Next Article in Journal
Review of Pharmacotherapy for Tinnitus
Next Article in Special Issue
Arguments for Using Direct Oral Anticoagulants in Cancer-Related Venous Thromboembolism
Previous Article in Journal
Work Fatigue in a Hospital Setting: The Experience at Cheng Hsin General Hospital
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Healthcare
Volume 9
Issue 6
10.3390/healthcare9060778
healthcare-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Risk Factors and Prediction Models for Venous Thromboembolism in Ambulatory Patients with Lung Cancer

by
Ann-Rong Yan
1,
Indira Samarawickrema
2,
Mark Naunton
1,
Gregory M. Peterson
1,3,
Desmond Yip
1,4,5,
Salvatore De Rosa
6 and
Reza Mortazavi
1,7,*
1
School of Health Sciences, Faculty of Health, University of Canberra, Canberra 2617, Australia
2
School of Nursing, Midwifery and Public Health, Faculty of Health, University of Canberra, Canberra 2617, Australia
3
College of Health and Medicine, University of Tasmania, Hobart 7005, Australia
4
Department of Medical Oncology, The Canberra Hospital, Garran 2605, Australia
5
ANU Medical School, Australian National University, Canberra 0200, Australia
6
Department of Medical and Surgical Science, Magna Graecia University, 88100 Catanzaro, Italy
7
Prehab Activity Cancer Exercise Survivorship Research Group, Faculty of Health, University of Canberra, Canberra 2617, Australia
*
Author to whom correspondence should be addressed.
Healthcare 2021, 9(6), 778; https://doi.org/10.3390/healthcare9060778
Submission received: 21 May 2021 / Revised: 13 June 2021 / Accepted: 17 June 2021 / Published: 21 June 2021
(This article belongs to the Special Issue Recent Advances in Haemostasis and Thrombosis Research in Cancer)
Versions Notes

Abstract

:
Venous thromboembolism (VTE) is a significant cause of mortality in patients with lung cancer. Despite the availability of a wide range of anticoagulants to help prevent thrombosis, thromboprophylaxis in ambulatory patients is a challenge due to its associated risk of haemorrhage. As a result, anticoagulation is only recommended in patients with a relatively high risk of VTE. Efforts have been made to develop predictive models for VTE risk assessment in cancer patients, but the availability of a reliable predictive model for ambulate patients with lung cancer is unclear. We have analysed the latest information on this topic, with a focus on the lung cancer-related risk factors for VTE, and risk prediction models developed and validated in this group of patients. The existing risk models, such as the Khorana score, the PROTECHT score and the CONKO score, have shown poor performance in external validations, failing to identify many high-risk individuals. Some of the newly developed and updated models may be promising, but their further validation is needed.

1. Background and Introduction

Cancer is a major risk factor for venous thromboembolism (VTE), which includes deep vein thrombosis and pulmonary embolism. VTE has an annual incidence of around 0.5% in cancer patients compared to around 0.1% in the general population [1]. The incidence of VTE in patients with cancer varies with cancer type, stage, and aggressiveness [2]. Among all cancer types, lung cancer has the second highest risk of VTE [3]. In cohort studies, the incidence of VTE in patients with lung cancer receiving chemotherapy was variously reported as 16.8% at three months and 14.1% at six months after the start of chemotherapy [4], and 13.9% after a median follow-up period of 12 months [5]. Having a VTE is a significant predictor of death within 2 years in patients with primary lung cancer, with hazard ratios (HRs) of 2.3 (95% CI 2.2–2.4) and 1.5 (95% CI 1.3–1.7) for non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC), respectively [6]. This matter highlights the importance of the identification of patients at risk of developing VTE so that therapeutic or preventive measures are implemented in a timely manner.
Thromboprophylaxis is suggested in hospitalised patients with lung cancer and those undergoing surgery, but the use of primary prevention of VTE in ambulatory patients with lung cancer is still debatable [7]. Choice of the anticoagulation therapy is particularly challenging in patients undergoing antineoplastic chemotherapy. On one hand, these patients are at risk of VTE over the course of therapy and beyond. On the other hand, anticoagulation is associated with a high bleeding risk, which could be life-threatening [8]. Low-molecular-weight heparins (LMWH) can reduce the risk of VTE, but current practice guidelines do not recommend their routine use, while direct-acting oral anticoagulants (DOACs) are an interesting alternative to LMWH in cancer patients [9,10,11]. Recent studies have confirmed their efficacy and safety in these patients [12]. Scientific societies such as the National Comprehensive Cancer Network (NCCN), the International Society on Thrombosis and Haemostasis (ISTH), and more recently, the American Society of Clinical Oncology (ASCO) and The International Initiative on Thrombosis and Cancer (ITAC) have supported the use of DOACs [9,10,11]. Nevertheless, the use of DOACs in this scenario should be carefully weighed against the bleeding risk, as evidence for higher risks of bleeding has emerged in studies of the general cancer population [13] and in patients starting chemotherapy [14]. As a result, it is recommended that anticoagulation is only offered to patients with a high risk of VTE, and for this we need to have robust and reliable risk assessment tools [15]. This requires a thorough understanding of the VTE risk factors and clinical prediction models to identify high-risk patients.
Clinical prediction models are epidemiological/statistical tools, which use a small number of parameters (related to the individual, or the disease, or the treatment) to estimate a likelihood in which a specific outcome (e.g., VTE) could happen. Prediction models can help clinicians better understand the individual patient’s conditions or risks, so they are able to devise a personalised treatment regimen for them [16]. In 2008, Khorana et al. established a predictive model to assess individual risk of VTE in ambulatory cancer patients receiving chemotherapy [11]. With this model, patients are assigned to one of three risk groups: low (score = 0), intermediate (score = 1–2) and high risk (score ≥ 3) [17]. Using this model, patients with lung cancer are stratified as either intermediate or high risk of developing VTE [17]. The pooled data from 45 studies including various types of cancer showed that only 23.4% (95% CI: 18.4–29.4%) of the patients who developed VTE in the first six months had been classified as being at high risk according to the Khorana score [18]. This poor performance led to the development of several modifications for the Khorana score over the years, including the Vienna Modification [19], PROTECHT [20], CONKO [21], and COPASS-CAT [22], with varying degrees of predictive ability. In this article, we have reviewed the risk factors for VTE in ambulatory patients with lung cancer, discussed some main risk assessment models for VTE in this group of patients, and reflected upon advantages and disadvantages of the models. We have also explored literature gaps and provided suggestions for further research.

