Risk scores for occult cancer in patients with unprovoked venous thromboembolism: Results from an individual patient data meta-analysis

Development of a Predictive Model of Occult Cancer After a Venous Thromboembolism Event Using Machine Learning: The CLOVER Study

Background and Objectives: Venous thromboembolism (VTE) can be the first manifestation of an underlying cancer. This study aimed to develop a predictive model to assess the risk of occult cancer between 30 days and 24 months after a venous thrombotic event using machine learning (ML). Materials and Methods: We designed a case-control study nested in a cohort of patients with VTE included in a prospective registry from two Spanish hospitals between 2005 and 2021. Both clinically and ML-driven feature selection were performed to identify predictors for occult cancer. XGBoost, LightGBM, and CatBoost algorithms were used to train different prediction models, which were subsequently validated in a hold-out dataset. Results: A total of 815 patients with VTE were included (51.5% male and median age of 59). During follow-up, 56 patients (6.9%) were diagnosed with cancer. One hundred and twenty-one variables were explored for the predictive analysis. CatBoost obtained better performance metrics among the ML models analyzed. The final CatBoost model included, among the top 15 variables to predict hidden malignancy, age, gender, systolic blood pressure, heart rate, weight, chronic lung disease, D-dimer, alanine aminotransferase, hemoglobin, serum creatinine, cholesterol, platelets, triglycerides, leukocyte count and previous VTE. The model had an ROC-AUC of 0.86 (95% CI, 0.83-0.87) in the test set. Sensitivity, specificity, and negative and positive predictive values were 62%, 94%, 93% and 75%, respectively. Conclusions: This is the first risk score developed for identifying patients with VTE who are at increased risk of occult cancer using ML tools, obtaining a remarkably high diagnostic accuracy. This study's limitations include potential information bias from electronic health records and a small cancer sample size. In addition, variability in detection protocols and evolving clinical practices may affect model accuracy. Our score needs external validation.

Newly Diagnosed Cancer After Diagnosis of Venous Thromboembolism - Insights From the COMMAND VTE Registry-2

BACKGROUND

Previous randomized clinical trials did not support a benefit of screening for occult cancer after diagnosis of venous thromboembolism (VTE), although screening may be of potential benefit for selected high-risk patients.

METHODS AND RESULTS

The COMMAND VTE Registry-2 enrolled consecutive patients with acute symptomatic VTE between 2015 and 2020 from 31 centers across Japan. The 3,706 patients in the registry without known active cancer at the time of VTE diagnosis were divided into 2 groups: those with (n=250) and without (n=3,456) newly diagnosed cancer during the follow-up period. The cumulative incidence of newly diagnosed cancer was 1.5% at 30 days, 3.7% at 1 year, and 7.0% at 3 years. The multivariable Cox proportional hazard model demonstrated that older age (hazard ratio [HR] 1.02 per 1 year increase; 95% confidence interval [CI] 1.01-1.03; P<0.001), a history of cancer (HR 3.57; 95% CI 2.73-4.64; P<0.001), autoimmune disorders (HR 1.48; 95% CI 1.06-2.02; P=0.02), a history of major bleeding (HR 1.64; 95% CI 1.04-2.48; P=0.04), and the absence of transient provoking risk factors for VTE (HR 1.44; 95% CI 1.08-1.92; P=0.01) were independently associated with newly diagnosed cancer.

CONCLUSIONS

The incidence of newly diagnosed cancer after VTE diagnosis was 3.7% at 1 year, and several independent risk factors for newly diagnosed cancer after VTE diagnosis were identified.

External validation of the RIETE and SOME scores for occult cancer in patients with venous thromboembolism: a multicentre cohort study

INTRODUCTION

Venous thromboembolism (VTE) may be the first sign of an undiagnosed cancer. The RIETE and SOME scores aim to identify patients with acute VTE at high risk of occult cancer. In the present study, we evaluated the performance of both scores.

