5%–7%), and 28% (95% CI = 25%–32%).
The estimated 10-year distant metastasis rates were 3% (95% CI = 1%–4%), 6% (95% CI = 5%–7%), and 17% (95% CI = 15%–18%) for NCCN low-, intermediate-, and high-risk groups, respectively, and were 3% (95% CI = 3%–4%), 12% (95% CI = 10%–14%), and 37% (95% CI = 33%–41%) for MMAI risk groups. Within NCCN intermediate-risk, MMAI identified 83% favorable-risk patients with a 10-year distant metastasis rate of 4%, and within NCCN high-risk, MMAI identified 13% favorable-risk patients with a 10-year distant metastasis rate of 4%.
The study authors concluded, “The three-tier MMAI models integrating clinical and digital histopathology data demonstrated improved discrimination of distant metastasis risk compared to NCCN. The MMAI model identified sixfold more patients with a lower risk of distant metastasis compared to NCCN. Conversely, patients in the MMAI unfavorable-risk group had substantially greater distant metastasis risk than NCCN high-risk. Use of these MMAI models may improve prognostication, enabling more informed decision-making for patient care.”
Disclosure: This project was supported by the National Cancer Institute (NCI). This project was led by NRG Oncology with participation of other NCI-funded network groups. For full disclosures of the study authors, visit plan.core-apps.com/myastroapp2022.