2. VTE Risk Factors in Ambulatory Patients with Lung Cancer

Risk factors for VTE are grouped into three categories: patient-related, cancer-related, and biomarkers [1]. Regarding the patient-related risk factors for VTE, co-morbidities such as atrial fibrillation, chronic kidney disease [23], cardiovascular conditions, and overweight or obesity [24] increase the risk of VTE. Smoking [25] and recent hospitalisation [23,24] also raise the risk of VTE. It is believed that the Asian race has a lower risk than other races [26]. Factor V Leiden and prothrombin 20210A mutations are relatively common in Caucasians whereas they are very rare in Asians [27,28]. These mutations have been identified as additional risk factors for VTE in cancer patients [3]. Cancer patients carrying Factor V Leiden mutation or prothrombin 20210A mutation had a 12.1-fold (95% CI 1.6–88.1) and2.3-fold (95% CI 1.6–3.3) higher risk of VTE, respectively [3].
Similar to variations in the observed VTE risk among other types of cancer [3], some subtypes of lung cancer have higher risks of VTE compared to others; for example, lung adenocarcinoma had a higher risk of VTE occurrence than squamous cell carcinoma [6]. However, within the NSCLC group, the association of oncogene mutations with the risk of VTE is debatable. A systematic review by Liu et al. of 20 retrospective studies showed that anaplastic lymphoma kinase (ALK) mutation has a higher risk of VTE than epidermal growth factor receptor (EGFR) mutation [29], while another systematic review by Alexander et al. reported that EGFR mutation was the strongest risk factor for VTE in patients with lung cancer [30]. Several studies showed patients with an ALK mutation had a higher VTE risk than those without [31], with a hazard ratio of 2.47 (95% CI 1.04–5.90) for VTE in a median follow-up period of 7.5 months (95% CI 3.1–15.4 months) [32] or increasing the risk of VTE by 3 to 5 times over a median follow-up period of 22 months comparing to the general NSCLC population [33]. One possible reason is the expression of tissue factor (TF) gene was elevated in around 41.7% of ALK-positive, but only 11.5% of ALK-negative tissue in patients with lung cancer (p = 0.015) [34]. On the other hand, the prospective interventional studies using ALK inhibitor found a lower incidence of VTE than that in retrospective studies, which may be due to reduced use of chemotherapy [35].
Cancer treatment is a strong risk factor for VTE, and commonly used chemotherapy drugs can increase the risk of this clinical condition [25,26,36]. For example, gemcitabine-based chemotherapy increases the risk of VTE, with a reported odds ratio of 3.37 (95% CI 1.09–10.39) [23]. In addition, VTE was more likely to occur within the first six months of the commencement of standard chemotherapy [24]. In terms of the mechanisms involved, they are probably related to vascular endothelial damage caused by chemotherapy, especially in patients who receive the medications through central venous catheterisation over a long period [25,26,36].
There are a limited number of studies which have investigated the impacts of novel anti-cancer treatments on the haemostatic system. For instance, in patients treated with immune checkpoint inhibitors (ICIs), the cumulative incidence of VTE over a median follow-up of 8.5 months was 12.9% (95% CI 8.2–18.5%) [37]. Sato et al. found that in patients with NSCLC, anti-programmed cell death 1 (PD-1)/programmed cell death ligand 1 (PD-L1) monoclonal antibodies might impose procoagulant effects on the haemostatic system [38]. It has been reported that the activation of T-cells in vitro induced the production of TF in high-PD-L1-expressed monocytes [38]. Additionally, in cancer patients receiving ICIs, some biomarkers such as vascular cell adhesion molecule 1 (sVCAM-1) and interleukin 8 (IL-8) were associated with the occurrence of VTE [39]. The administration of ICIs and associated biomarkers have not been investigated in a VTE risk model yet, and they may be considered as candidates for a new predictive model, although the feasibility of the measurements could be a problem in clinical settings.
Some tissue or blood biomarkers have been found to have significant associations with hypercoagulable conditions in patients with cancer; therefore, they could be considered as candidate predictors in VTE risk assessment. For instance, platelets and von-Willebrand factor (vWF) not only play vital roles in preventing blood loss following a blood vessel injury by facilitating coagulation [40], but also contribute to thrombosis as well as cancer metastasis [15,41]. It has been reported that high platelet counts and/or increased platelet activity can increase the risk of developing VTE in patients with cancer [17,42]. Additionally, elevated levels of plasma vWF, particularly the high molecular weight multimers, can induce the formation of vWF-mediated platelet thrombi by facilitating platelet aggregation under high shear rate conditions [43]. In different studies, high plasma vWF was associated with the occurrence of VTE in cancer patients [44,45]. Meanwhile, increased levels of thrombin-antithrombin III complex (TAT) (mean 4.7 [95% CI 4–6] vs. 2.5 [95% CI 2–3] μg/L, p < 0.0001) and elevated prothrombin fragment 1+2 (F1+2), as a by-product of thrombin generation (mean 267 [95% CI 244–289] vs. 183 [95% CI 171–196] pmol/L, p < 0.00001) have been reported in primary lung cancer patients with VTE [46]. In addition, increased levels of D-dimer and plasmin-α2-antiplasmin inhibitor complex (PIC) [47], generated in fibrinolysis, and serum ionised calcium have shown associations with the occurrence of VTE in patients with lung cancer [48]. Other biomarkers or laboratory parameters, such as fibrinogen, activated partial thromboplastin time (aPTT) and albumin, showed associations with VTE [49,50], but are lacking external validation of their predictive roles. Most biomarkers, as predictors of VTE, have been typically used at the diagnosis of cancer or before the beginning of chemotherapy, whereas D-dimer has also been investigated in longitudinal studies following the diagnosis [30,51]. Some biomarkers have been studied as predictors of VTE and have been used for prediction model development. For example, platelet count, white cell count and haemoglobin have been incorporated into the Khorana score [17], although their effects on the occurrence of VTE in patients with lung cancer are controversial [30].
Inflammatory mediators (particularly those cytokines which are associated with vascular inflammation) are another group of biomarkers which could be used as predictors for new VTE prediction model development. Cancer cells produce and secrete inflammatory cytokines such as interleukin 1 (IL1), tumour necrosis factor alpha (TNF-α) and vascular endothelial growth factor (VEGF), which mediate interactions between cancer cells and host cells, including endothelial cells, platelets, monocytes and neutrophils [52]. Furthermore, the blood levels of intracellular granular proteins or surface-expressed proteins released from the activated host cells [15], including procoagulant platelets [53], may be used as predictors of a hypercoagulable status in patients with solid tumours [52]. For instance, activated monocytes in cancer patients produced more TF than resting monocytes [15]. Additionally, neutrophil elastase (NE) released by activated neutrophils showed correlations with aPTT, D-dimer, TAT, PIC and fibrinogen levels in lung cancer patients, with stronger correlations seen in NSCLC [54].

3. Risk Prediction Models for VTE in Patients with Cancer

Risk prediction models, also referred to as risk assessment tools or clinical prediction rules, are prognostic models, which use predictors to estimate the probability for individuals to develop a condition in the future [55]. This review covers the particulars of some available VTE risk prediction models which have been developed and/or validated in ambulatory patients with lung cancer. A summary of the main features can be found in Table 1.