METHODS

The scores were evaluated in a retrospective cohort from two centers. The area under the receiver-operating characteristics curve (AUC) evaluated the discriminatory performance.

RESULTS

The RIETE score was applied to 815 patients with provoked and unprovoked VTE, of whom 56 (6.9%) were diagnosed with cancer. Of the 203 patients classified as high-risk, 18 were diagnosed with cancer, representing 32.1% (18/56) of the total cancer diagnoses. In the group of 612 low-risk patients, 67.9% of the cancer cases were diagnosed (38/56). Sensitivity, specificity, negative and positive predictive values, and AUC were 32%, 76%, 94%, 9%, and 0.430 (95% confidence interval [CI], 0.38‒0.47), respectively. The SOME score could be calculated in 418 patients with unprovoked VTE, of whom 33 (7.9%) were diagnosed with cancer. Of the 45 patients classified as high-risk, three were diagnosed with cancer, representing 9.1% (3/33) of the total cancer diagnoses. In the group of 373 low-risk patients, 90.9% of the cancer cases were diagnosed (30/33). Sensitivity, specificity, negative and positive predictive values, and AUC were 33%, 88%, 94%, 20%, and 0.351 (95% CI, 0.27‒0.43), respectively.

CONCLUSIONS

The performance of both scores was poor. Our results highlight the need to develop new models to identify high-risk patients who may benefit from an extensive cancer screening strategy.

Cardio-Oncology: A New Discipline in Medicine and Its Relevance to Hematology

Cardio-oncology, a burgeoning subspecialty, addresses the complex interplay between cardiology and oncology, particularly in light of increased cardiovascular (CV) disease mortality in cancer patients. This review provides a comprehensive overview of cardio-oncology with a focus on the therapies used in hematological malignancies. We explore the bidirectional relationship between heart failure and cancer, emphasizing the need for collaborative care. The review discusses risk stratification, highlighting the importance of baseline CV risk assessment and personalized surveillance regimens. Primary and secondary prevention strategies, including pharmacological interventions, are outlined. The review also delves into the cardiotoxicity associated with hematological cancer therapies, focusing on anthracyclines, Bruton kinase inhibitors, BCR-ABL tyrosine kinase inhibitors, CAR-T cell therapy, immune checkpoint inhibitors, multiple myeloma treatments, and hematopoietic stem cell transplantation. We then highlight the high risk of venous and arterial thromboembolisms in cancer patients and the challenges of anticoagulation management in cardio-oncology. Finally, the review touches on the importance of long-term follow-up and appropriate screening in cancer survivors at high risk of CV morbidity and mortality, based on their CV risk profile and the type and dose of cardiotoxic therapies they received such as anthracyclines or high radiation doses.

Unprovoked venous thromboembolism recurrence and arterial embolism revealing lung cancer: a case report

The link between venous thromboembolism (VTE) and cancer is well known. VTE could be the initial sign of an occult malignancy. There are more diagnoses of cancer after an unprovoked VTE compared to a provoked VTE, with a reported prevalence between 4.5% and 5.6% over 12 months, within the first 6 months of VTE diagnosis. There are no recommended guidelines and scores yet adopted in clinical practice, but many studies support occult cancer screening in unprovoked VTE patients. We report the case of a patient with a history of unprovoked pulmonary embolism (PE) diagnosed with bronchopulmonary neoplasm in an advanced stage one year after the thromboembolic event. When the cancer was first diagnosed, the patient's condition was already serious, being too late for the adoption of measures meant to decrease the risk of mortality and increase the duration of survival. We wanted to emphasize the importance of occult cancer screening in patients with unprovoked VTE and the fact that early cancer diagnosis reduces the risk of cancer progression, decreasing mortality and morbidity related to it.

