Clinical genitourinary cancer · 2026
Prognostic Impact of Diagnostic Delay in Testicular Cancer: A Single-center Mexican Cohort Study With Machine Learning Risk Prediction.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Cantellano-Orozco Mauricio, Anceno Alec, Silva-Mendoza César A, Garzón-Ortega Victor H, Hosseini Hirad, Angulo-Sánchez José R, Magdaleno-Rodríguez Daniel R, Cervantes-Zorrilla Ricardo, Martínez-Arroyo Carlos, Fernández-Noyola Gerardo, Morales-Montor Gustavo, Ortega-González Mario, Pacheco-Gahbler Carlos
Аннотация
INTRODUCTION: Testicular cancer (TC), though highly curable, can carry significant morbidity and mortality when diagnosis is delayed. Herein, we characterize the distribution and prognostic impact of diagnostic delay on stage at presentation, disease progression, and mortality in a Mexican TC cohort.
PATIENTS: We retrospectively analyzed 223 men with histologically confirmed TC treated at a single tertiary referral center in Mexico City between January 2017 and January 2025. The analyzable cohort was defined after excluding patients with incomplete clinical data or fewer than 1 month of post-orchiectomy follow-up.
METHODS: The primary analysis evaluated associations between diagnostic delay and clinical outcomes, using penalized regression for mortality because of complete separation. Secondarily, 5 machine-learning models were cross-validated to assess whether routine clinical variables improved prediction beyond established prognostic factors.
RESULTS: Median age was 26 years; 53.4% had non-seminomatous histology and 25.6% had diagnostic delay > 6 months. All 13 deaths occurred in stage IIIC disease, while 56/57 delayed patients progressed. Mortality was associated with delay > 6 months (OR 10.3), poor-risk IGCCCG disease (OR 17.2), and AFP > 10,000 ng/mL (OR 5.3). Cross-validated AUCs ranged from 0.830 to 0.863, with no consistently superior model.
CONCLUSION: Diagnostic delay > 6 months was strongly associated with advanced disease, progression, and mortality, although this association was largely mediated by stage at presentation. Machine learning models did not identify prognostic information beyond established clinical factors.
PATIENTS: We retrospectively analyzed 223 men with histologically confirmed TC treated at a single tertiary referral center in Mexico City between January 2017 and January 2025. The analyzable cohort was defined after excluding patients with incomplete clinical data or fewer than 1 month of post-orchiectomy follow-up.
METHODS: The primary analysis evaluated associations between diagnostic delay and clinical outcomes, using penalized regression for mortality because of complete separation. Secondarily, 5 machine-learning models were cross-validated to assess whether routine clinical variables improved prediction beyond established prognostic factors.
RESULTS: Median age was 26 years; 53.4% had non-seminomatous histology and 25.6% had diagnostic delay > 6 months. All 13 deaths occurred in stage IIIC disease, while 56/57 delayed patients progressed. Mortality was associated with delay > 6 months (OR 10.3), poor-risk IGCCCG disease (OR 17.2), and AFP > 10,000 ng/mL (OR 5.3). Cross-validated AUCs ranged from 0.830 to 0.863, with no consistently superior model.
CONCLUSION: Diagnostic delay > 6 months was strongly associated with advanced disease, progression, and mortality, although this association was largely mediated by stage at presentation. Machine learning models did not identify prognostic information beyond established clinical factors.