3.1. The Khorana Score

The Khorana Score is the first and most frequently investigated VTE risk prediction model in patients with cancer. It uses the following five predictors: cancer site, platelet count, leucocyte count, haemoglobin, and body mass index (BMI). In assessing an individual’s VTE risk, 2 points are assigned for very high-risk cancer types (e.g., stomach and pancreas), 1 point for high-risk cancer (e.g., lung, lymphoma, gynaecological, bladder, testicular), 1 point for baseline platelet count ≥ 350 × 109/L, 1 point for baseline leukocyte count > 11 × 109/L, 1 point for baseline haemoglobin level < 100 g/L or the use of erythropoietin, and 1 point for BMI ≥ 35 kg/m2 (Table 1) [17]. The main advantage of the Khorana score is that all the predictors used are among the routinely measured clinical or laboratory parameters. Additionally, this score has a high specificity. However, the score has a poor sensitivity. Multiple studies showed a low sensitivity of 10–25%, a high specificity of 76–100%, and a poor discrimination (C-index) of about 0.50 with the high-risk threshold of 3 points in ambulatory patients with lung cancer [23,25,51,56,57].
On the other hand, a subgroup analysis between lung cancer and non-lung cancer patients (including colorectal, pancreatic, stomach, ovarian, breast, brain and bladder cancers) showed a different discriminatory performance for the Khorana score [58]. The dichotomous Khorana score did not identify higher risk of VTE compared with low–moderate risk in ambulatory patients with lung cancer in a meta-analysis (OR = 1.1, 95% CI 0.72–1.7), whereas it was a strong predictor for other types of cancers (OR = 3.2, 95% CI 1.8–5.6, p for interaction = 0.002) [58]. This finding emphasises the need for a better VTE model for ambulatory patients with lung cancer.
Recently, a change in the high-risk threshold from 3 to 2 of the Khorana score was introduced and used in randomised controlled trials of direct inhibitors of Factor Xa in ambulatory patients with cancer at a high risk of VTE [9,61]. In the stratified group with a Khorana score ≥ 2, apixaban lowered incidence of VTE with increasing major bleeding risk [9], while rivaroxaban did not lower incidence of VTE significantly [61]. This cut-off value of the Khorana score has been adopted in the latest ASCO guidelines for thromboprophylaxis in cancer (recommended for patients with a Khorana score of 2 or higher) [10]. However, so far, the latter threshold is yet to be validated in a homogeneous lung cancer population.

3.2. Modifications of the Khorana Score

Since its introduction in 2008, the Khorana model has been modified multiple times by the addition and/or replacement of biomarkers. For example, by adding D-dimer and soluble P-selection to it, the Vienna Modification (CATS score) was introduced in 2010 [19]. Later, by the addition of treatment-related factors, such as gemcitabine or platinum-based chemotherapy, the PROTECHT score was developed [20]. By the addition of World Health Organisation (WHO) performance status and omission of BMI, the CONKO score was developed in patients with advanced pancreatic cancer [21]. More precisely, the PROTECHT score adds 1 point for each of gemcitabine or platinum chemotherapies, and the CONKO score removes BMI but adds the Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≥ 2 for 1 point, while the cut-off value is still 3 points in both scores (Table 1).
The PROTECHT score and the CONKO score have been validated in ambulatory patients with lung cancer [23,51]. Compared to the Khorana score, more patients (22–48%) were categorised into the high-risk group by the CONKO score and even more so by the PROTECHT score (52–64%), but there was no significant difference in the incidence of VTE between high- and low-risk groups stratified by either of these models. A C-index around 0.50 consistently showed a poor discrimination for both models [23,51].
The VTE risk prediction model developed by Ferroni et al. is another extension of the Khorana model with an additional predictor, which is high-sensitive D-dimer (Table 1) [59]. According to the Khorana score, patients with lung cancer are given a score of 1 point or more, having either intermediate or high risks for VTE [17]. This model is not applicable to all ambulatory patients with lung cancer, but only to those with an intermediate risk of VTE stratified by the Khorana score. A high-sensitive (HS) D-dimer level is used to select patients with a higher risk of VTE from those within the intermediate risk group stratified by a Khorana score of 1–2 points [59]. This study compared the relatively high-risk subgroup to the relatively low risk subgroup within the Khorana intermediate risk group. With a threshold of 1500 ng/mL for HS D-dimer at baseline, both sensitivity and specificity were moderate at 81.3% and 68.5%, respectively, with a moderate discriminating capacity of 0.704 [59].
There are also some other types of VTE risk models, such as a simple model with only two factors including distant metastases and platinum therapy [62], and a model using cancer site and continuous D-dimer concentration rather than a cut-off value, called the CATS-MICA model [63]. Neither of these risk prediction models has been externally validated in patients with lung cancer.

3.3. The COMPASS-CAT Score

The COMPASS-CAT score is a more complicated model with both cancer-related and patient-related factors, which are anthracycline treatment (6 points), time since cancer diagnosis ≤ 6 months (4 points), central venous catheter use (3 points), advanced stage of cancer (2 points), cardiovascular risk factors present (5 points), recent hospitalisation for acute medical illness (2 points), a history of VTE (1 point), and platelet count ≥350 × 109/L (2 points) (Table 1) [22]. This score has been validated in ambulatory patients with lung cancer, but two different cut-off values for risk stratification have been used in different studies [23,24,60]. The percentage of high-risk patients was 90% and 71% with a cut-off value of 7 and 11, respectively. There was an obvious difference in the incidence of VTE between high- and low-risk groups, which was 10.8% vs. 6.6% (p value not reported) and 23.8% vs. 0% (OR 9.65 [95% CI 1.24–75.24], p = 0.031) in the models with the cut-off values of 7 and 11, respectively [23,60].
Compared to the Khorana score and its variations, the COMPASS-CAT score had a higher sensitivity, which was 83% and 100% with a cut-off value of 7 and 11, respectively. However, the specificity decreased to 35–51%, which means about one-half to two-thirds of patients who did not develop VTE had been classified as high risk [23,24]. The COMPASS-CAT score with the cut-off value of 11 showed a high discrimination with a C-index of 0.89 [23]; however, with the change in the cut-off value from 7 to 11 this model needs further validation.
The varying performance between these models may reflect the composite of the different categories of predictors and their weights. The COMPASS-CAT score only comprises one biomarker, that is platelet count [22], while the Khorana score, the PROTECHT score and the CONKO score also include leukocyte count and haemoglobin level; meanwhile, the COMPASS-CAT score also places more weight on cancer-related and patient-related risk factors for VTE than the Khorana score and its variations do [17,20,21].