The Conundrum of Occult Cancer Screening in Venous Thromboembolism: Lessons from the REMOTEV Registry

(1) Background and Objectives: Venous thromboembolism (VTE) is strongly associated with cancer, and may be the first event revealing occult neoplasia. Nonetheless, the reasonable extent of the etiological assessment after an unprovoked VTE event remains debated. The main objective of this study was to evaluate the incidence of occult neoplasia one year after an episode of VTE, in consecutively hospitalized patients for VTE from the REMOTEV registry. The secondary objectives were to assess the performance of the various tests used for occult cancer screening in a real-life setting and analyze the risk factors associated with the discovery of cancer and the 1-year prognosis. (2) Methods: REMOTEV is a prospective, non-interventional cohort study of patients with acute VTE. Patients included in the registry from 23 October 2013 to 28 July 2018 were analyzed after a follow-up of 12 months. Cancer detection was performed according to local practices and consisted of a limited strategy to which an abdominal ultrasound was added. In the presence of suggestive clinical manifestations, further examinations were performed on an individual basis. (3) Results: A total of 993 patients were included in the study. At 1 year, the incidence of newly diagnosed cancer was low (5.3%). Half of the detected cancers were metastatic at discovery (51%) and had a poor global prognosis (32% of mortality at 1 year). Admission pulmonary CT scans as well as (thoracic)-abdomino-pelvic CT scans (when performed) were responsible for the majority of detected cancers. Age over 65 years and the concomitant presence of an unusual site and lower-limb deep vein thrombosis were the only factors associated with occult neoplasia in this cohort. After 1-year FU, mortality was higher in cancer patients (HR 6.0 (CI 95% 3.5−10.3, p < 0.0001)), and cancer evolution was the leading cause of death in the cancer group. (4) Conclusions: In REMOTEV, VTE-revealed occult cancer prevalence was low, but similar to recent reports and associated with higher age, multiple thrombotic sites and worse prognosis.

Predictors for thromboembolism in patients with cholangiocarcinoma

BACKGROUND

Patients with cancer are at increased risk of thromboembolic events contributing significantly to cancer-related morbidity and mortality. Because cholangiocarcinoma is a rare type of cancer, the incidence of thromboembolism in this patient population is not well defined.

METHODS

Patients with cholangiocarcinoma treated at the University Cancer Center Leipzig between January 2014 and December 2018 were analyzed retrospectively regarding the incidence of arterial and venous thromboembolism.

RESULTS

A total of 133 newly and consecutively diagnosed patients were included, of whom 22% had stage IV disease. Thromboembolism was diagnosed in 39 (29.3%), with 48% of the events occurring between 60 days prior and 30 days after the initial diagnosis. Arterial thrombosis accounted for 19% and portal venous thrombosis for 33% of the events, while the rest of events occurred in the non-portal venous system. In multivariable analysis, an ONKOTEV score ≥ 2 was the only independent predictor for thromboembolism. Serum CA 19-9 was available in 87 patients (65.4%). In this subgroup, CA 19-9 above the median of 97.7 U/ml and vascular or lymphatic compression were independent predictors for thromboembolism in the first year and CA 19-9 alone remained a significant predictor over the whole observation period. An ONKOTEV score ≥ 2 and increasing age were predictors of survival.

CONCLUSIONS

A very high thromboembolic risk was observed in cholangiocarcinoma, comparable to the risk situation in pancreatic and gastric cancer. The ONKOTEV score and serum CA 19-9 are independent predictors of thromboembolic events. Prospective validation of our observations in this patient population is warranted.

Clinical presentation and long-term follow-up of 45 patients with Mondor disease: A single-center longitudinal study

