Predictive value of HDL cholesterol for cancer-associated venous thromboembolism during chemotherapy

The ICU-venous thromboembolism score and tumor grade can predict inhospital venous thromboembolism occurrence in critical patients with tumors

Background

Venous thromboembolism (VTE) is a threat to the prognosis of tumor patients, especially for critically ill patients. No uniform standard model of VTE risk for critically ill patients with tumors was formatted by now. We thus analyzed risk factors of VTE from the perspectives of patient, tumor, and treatment and assessed the predictive value of the ICU-VTE score, which consisted of six independent risk factors (central venous catheterization, 5 points; immobilization ≥ 4 days, 4 points; prior VTE, 4 points; mechanical ventilation, 2 points; lowest hemoglobin during hospitalization ≥ 90 g/L, 2 points; and baseline platelet count > 250,000/μL, 1 points).

Methods

We evaluated the data of tumor patients admitted to the intensive care unit of the Peking University Cancer Hospital between November 2011 and January 2022; 560 cases who received VTE-related screening during hospitalization were chosen for this retrospective study.

Results

The inhospital VTE occurrence rate in our cohort was 55.7% (312/560), with a median interval from ICU admission to VTE diagnosis of 8.0 days. After the multivariate logistic regression analysis, several factors were proved to be significantly associated with inhospital VTE: age ≥ 65 years, high tumor grade (G3–4), medical diseases, fresh frozen plasma transfusion, and anticoagulant prophylaxis. The medium-high risk group according to the ICU-VTE score was positively correlated with VTE when compared with the low-risk group (9–18 points vs. 0–8 points; OR, 3.13; 95% CI, 2.01–4.85, P < 0.001). The AUC of the ICU-VTE scores according to the ROC curve was 0.714 (95% CI, 0.67–0.75, P < 0.001).

Conclusions

The ICU-VTE score, as well as tumor grade, might assist in the assessment of inhospital VTE risk for critically ill patients with tumors. The predictive accuracy might be improved when combining two of them; further follow-up researches are needed to confirm it.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-022-02705-z.

Red Cell Distribution Width-to-High-Density Lipoprotein Cholesterol Ratio (RHR): A Promising Novel Predictor for Preoperative Deep Vein Thrombosis in Geriatric Patients with Hip Fracture

Background

Deep vein thrombosis (DVT) is a devastating complication in geriatric patients before hip fracture surgery, and the predictive value of red cell distribution width (RDW) and high-density lipoprotein cholesterol (HDL-C) for DVTs after hip fracture remains to be established. This study aimed to assess the predictive value of RDW, HDL-C, and RDW-to-HDL-C ratio (RHR) in preoperative DVTs screening.

Methods

We retrospectively analyzed the data of geriatric patients (≥65 years old) admitted for hip fracture surgery between 2015 and 2020. The receiver operating characteristic (ROC) curve and related parameters were used to evaluate the predictive value of the biomarkers. Patients were divided into two groups according to the cutoff value of RHR, and propensity score matching (PSM) and subgroup analyses were performed to assess the true correlations between RHR and DVT.

Results

Among 2566 eligible patients included, we identified RDW with the area under ROC curve (AUC) of 0.532, cut-off value of 15.89, specificity of 88.2%, sensitivity of 18.2%, HDL-C with AUC of 0.574, cut-off value of 1.20, specificity of 55.6%, sensitivity of 59.3%, and RHR with AUC of 0.578, cut-off value of 13.45, specificity of 71.3%, sensitivity of 43.4%. RHR (>13.45) was independently associated with 1.54-fold risk (95% CI: 1.11–2.14, P=0.011) of DVTs among the post-PSM cohort. And compared with the counterparts, the relative risk of RHR associated with DVT was higher in the subgroups of aged 65–79 years (1.61 vs 1.45), non-hypoproteinemia (2.70 vs 1.29), non-diabetic (1.58 vs 1.41), non-hypertension (2.40 vs 1.06), ASA score I-II (2.38 vs 1.04), and femoral neck fracture (1.70 vs 1.50).

Conclusion

RDW, HDL-C and RHR were valuable biomarkers in predicting preoperative DVTs in geriatric patients with hip fracture, and RHR would be more efficient in the subgroups of younger age, better medical condition or femoral neck fracture.

Cancer-Related Venous Thromboembolism: From Pathogenesis to Risk Assessment

Cancer-related venous thromboembolism (VTE) remains a major health problem, accounting for at least 18% of all cases of VTE. Cancer patients with VTE have worse prognosis than those without VTE. Prophylaxis reduces VTE risk, but it is not feasible for all outpatients with cancer due to an increased bleeding risk. The factors involved in the pathogenesis of cancer-related VTE are direct coagulation activation, platelet activation, induction of inflammatory responses, and inhibition of fibrinolysis. Direct coagulation activation can be due to cancer procoagulant (a cysteine protease), microvesicles, or other prothrombotic abnormalities. Risk factors for developing VTE in cancer patients can be divided into four groups: tumor-related risk factors, patient-related risk factors, treatment-related risk factors, and biomarkers. Cancers of the pancreas, kidney, ovary, lung, and stomach have the highest rates of VTE. Patient-related risk factors such as age, obesity, or the presence of medical comorbidities can contribute to VTE. Platinum-based chemotherapies and antiangiogenesis treatments have also been associated with VTE. Biomarkers identified as risk factors include high platelet count, high leukocyte count, P-selectin, prothrombin fragments, D-dimer, and C-reactive protein. Based on the known risk factors, risk assessment models were developed to stratify patients who would benefit from thromboprophylaxis. The Khorana model was the first and is still the most widely used model. Because of its low sensitivity for certain tumor types, four new models have been developed in recent years. In this review, we describe the current knowledge about the pathogenesis and risk factors for cancer-related VTE, hoping to contribute to further research on the still many obscure aspects of this topic.