3.4. Models Developed in Ambulatory Patients with Lung Cancer

A dynamic model developed in NSCLC patients by Alexander et al. involves only two biomarkers: baseline fibrinogen level and dynamic D-dimer levels at baseline and after one month [51]. This risk model is described as three conditions: baseline fibrinogen ≥ 4.0 g/L and baseline D-dimer ≥ 0.5 mg/L, baseline D-dimer ≥ 1.5 mg/L, and month-1 D-dimer ≥ 1.5 mg/L; each condition scores 1 point and a high risk is assigned if the patient is receiving chemotherapy and the score ≥1 [51]. Seventy-two percent of a cohort of outpatients with lung cancer had a high risk of thromboembolism, and the incidence of thromboembolism at 6 months was 26.5% vs. 0% in the high- and low-risk groups, respectively [51]. This model had a high sensitivity (100%, 95% CI 79–100%) but a low specificity (34%, 95% CI 23–47%), while the discriminating capacity was moderate (C-index 0.67, 95% CI 0.61–0.73) [50]. This model may not be applicable in patients with lung cancer who do not receive chemotherapy.
The ROADMAP-CAT model, which was developed in outpatients with lung adenocarcinoma, involves laboratory parameters only, which are procoagulant phospholipid-dependent clotting time (Procoag-PPL) and mean rate index (MRI) of thrombin generation [24]. This model is a binary scoring system: patients with Procoag-PPL < 44 s and MRI < 125 nM/min are identified with a high risk of VTE, while those with Procoag-PPL > 44 s or MRI > 125 nM/min are considered low risk [24]. The incidence of VTE in a 6-month follow-up period was 12.2% and 3.4% in high- and low-risk groups, respectively. The sensitivity was relatively high (88%), but the specificity was relatively low (52%), with a C-index of 0.77 showing moderate discriminatory performance [24].

4. Risk of Bias in VTE Risk Model Development and Validation Studies

We used the Prediction model Risk Of Bias ASessment Tool (PROBAST) [64] to identify potential biases in the development or validation of current VTE risk models. The first issue is the small sample size of many studies. As a rule of thumb, for risk model development studies, events per variable (EPV) should not be less than 10, while for risk model validation studies, there should be more than 100 participants with the occurrence of the outcome, in this case, VTE [65]. Secondly, dichotomisation of continuous predictors, such as white cell count, platelet count, BMI, D-dimer and fibrinogen, occurs in almost all models, which may lead to loss of linear information [65]. In addition, univariable analysis is a popular approach for predictor selection; however, this may miss some important predictors that are confounded by other predictors [65]. Furthermore, internal validation is necessary for directing an adjustment to build a robust risk prediction model, but in the VTE risk model development studies in ambulatory patients with lung cancer, internal validation was overlooked [65].
Model overfitting in risk model development could arise from issues such as small EPV, dichotomisation of continuous predictors, selecting predictors solely based on univariable analysis, or lack of internal validation with bootstrapping or cross-validation in the development studies. A lack of calibration is the next problem in both risk model development and validation. Calibration indicates the accuracy of a risk model by showing agreement between the expected number of events based on the risk model and the observed number of events [66]. Calibration is indispensable in the external validation of prognostic risk models, and calibration plots are even suggested at more than one time point for those models with competing risks [67]. Last but not least, there is use of a derived clinical score in risk models that does not reflect the actual weight of a predictor from the multivariable analysis in VTE risk model development [51]. The logistic regression equation is the actual expression of the risk model developed from the data [23].
Validation of VTE risk models should include patients receiving all currently approved medications, including anticoagulation therapies in this context. Ideally, a VTE risk model to be used in cancer patients should also provide some hint as to the most suitable medication to be used [68]. To address the growing clinical complexity, use of novel technologies, such as information-technology-powered decision support systems and Artificial Intelligence (AI) algorithms might be helpful both in the development and the validation phases. In this regard, preliminary studies have recently provided promising results [68,69].

5. Conclusions

Anticoagulant thromboprophylaxis in ambulatory patients with lung cancer could be a life-saving treatment by preventing VTE. However, since this treatment increases the risk of bleeding, it should be administered only in patients who are at a high risk of developing thromboembolic events. In this literature review, we have discussed some of the available risk prediction models for VTE which could be used in ambulatory patients with lung cancer. The original Khorana score and its modifications may not be sensitive enough to identify high-risk ambulatory patients with lung cancer. With the cut-off value changing from 3 to 2 points, the Khorana score may be effective, but it needs to be validated in ambulatory patients with lung cancer. The COMPASS-CAT score and some other developed risk models may be promising, but further validation is needed. Correct statistical methods and processes, including sufficient sample size, pre-specified predictor candidates, calibration plots and internal validation, should be considered and used in both risk model development and validation studies to ensure a low risk of bias.

Author Contributions

Writing—original draft preparation, A.-R.Y.; writing—review and editing, I.S., M.N., G.M.P., D.Y., S.D.R. and R.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Ethical review and approval were waived for this study, because this study is a systematic review.

Informed Consent Statement

Patient consent was waived because this is a systematic review of published original studies, in each of which patient consent has been obtained.