Mondor disease is characterized by an acute painful thrombophlebitis occurring at specific anatomical sites. Data on its incidence, characteristics of clinical presentation, and course are unavailable to date. We studied the course of Mondor disease in patients diagnosed and followed at the University Hospital Zurich (Switzerland) between 2004 and 2020. The primary study outcomes were a diagnosis of active cancer either at the time of clinical diagnosis of Mondor disease or within 1 year, as well as 1-year all-cause death and recurrent Mondor disease. We included 45 patients and classified them into one of the three Mondor disease subgroups: thoracic (n = 26), penile (n = 12), or axillary (n = 7). The median age was 39 (Q1-Q3: 30-45) years and 44% of patients were men. Surgery was the likely cause of Mondor disease in 53.8% of patients with a thoracic form, 41.7% of those with a penile location, and all of those with an axillary location. Known active cancer was present in nine (20%) of 45 patients at baseline. One-year follow-up was available for 43 patients (median 94 months), whereas 6-month data were available for the remaining two patients. During the available follow-up, no patient had a new diagnosis of cancer. In conclusion, one in five patients with Mondor disease had known cancer at the time of diagnosis. During follow-up, the rate of new cancer diagnosis and death was negligible, providing reassurance about the good prognosis of this condition. Based on these preliminary data, extended cancer screening besides what is recommended by current guidelines for the general population might not be necessary in patients with Mondor disease.

Screening for Cancer in Patients with Acute Venous Thromboembolic Disease

Active cancer causes approximately 25% of all acute events of venous thromboembolism (VTE). While most of the cancer diagnoses are known or clinically apparent at the time of VTE, care providers and patients may be worried about the 3 to 8% risk of occult cancer occurring in the year after VTE. Several studies have compared limited to extensive cancer screening after acute VTE, especially with the addition of abdominal computed tomography (CT) or whole-body PET-CT, with the hope to shorten the time to cancer diagnosis and lead to less advanced cancer stages. These studies have not shown improved clinical outcomes with an extensive screening, and have led to current recommendations of limited screening for cancer in patients with acute VTE, including unprovoked cases. Several risk assessment models have been developed to identify patients at greatest risk of occult cancer, however, with low discriminative performances and no current clinical usefulness. Some clinical situations may empirically deserve a more thorough cancer screening, such as unprovoked upper extremity deep vein thrombosis (DVT), bilateral leg DVT, descending leg DVT, or recurrent VTE during anticoagulation.

Risk scores for occult cancer in patients with unprovoked venous thromboembolism: Results from an individual patient data meta-analysis

BACKGROUND

The Registro Informatizado de Pacientes con Enfermedad TromboEmbólica (RIETE) score and the Screening for Occult Malignancy in Patients with Idiopathic Venous Thromboembolism (SOME) risk scores aim to identify patients with acute unprovoked venous thromboembolism (VTE) at high risk of occult cancer, but their predictive performance is unclear.

METHODS

The scores were evaluated in an individual patient data meta-analysis. Studies were eligible if enrolling consecutive adults with unprovoked VTE who underwent protocol-mandated screening for cancer. The primary outcome was a cancer diagnosis between 30 days and 2 years of follow-up. The discriminatory performance was evaluated by computing the area under the receiver (ROC) curve in random-effects meta-analyses.

RESULTS

The RIETE score could be calculated in 1753 patients, of whom 63 (3.6%) were diagnosed with cancer. The pooled area under the ROC curve was 0.59 (95% confidence interval [CI], 0.52-0.66; I2  = 0%). Of the 427 patients (24%) classified as high risk, 25 (5.9%) were diagnosed with cancer compared with 38 of 1326 (2.9%) low-risk patients (hazard ratio [HR], 2.0; 95% CI, 1.3-3.4). The SOME score was calculated in 925 patients, of whom 37 (4.0%) were diagnosed with cancer. The pooled area under the ROC curve was 0.56 (95% CI, 0.46-0.65; I2  = 46%). Of the 161 patients (17%) classified as high risk (≥2 points), eight (5.0%) were diagnosed with cancer compared with 29 of 764 (3.8%) low-risk patients (HR, 1.2; 95% CI, 0.55-2.7).

CONCLUSIONS

The predictive discriminatory performance of both scores is poor. When used dichotomously, the RIETE score is able to discriminate between low- and high-risk patients. Because this is largely driven by advanced age, these results do not support the use of these scores in daily clinical practice.