Khorana and PROTECHT scores in predicting the risk of venous thromboembolism in pancreatic cancer: Which performed better?

Venous thromboembolism (VTE) represents a major cause of morbidity and mortality in cancer patients. Several scores have been developed to predict the risk of cancer-associated VTE and help clinicians to select patients for thromboprophylaxis. The best known is the Khorana score, which stratifies patients as low (0 points), intermediate (1-2 points) and high (>= 3 points) risk. Pancreatic cancer patients score at least an intermediate risk due to tumor location parameter. Previous studies concluded that the Khorana score was not able to discriminate between intermediate and high-risk cancer patients, and the incidence of VTE was instead higher in intermediate-risk patients. Other scores have been developed to serve as better tools to stratify VTE risk in cancer patients. The PROTECHT score accounts for the chemotherapy regimen to be initiated. Some analyses suggest that the PROTECHT score provides better discrimination between low and high-risk patients, but further investigation is needed. The aim of this study is to compare the Khorana and PROTECHT scores in discriminating VTE risk in a cohort of pancreatic cancer patients, as well as to assess other potential risk factors for VTE. This is a monocentric, retrospective study of 91 patients with pancreatic cancer with I-IV stage disease (AJCC 8th ed.) that received any systemic treatment with neoadjuvant, adjuvant or palliative intention between 2016 and 2019 at an Oncology department. Exclusion criteria: absence of chemotherapy/other systemic treatment; ongoing anticoagulation previous to oncological treatment; presence of any hematologic disease. Khorana and PROTECHT scores were calculated according to literature. In our sample, most patients were female (54,9%, n=50), median age at diagnosis was 70 years old [34-89] and 82,4% (n=75) had a stage III-IV disease. Median time of follow-up was 7,59 months. Median overall survival was 10,72 months. According to Khorana score, 62,6% (n=57) were classified as intermediate risk and 37,4% (n=34) as high risk. According to PROTECHT score, 97,8% (n=89) were stratified as high risk. Thromboprophylaxis was initiated in only 5 patients. 22 VTE events were documented, 40,9% (n=9) in deep veins of the lower limbs and 31,8% (n=7) in the mesenteric venous system. For Khorana intermediate-risk patients, the odds of a VTE event was 0,39 versus 0,21 for high risk [OR 0,537, CI 95% 0,19-1,57]. The odds of VTE for PROTECHT high-risk patients was 0,32. None of low-risk patients had an event. No other clinical and pathological parameters were significantly associated with VTE risk. In our analysis, the Khorana score failed to discriminate the risk of VTE between intermediate and high-risk patients. The odds of patients classified as intermediate risk to develop a thrombotic event was 1,86 times higher than those classified as high risk, with the majority of events occurring in the first group. The PROTECHT score performed better at stratifying our patients. Other parameters such as staging, histological subtype, grade, and tumor markers were not predictive of VTE which can be due to our limited number of events. These findings would merit further validation in larger prospective studies.