Data Availability Statement

No data have been synthesised and reported in this study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Fernandes, C.; Morinaga, L.; Alves, J.; Castro, M.; Calderaro, D.; Jardim, C.; Souza, R. Cancer-associated thrombosis: The when, how and why. Eur. Respir. Rev. 2019, 28, 180119. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  2. Timp, J.; Braekkan, S.K.; Versteeg, H.H.; Cannegieter, S.C. Epidemiology of cancer-associated venous thrombosis. Blood 2013, 122. [Google Scholar] [CrossRef] [Green Version]
  3. Blom, J.W.; Doggen, C.J.; Osanto, S.; Rosendaal, F.R. Malignancies, prothrombotic mutations, and the risk of venous thrombosis. JAMA 2005, 293, 715–722. [Google Scholar] [CrossRef]
  4. Kenmotsu, H.; Notsu, A.; Mori, K.; Omori, S.; Tsushima, T.; Satake, Y.; Miki, Y.; Abe, M.; Ogiku, M.; Nakamura, T.; et al. Cumulative incidence of venous thromboembolism in patients with advanced cancer in prospective observational study. Cancer Med. 2021, 10, 895–904. [Google Scholar] [CrossRef] [PubMed]
  5. Connolly, G.C.; Dalal, M.; Lin, J.; Khorana, A.A. Incidence and predictors of venous thromboembolism (VTE) among ambulatory patients with lung cancer. Lung Cancer 2012, 78. [Google Scholar] [CrossRef] [PubMed]
  6. Chew, H.K.; Davies, A.M.; Wun, T.; Harvey, D.; Zhou, H.; White, R.H. The incidence of venous thromboembolism among patients with primary lung cancer. J. Thromb. Haemost. 2008, 6. [Google Scholar] [CrossRef]
  7. Streiff, M.; Abutalib, S.A.; Farge, D.; Murphy, M.; Connors, J.M.; Piazza, G. Update on Guidelines for the Management of Cancer-Associated Thrombosis. Oncologist 2021, 26, e24–e40. [Google Scholar] [CrossRef]
  8. Fuentes, H.; Oramas, D.M.; Paz, L.H.; Casanegra, A.I.; Mansfield, A.S.; Tafur, A.J. Meta-analysis on anticoagulation and prevention of thrombosis and mortality among patients with lung cancer. Thromb. Res. 2017, 154, 28–34. [Google Scholar] [CrossRef]
  9. Carrier, M.; Abou-Nassar, K.; Mallick, R.; Tagalakis, V.; Shivakumar, S.; Schattner, A.; Kuruvilla, P.; Hill, D.; Spadafora, S.; Marquis, K.; et al. Apixaban to Prevent Venous Thromboembolism in Patients with Cancer. N. Engl. J. Med. 2019, 380, 711–719. [Google Scholar] [CrossRef]
  10. Key, N.S.; Khorana, A.A.; Kuderer, N.M.; Bohlke, K.; Lee, A.Y.Y.; Arcelus, J.I.; Wong, S.L.; Balaban, E.P.; Flowers, C.R.; Francis, C.W.; et al. Venous Thromboembolism Prophylaxis and Treatment in Patients With Cancer: ASCO Clinical Practice Guideline Update. J. Clin. Oncol. 2020, 38, 496–520. [Google Scholar] [CrossRef]
  11. Wang, T.F.; Zwicker, J.I.; Ay, C.; Pabinger, I.; Falanga, A.; Antic, D.; Noble, S.; Khorana, A.A.; Carrier, M.; Meyer, G. The use of direct oral anticoagulants for primary thromboprophylaxis in ambulatory cancer patients: Guidance from the SSC of the ISTH. J. Thromb. Haemost. 2019, 17, 1772–1778. [Google Scholar] [CrossRef] [Green Version]
  12. NCCN. Guidelines for Venous Thromboembolic Disease; Version 1; NCCN: Plymouth Meeting, PA, USA, 2019. [Google Scholar]
  13. Sabatino, J.; De Rosa, S.; Polimeni, A.; Sorrentino, S.; Indolfi, C. Direct Oral Anticoagulants in Patients With Active Cancer: A Systematic Review and Meta-Analysis. JACC CardioOncol. 2020, 2, 428–440. [Google Scholar] [CrossRef]
  14. Li, A.; Kuderer, N.M.; Garcia, D.A.; Khorana, A.A.; Wells, P.S.; Carrier, M.; Lyman, G.H. Direct oral anticoagulant for the prevention of thrombosis in ambulatory patients with cancer: A systematic review and meta-analysis. J. Thromb. Haemost. 2019, 17, 2141–2151. [Google Scholar] [CrossRef]
  15. Roselli, M.; Riondino, S.; Mariotti, S.; La Farina, F.; Ferroni, P.; Guadagni, F. Clinical models and biochemical predictors of VTE in lung cancer. Cancer Metastasis Rev. 2014, 33, 771–789. [Google Scholar] [CrossRef]
  16. Steyerberg, E.W. Clinical Prediction Models: A Practical Approach to Development, Validation and Updating, 2nd ed.; Springer: Cham, Switzerland, 2019; p. 3. [Google Scholar]
  17. Khorana, A.; Kuderer, N.M.; Culakova, E.; Lyman, G.H.; Francis, C.W. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood 2008, 111. [Google Scholar] [CrossRef] [Green Version]
  18. Mulder, F.I.; Candeloro, M.; Kamphuisen, P.W.; Di Nisio, M.; Bossuyt, P.M.; Guman, N.; Smit, K.; Buller, H.R.; van Es, N. The Khorana score for prediction of venous thromboembolism in cancer patients: A systematic review and meta-analysis. Haematologica 2019, 104. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  19. Ay, C.; Dunkler, D.; Marosi, C.; Chiriac, A.; Vormittag, R.; Simanek, R.; Quehenberger, P.; Zielinski, C.; Pabinger, I. Prediction of venous thromboembolism in cancer patients. Blood 2010, 116. [Google Scholar] [CrossRef] [PubMed]
  20. Verso, M.; Agnelli, G.; Barni, S.; Gasparini, G.; Labianca, R. A modified Khorana risk assessment score for venous thromboembolism in cancer patients receiving chemotherapy: The Protecht score. Intern. Emerg. Med. 2012, 7. [Google Scholar] [CrossRef] [PubMed]
  21. Pelzer, U.; Sinn, M.; Stieler, J.; Riess, H. Primary pharmacological prevention of thromboembolic events in ambulatory patients with advanced pancreatic cancer treated with chemotherapy. Dtsch. Med. Wochenschr. 2013, 138. [Google Scholar] [CrossRef]
  22. Gerotziafas, G.T.; Taher, A.; Abdel-Razeq, H.; AboElnazar, E.; Spyropoulos, A.C.; El Shemmari, S.; Larsen, A.K.; Elalamy, I. A Predictive Score for Thrombosis Associated with Breast, Colorectal, Lung, or Ovarian Cancer: The Prospective COMPASS-Cancer-Associated Thrombosis Study. Oncologist 2017, 22, 1222–1231. [Google Scholar] [CrossRef] [Green Version]