From Classical Laboratory Parameters to Novel Biomarkers for the Diagnosis of Venous Thrombosis

Venous thrombosis is a common and potentially fatal disease, because of its high morbidity and mortality, especially in hospitalized patients. To establish the diagnosis of venous thrombosis, in the last years, a multi-modality approach that involves not only imaging modalities but also serology has been evolving. Multiple studies have demonstrated the use of some biomarkers, such as D-dimer, selectins, microparticles or inflammatory cytokines, for the diagnosis and treatment of venous thrombosis, but there is no single biomarker available to exclusively confirm the diagnosis of venous thrombosis. Considering the fact that there are some issues surrounding the management of patients with venous thrombosis and the duration of treatment, recent studies support the idea that these biomarkers may help guide the length of appropriate anticoagulation treatment, by identifying patients at high risk of recurrence. At the same time, biomarkers may help predict thrombus evolution, potentially identifying patients that would benefit from more aggressive therapies. This review focuses on classic and novel biomarkers currently under investigation, discussing their diagnostic performance and potential benefit in guiding the therapy for venous thrombosis.

Predicting VTE in Cancer Patients: Candidate Biomarkers and Risk Assessment Models

Risk prediction of chemotherapy-associated venous thromboembolism (VTE) is a compelling challenge in contemporary oncology, as VTE may result in treatment delays, impaired quality of life, and increased mortality. Current guidelines do not recommend thromboprophylaxis for primary prevention, but assessment of the patient's individual risk of VTE prior to chemotherapy is generally advocated. In recent years, efforts have been devoted to building accurate predictive tools for VTE risk assessment in cancer patients. This review focuses on candidate biomarkers and prediction models currently under investigation, considering their advantages and disadvantages, and discussing their diagnostic performance and potential pitfalls.

Validation of a Machine Learning Approach for Venous Thromboembolism Risk Prediction in Oncology

Using kernel machine learning (ML) and random optimization (RO) techniques, we recently developed a set of venous thromboembolism (VTE) risk predictors, which could be useful to devise a web interface for VTE risk stratification in chemotherapy-treated cancer patients. This study was designed to validate a model incorporating the two best predictors and to compare their combined performance with that of the currently recommended Khorana score (KS). Age, sex, tumor site/stage, hematological attributes, blood lipids, glycemic indexes, liver and kidney function, BMI, performance status, and supportive and anticancer drugs of 608 cancer outpatients were all entered in the model, with numerical attributes analyzed as continuous values. VTE rate was 7.1%. The VTE risk prediction performance of the combined model resulted in 2.30 positive likelihood ratio (+LR), 0.46 negative LR (−LR), and 4.88 HR (95% CI: 2.54–9.37), with a significant improvement over the KS [HR 1.73 (95% CI: 0.47–6.37)]. These results confirm that a ML approach might be of clinical value for VTE risk stratification in chemotherapy-treated cancer outpatients and suggest that the ML-RO model proposed could be useful to design a web service able to provide physicians with a graphical interface helping in the critical phase of decision making.

Risk Assessment for Venous Thromboembolism in Chemotherapy-Treated Ambulatory Cancer Patients

OBJECTIVE

To design a precision medicine approach aimed at exploiting significant patterns in data, in order to produce venous thromboembolism (VTE) risk predictors for cancer outpatients that might be of advantage over the currently recommended model (Khorana score).

DESIGN

Multiple kernel learning (MKL) based on support vector machines and random optimization (RO) models were used to produce VTE risk predictors (referred to as machine learning [ML]-RO) yielding the best classification performance over a training (3-fold cross-validation) and testing set.

RESULTS

Attributes of the patient data set ( n = 1179) were clustered into 9 groups according to clinical significance. Our analysis produced 6 ML-RO models in the training set, which yielded better likelihood ratios (LRs) than baseline models. Of interest, the most significant LRs were observed in 2 ML-RO approaches not including the Khorana score (ML-RO-2: positive likelihood ratio [+LR] = 1.68, negative likelihood ratio [-LR] = 0.24; ML-RO-3: +LR = 1.64, -LR = 0.37). The enhanced performance of ML-RO approaches over the Khorana score was further confirmed by the analysis of the areas under the Precision-Recall curve (AUCPR), and the approaches were superior in the ML-RO approaches (best performances: ML-RO-2: AUCPR = 0.212; ML-RO-3-K: AUCPR = 0.146) compared with the Khorana score (AUCPR = 0.096). Of interest, the best-fitting model was ML-RO-2, in which blood lipids and body mass index/performance status retained the strongest weights, with a weaker association with tumor site/stage and drugs.

CONCLUSIONS

Although the monocentric validation of the presented predictors might represent a limitation, these results demonstrate that a model based on MKL and RO may represent a novel methodological approach to derive VTE risk classifiers. Moreover, this study highlights the advantages of optimizing the relative importance of groups of clinical attributes in the selection of VTE risk predictors.