  23. Rupa-Matysek, J.; Lembicz, M.; Rogowska, E.K.; Gil, L.; Komarnicki, M.; Batura-Gabryel, H. Evaluation of risk factors and assessment models for predicting venous thromboembolism in lung cancer patients. Med. Oncol. 2018, 35, 63. [Google Scholar] [CrossRef] [Green Version]
  24. Syrigos, K.; Grapsa, D.; Sangare, R.; Evmorfiadis, I.; Larsen, A.K.; Van Dreden, P.; Boura, P.; Charpidou, A.; Kotteas, E.; Sergentanis, T.N.; et al. Prospective Assessment of Clinical Risk Factors and Biomarkers of Hypercoagulability for the Identification of Patients with Lung Adenocarcinoma at Risk for Cancer-Associated Thrombosis: The Observational ROADMAP-CAT Study. Oncologist 2018, 23, 1372–1381. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  25. Kuderer, N.M.; Poniewierski, M.S.; Culakova, E.; Lyman, G.H.; Khorana, A.A.; Pabinger, I.; Agnelli, G.; Liebman, H.A.; Vicaut, E.; Meyer, G.; et al. Predictors of Venous Thromboembolism and Early Mortality in Lung Cancer: Results from a Global Prospective Study (CANTARISK). Oncologist 2018, 23, 247–255. [Google Scholar] [CrossRef] [Green Version]
  26. Li, Z.; Zhang, G.; Zhang, M.; Mei, J.; Weng, H.; Peng, Z. Development and Validation of a Risk Score for Prediction of Venous Thromboembolism in Patients with Lung Cancer. Clin. Appl. Thromb. Hemost. 2020, 26, 1076029620910793. [Google Scholar] [CrossRef] [Green Version]
  27. Jadaon, M.M. Epidemiology of Prothrombin G20210A Mutation in the Mediterranean Region. Mediterr. J. Hematol. Infect. Dis. 2011, 3, e2011054. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  28. De Stefano, V.; Chiusolo, P.; Paciaroni, K.; Leone, G. Epidemiology of factor V Leiden: Clinical implications. Semin. Thromb. Hemost. 1998, 24, 367–379. [Google Scholar] [CrossRef] [PubMed]
  29. Liu, Y.; Wang, W.; Wu, F.; Gao, G.; Xu, J.; Li, X.; Zhao, C.; Yang, S.; Mao, S.; Pan, Y.; et al. High discrepancy in thrombotic events in non-small cell lung cancer patients with different genomic alterations. Transl. Lung Cancer Res. 2021, 10, 1512–1524. [Google Scholar] [CrossRef]
  30. Alexander, M.; Burbury, K. A systematic review of biomarkers for the prediction of thromboembolism in lung cancer—Results, practical issues and proposed strategies for future risk prediction models. Thromb. Res. 2016, 148, 63–69. [Google Scholar] [CrossRef]
  31. Al-Samkari, H.; Leiva, O.; Dagogo-Jack, I.; Shaw, A.; Lennerz, J.; Iafrate, A.J.; Bendapudi, P.K.; Connors, J.M. Impact of ALK Rearrangement on Venous and Arterial Thrombotic Risk in NSCLC. J. Thorac. Oncol. 2020, 15, 1497–1506. [Google Scholar] [CrossRef]
  32. Dou, F.; Zhang, Y.; Yi, J.; Zhu, M.; Zhang, S.; Zhang, D.; Zhang, Y. Association of ALK rearrangement and risk of venous thromboembolism in patients with non-small cell lung cancer: A prospective cohort study. Thromb. Res. 2020, 186, 36–41. [Google Scholar] [CrossRef]
  33. Zer, A.; Moskovitz, M.; Hwang, D.M.; Hershko-Klement, A.; Fridel, L.; Korpanty, G.J.; Dudnik, E.; Peled, N.; Shochat, T.; Leighl, N.B.; et al. ALK-Rearranged Non-Small-Cell Lung Cancer Is Associated With a High Rate of Venous Thromboembolism. Clin. Lung Cancer 2017, 18, 156–161. [Google Scholar] [CrossRef]
  34. Yang, S.; Yang, L.; Wu, Y.; Zhang, C.; Wang, S.; Ma, N.; Wang, L.; Wang, Q. Anaplastic lymphoma kinase rearrangement may increase the incidence of venous thromboembolism by increasing tissue factor expression in advanced lung adenocarcinoma. Ann. Transl. Med. 2020, 8, 1307. [Google Scholar] [CrossRef] [PubMed]
  35. Alexander, M.; Solomon, B.; Burbury, K. Thromboembolism in Anaplastic Lymphoma Kinase-Rearranged Non-Small Cell Lung Cancer. Clin. Lung Cancer 2018, 19, e71–e72. [Google Scholar] [CrossRef]
  36. Hohl Moinat, C.; Périard, D.; Grueber, A.; Hayoz, D.; Magnin, J.-L.; André, P.; Kung, M.; Betticher, D.C. Predictors of venous thromboembolic events associated with central venous port insertion in cancer patients. J. Oncol. 2014, 2014, 743181. [Google Scholar] [CrossRef] [PubMed]
  37. Moik, F.; Chan, W.E.; Wiedemann, S.; Hoeller, C.; Tuchmann, F.; Aretin, M.B.; Fuereder, T.; Zöchbauer-Müller, S.; Preusser, M.; Pabinger, I.; et al. Incidence, risk factors, and outcomes of venous and arterial thromboembolism in immune checkpoint inhibitor therapy. Blood 2021, 137, 1669–1678. [Google Scholar] [CrossRef]
  38. Sato, R.; Imamura, K.; Sakata, S.; Ikeda, T.; Horio, Y.; Iyama, S.; Akaike, K.; Hamada, S.; Jodai, T.; Nakashima, K.; et al. Disorder of Coagulation-Fibrinolysis System: An Emerging Toxicity of Anti-PD-1/PD-L1 Monoclonal Antibodies. J. Clin. Med. 2019, 8, 762. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  39. Swaidani, S.; Roopkumar, J.; Jun-Shim, Y.; Charles, C.; Paul, S.; Kundu, S.; Funchain, P.; Rayman, P.; Pavicic, P.G.; Diaz-Montero, C.; et al. Biomarker Assessment of Venous Thromboembolism in Cancer Patients Receiving Checkpoint Blockade. Blood 2019, 134 (Suppl. 1). [Google Scholar] [CrossRef]
  40. Hoffbrand, A.; Moss, P.A.H. Thrombosis 1: Pathogenesis and Diagnosis. In Hoffbrand’s Essential Haematology, 7th ed.; John Wiley & Sons: Chichester, UK, 2016; Charpter 27. [Google Scholar]
  41. Koupenova, M.; Clancy, L.; Corkrey, H.A.; Freedman, J.E. Circulating Platelets as Mediators of Immunity, Inflammation, and Thrombosis. Circ. Res. 2018, 122, 337–351. [Google Scholar] [CrossRef]
  42. Ma, R.; Bi, Y.; Kou, J.; Zhou, J.; Shi, J. Enhanced procoagulant activity of platelets after chemotherapy in non-small cell lung cancer. Cancer Biol. Ther. 2017, 18, 627–634. [Google Scholar] [CrossRef] [Green Version]
  43. Rangarajan, S. Von Willebrand factor—Two sides and the edge of a coin. Haemophilia 2011, 17, 61–64. [Google Scholar] [CrossRef]
  44. Pépin, M.; Kleinjan, A.; Hajage, D.; Büller, H.R.; Di Nisio, M.; Kamphuisen, P.W.; Salomon, L.; Veyradier, A.; Stepanian, A.; Mahé, I. ADAMTS-13 and von Willebrand factor predict venous thromboembolism in patients with cancer. J. Thromb. Haemost. 2016, 14, 306–315. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  45. Obermeier, H.L.; Riedl, J.; Ay, C.; Koder, S.; Quehenberger, P.; Bartsch, R.; Kaider, A.; Zielinski, C.C.; Pabinger, I. The role of ADAMTS-13 and von Willebrand factor in cancer patients: Results from the Vienna Cancer and Thrombosis Study. Res. Pract. Thromb. Haemost. 2019, 3, 503–514. [Google Scholar] [CrossRef] [Green Version]
  46. Lundbech, M.; Krag, A.E.; Christensen, T.D.; Hvas, A.M. Thrombin generation, thrombin-antithrombin complex, and prothrombin fragment F1+2 as biomarkers for hypercoagulability in cancer patients. Thromb. Res. 2020, 186, 80–85. [Google Scholar] [CrossRef]
  47. Gabazza, E.; Taguchi, O.; Yamakami, T.; Machishi, M.; Ibata, H.; Suzuki, S. Evaluating prethrombotic state in lung cancer using molecular markers. Chest 1993, 103, 196–200. [Google Scholar] [CrossRef]
  48. Xiong, W.; Zhao, Y.; Xiong, Y.; Xu, M.; Pudasaini, B.; Du, H.; Guo, X. Coagulation factor IV is an indicator of symptomatic pulmonary embolism in patients with primary lung cancer. Clin. Respir. J. 2020, 14, 124–131. [Google Scholar] [CrossRef] [PubMed]
  49. Iiu, Y.; Gu, Y.; Yi, F.; Cao, B. Retrospective Analysis of Risk Factors for Venous Thromboembolism in 283 Patients with Lung Cancer during Systemic Therapy. Chin. J. Lung Cancer 2019, 22, 419–426. [Google Scholar] [CrossRef]
  50. Kadlec, B.; Skrickova, J.; Merta, Z.; Dusek, L.; Jarkovsky, J. The incidence and predictors of thromboembolic events in patients with lung cancer. Sci. World J. 2014, 2014, 125706. [Google Scholar] [CrossRef]
  51. Alexander, M.; Ball, D.; Solomon, B.; MacManus, M.; Manser, R.; Riedel, B.; Westerman, D.; Evans, S.M.; Wolfe, R.; Burbury, K. Dynamic Thromboembolic Risk Modelling to Target Appropriate Preventative Strategies for Patients with Non-Small Cell Lung Cancer. Cancers 2019, 11, 50. [Google Scholar] [CrossRef] [Green Version]
  52. Falanga, A.; Russo, L.; Milesi, V.; Vignoli, A. Mechanisms and risk factors of thrombosis in cancer. Crit. Rev. Oncol. Hematol. 2017, 118, 79–83. [Google Scholar] [CrossRef] [PubMed]
  53. Faria, A.; Andrade, S.S.; Peppelenbosch, M.P.; Ferreira-Halder, C.V.; Fuhler, G.M. Platelets in aging and cancer—“Double-edged sword”. Cancer Metastasis Rev. 2020, 39, 1205–1221. [Google Scholar] [CrossRef]
  54. Gabazza, E.; Taguchi, O.; Yamakami, T.; Machishi, M.; Ibata, H.; Suzuki, S. Correlation between increased granulocyte elastase release and activation of blood coagulation in patients with lung cancer. Cancer 1993, 72, 2134–2140. [Google Scholar] [CrossRef]
  55. Cowley, L.E.; Farewell, D.M.; Maguire, S.; Kemp, A.M. Methodological standards for the development and evaluation of clinical prediction rules: A review of the literature. Diagn. Progn. Res. 2019, 3, 16. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  56. Vathiotis, I.; Dimakakos, E.P.; Boura, P.; Ntineri, A.; Charpidou, A.; Gerotziafas, G.; Syrigos, K. Khorana Score: Νew Predictor of Early Mortality in Patients with Lung Adenocarcinoma. Clin. Appl. Thromb. Hemost. 2018, 24, 1347–1351. [Google Scholar] [CrossRef]
  57. Mansfield, A.S.; Tafur, A.J.; Wang, C.E.; Kourelis, T.V.; Wysokinska, E.M.; Yang, P. Predictors of active cancer thromboembolic outcomes: Validation of the Khorana score among patients with lung cancer. J. Thromb. Haemost. 2016, 14, 1773–1778. [Google Scholar] [CrossRef] [Green Version]
  58. Van Es, N.; Ventresca, M.; Di Nisio, M.; Zhou, Q.; Noble, S.; Crowther, M.; Briel, M.; Garcia, D.; Lyman, G.H.; Macbeth, F.; et al. The Khorana score for prediction of venous thromboembolism in cancer patients: An individual patient data meta-analysis. J. Thromb. Haemost. 2020, 18, 1940–1951. [Google Scholar] [CrossRef]
  59. Ferroni, P.; Martini, F.; Portarena, I.; Massimiani, G.; Riondino, S.; La Farina, F.; Mariotti, S.; Guadagni, F.; Roselli, M. Novel high-sensitive D-dimer determination predicts chemotherapy-associated venous thromboembolism in intermediate risk lung cancer patients. Clin. Lung Cancer 2012, 13, 482–487. [Google Scholar] [CrossRef]
  60. Spyropoulos, A.; Eldredge, J.B.; Anand, L.N.; Zhang, M.Q.; Michael, N.; Soheila, R.; David, J. External Validation of a Venous Thromboembolic Risk Score for Cancer Outpatients with Solid Tumors: The COMPASS-CAT Venous Thromboembolism Risk Assessment Model. Oncologist 2020, 25, e1083–e1090. [Google Scholar] [CrossRef] [Green Version]
  61. Khorana, A.A.; Soff, G.A.; Kakkar, A.K.; Vadhan-Raj, S.; Riess, H.; Wun, T.; Streiff, M.B.; Garcia, D.A.; Liebman, H.A.; Belani, C.P.; et al. Rivaroxaban for Thromboprophylaxis in High-Risk Ambulatory Patients with Cancer. N. Engl. J. Med. 2019, 380, 720–728. [Google Scholar] [CrossRef]
  62. Rojas-Hernandez, C.; Tang, V.K.; Sanchez-Petitto, G.; Qiao, W.; Richardson, M.; Escalante, C. Development of a clinical prediction tool for cancer-associated venous thromboembolism (CAT): The MD Anderson Cancer Center CAT model. Support. Care Cancer 2020, 28, 3755–3761. [Google Scholar] [CrossRef]
  63. Pabinger, I.; van Es, N.; Heinze, G.; Posch, F.; Riedl, J.; Reitter, E.M.; Di Nisio, M.; Cesarman-Maus, G.; Kraaijpoel, N.; Zielinski, C.C.; et al. A clinical prediction model for cancer-associated venous thromboembolism: A development and validation study in two independent prospective cohorts. Lancet Haematol. 2018, 5, e289–e298. [Google Scholar] [CrossRef]
  64. Wolff, R.F.; Moons, K.G.M.; Riley, R.D.; Whiting, P.F.; Westwood, M.; Collins, G.S.; Reitsma, J.B.; Kleijnen, J.; Mallett, S.; PROBAST Group. PROBAST: A Tool to Assess the Risk of Bias and Applicability of Prediction Model Studies. Ann. Intern. Med. 2019, 170, 51–58. [Google Scholar] [CrossRef] [Green Version]
  65. Moons, K.G.M.; Wolff, R.F.; Riley, R.D.; Whiting, P.F.; Westwood, M.; Collins, G.S.; Reitsma, J.B.; Kleijnen, J.; Mallett, S. PROBAST: A Tool to Assess Risk of Bias and Applicability of Prediction Model Studies: Explanation and Elaboration. Ann. Intern. Med. 2019, 170, W1–W33. [Google Scholar] [CrossRef] [Green Version]
  66. Van Calster, B.; McLernon, D.J.; van Smeden, M.; Wynants, L.; Steyerberg, E.W.; On behalf of Topic Group ‘Evaluating diagnostic, tests prediction models’ of the Stratos initiative. Calibration: The Achilles heel of predictive analytics. BMC Med. 2019, 17, 230. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  67. Wolbers, M.; Koller, M.T.; Witteman, J.C.; Steyerberg, E.W. Prognostic models with competing risks: Methods and application to coronary risk prediction. Epidemiology 2009, 20, 555–561. [Google Scholar] [CrossRef] [PubMed]
  68. Martins, T.; Annichino-Bizzacchi, J.M.; Romano, A.V.C.; Maciel, F.R. Artificial neural networks for prediction of recurrent venous thromboembolism. Int. J. Med. Inform. 2020, 141, 104221. [Google Scholar] [CrossRef] [PubMed]
  69. Gerotziafas, G.; Mahé, I.; Lefkou, E.; AboElnazar, E.; Abdel-Razeq, H.; Taher, A.; Antic, D.; Elalamy, I.; Syrigos, K.; Van Dreden, P. Overview of risk assessment models for venous thromboembolism in ambulatory patients with cancer. Thromb. Res. 2020, 191 (Suppl. 1), S50–S57. [Google Scholar] [CrossRef]
Table
Table 1. Models for predicting VT E in ambulatory patients with lung cancer [17,20,21,22,23,24,25,51,56,57,58,59,60].
Table 1. Models for predicting VT E in ambulatory patients with lung cancer [17,20,21,22,23,24,25,51,56,57,58,59,60].
Name of Model (Author, Year)Cancer Type for Model DerivationPredictorsScoreHigh Risk Validated
By First Author, Year		Cancer Type for Model Validation
Khorana Score
(Khorana, 2008)
[17]		variousCancer tissue: Score ≥ 3 #Alexander 2019 [51]NSCLC
  • Very high-risk site (stomach, pancreas)
2
  • High-risk site (lung, lymphoma, gynaecologic, bladder, testicular)
1van Es 2020 [58]Various (lung cancer 58%)
Platelet count ≥ 350 × 109/L1Vathiotis 2018 [56]Lung adenocarcinoma
Haemoglobin < 100 g/L and/or use of ESA1Kuderer 2018 [25]Lung cancer (84% NSCLC)
Leukocyte count >11 × 109/L1Rupa-Matysek 2018 [23]Lung cancer (97/118 NSCLC)
BMI ≥ 35 kg/m21Mansfield 2016 [57]Lung cancer (87.1% NSCLC)
PROTECHT
(Verso 2012) [20]		variousAs Khorana Score, but Score ≥ 3Alexander 2019 [51]NSCLC
adds gemcitabine chemotherapy, and1Rupa-Matysek 2018 [23]Lung cancer (NSCLC 97/118)
platinum chemotherapy1
CONKO
(Pelzer 2013) [21]		variousAs Khorana Score, but s Score ≥ 3Alexander 2019 [51]NSCLC
removes BMI ≥ 35 kg/m2, and Rupa-Matysek 2018 [23]Lung cancer (NSCLC 97/118)
adds ECOG PS ≥ 21
COMPASS-CAT
(Gerotziafas 2017)
[22]		Various
(13% lung cancer)		Anthracycline treatment6Score ≥ 7Spyropoulos 2020 [60]Various (29.05% lung cancer)
Time since cancer diagnosis ≤ 6 months4
Central venous catheter3
Advanced stage of cancer2
Cardiovascular risk factors present5Syrigos 2018 [24]Lung adenocarcinoma
Hospitalisation for acute medical Illness2Score ≥ 11Rupa-Matysek 2018 [23]Lung cancer (97/118 NSCLC)
A history of VTE1
Platelet count ≥ 350 × 109/L2
ROADMAP-CAT
(Syrigos 2018) [24]		Lung adenocarcinomaProcoag-PPL < 44 s, and
MRI < 125 nM/min
1 †		score = 1
HS D-dimer
(Ferroni 2012) [59]		Lung cancerKhorana Score intermediate group
adds high-sensitive D-Dimer		 Khorana Score 1–2 and
HS D-dimer ≥ 1500 ng/mL
Model 1
(Alexander 2019)
[51]		NSCLCBaseline fibrinogen ≥ 4.0 g/L and baselie D-dimer ≥ 0.5 mg/L1Score ≥ 1Underway ACTRN12618000811202 [51]NSCLC
Baseline D-dimer ≥ 1.5 mg/L1
Month-1 D-dimer ≥ 1.5 mg/L1
# Score ≥ 2 was used to stratify high-risk patients in the CASINI clinical trial of primary thromboprophylaxis and has been incorporated into ASCO Guideline. † binary scoring: score = 1 if Procoag-PPL < 44 s and MRI < 125 nM/min; score = 0 if Procoag-PPL > 44 s or MRI >125 nM/min. VTE: venous thromboembolism; ESA: erythropoiesis stimulating agents; BMI: body mass index; NSCLC: non-small cell lung cancer; ECOG PS: Eastern Cooperative Oncology Group performance status; Procoag-PPL: procoagulant phospholipid-dependent clotting time; MRI: mean rate index of thrombin generation; HS: high-sensitive.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Yan, A.-R.; Samarawickrema, I.; Naunton, M.; Peterson, G.M.; Yip, D.; De Rosa, S.; Mortazavi, R. Risk Factors and Prediction Models for Venous Thromboembolism in Ambulatory Patients with Lung Cancer. Healthcare 2021, 9, 778. https://doi.org/10.3390/healthcare9060778

AMA Style

Yan A-R, Samarawickrema I, Naunton M, Peterson GM, Yip D, De Rosa S, Mortazavi R. Risk Factors and Prediction Models for Venous Thromboembolism in Ambulatory Patients with Lung Cancer. Healthcare. 2021; 9(6):778. https://doi.org/10.3390/healthcare9060778

Chicago/Turabian Style

Yan, Ann-Rong, Indira Samarawickrema, Mark Naunton, Gregory M. Peterson, Desmond Yip, Salvatore De Rosa, and Reza Mortazavi. 2021. "Risk Factors and Prediction Models for Venous Thromboembolism in Ambulatory Patients with Lung Cancer" Healthcare 9, no. 6: 778. https://doi.org/10.3390/healthcare9060778

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop