Risk stratification of PE in the cancer patient

During the ICTHIC webinar titled “Management of CAT: Lung Cancer Focus,” Professor António Araújo delivered a lecture focused on the stratification of pulmonary embolism risk in cancer patients. Here, we summarize the main points from his speech. You can watch António Araújo’s lecture in the video below and the full webinar recording here.

Western countries are dealing with a concerning incidence rate of pulmonary embolism (PE), estimated to range from 60 to 120 cases per 100,000 individuals annually. This condition carries a daunting in-hospital mortality rate of 14% and a 90-day mortality rate of 20%. PE also increases the risk of death and the potential for perilous complications, including pulmonary infarction and chronic thromboembolic pulmonary hypertension [1,2].

In addition, data from the USA reveals that nearly 5% of all cancer cases involve cancer-associated thrombosis (CAT), with PE ranking as the most common CAT event. Some of these cases may be attributed to incidental PE detected for frequent use of CT scans in cancer patients. Furthermore, European data show that incidental PE is responsible for almost half of pulmonary events [3,4].

When dissecting PE cases by cancer type, we observed that tumors with a higher thrombogenic potential, such as ovarian, pancreatic, lung, brain and gastric cancers, exhibit a greater incidence of PE. Indeed, PE ranks as the most frequent event in these cases [5].

Therefore, patient stratification becomes imperative because PE patients have a heightened risk of rehospitalization, with almost half requiring readmission within 1 year. Identifying these high-risk patients is crucial for developing targeted interventions to reduce the rehospitalization risk [6].

Prognostic risk stratification models

To address these challenges, prognostic stratification is essential. This stratification can categorize patients into different risk groups, namely low-risk and high-risk categories. Low-risk patients can potentially be managed in an outpatient setting with reduced supportive therapy, while high-risk patients may require hospitalization and close outpatient monitoring.

Prognostic stratification enhances patients’ quality of life and reduces costs within the healthcare system.

It is worth noting that in the general population, several clinical prognostic models exist, such as the Geneva Prognostic Score (GPS) model, the Pulmonary Embolism Severity Index (PESI) model, and the simplification of the PESI (sPESI) model, each utilizing various numbers of variables [7].

More recently, two additional clinical prognostic models have been published. The BOVA model incorporates variables such as clinical factors, radiologic right ventricle dysfunction, and biomarkers of myocardial damage, with cancer not being a prognostic variable. Another model, the FAST score, utilizes clinical variables, such as the presence of syncope and tachycardia [7].

However, it is important to note that these five clinical prognostic models are primarily designed for the general population. The issue arises when considering their applicability to cancer patients, as the proportion of cancer patients included in these prognostic model cohorts represents only 18–24% of cases [7]. Consequently, there is a need for cancer-specific tools that exhibit superior prognostic accuracy compared to their generic counterparts [8,9].

The POMPE-c, RIETE and FONT models

This need is addressed by models, such as the POMPE-c, RIETE and Font criteria, which identify a substantially larger proportion of PE patients, including those with cancer, who are likely to survive 30 days while maintaining comparable sensitivity to genetic tools [7].

The POMPE-c score is derived from the US EMPEROR registry, which involved approximately 1,880 PE patients, including 408 with cancer. This score takes into account the patient’s body weight and other clinical factors. However, it does not incorporate new oncological settings, such as the use of new drugs, and does not consider long life expectancy. Additionally, the POMPE-c score lacks external validation and does not discriminate between competing causes of death or contemplate incidental events [7].

Similarly, the RIETE score, which includes six clinical variables easily assessed at the bedside, was derived from 1,048 patients and internally validated in 508 patients. Like the POMPE-c score, the RIETE score does not incorporate new oncological settings and has a cutoff at 80 years old and a body weight of 60 kg. Additionally, it dichotomizes disease progression but lacks external validation and does not differentiate between competing causes of death or incidental events [7].

Lastly, the EPIPHANY index predicts serious complications within 15 days of diagnosis. It comprises six variables, three highly specific and commonly used in oncology. These variables include the Eastern Cooperative Oncology Group (ECOG) performance status, PE-specific symptoms, focal stock symptoms, and whether the patient has undergone surgery for the primary tumor [7].

Importantly, the EPIPHANY index has been externally validated for cancer patients in the emergency department and applies to a wide range of PE scenarios, encompassing both incidental and symptomatic events [10,11].

The EPIPHANY index stands out as the only model currently available that allows us to classify the risk of PE in cancer patients.

Development of the EPIPHANY index

The development of this index aimed to stratify the risk of serious complications within 15 days after a PE diagnosis, correlating the outcome with 39 variables and some clinical oncology practice variables to construct a decision tree model [12].

Through rigorous statistical analysis, variables with the most discriminatory and predictive power were identified, ultimately classifying patients into three risk groups: low, intermediate and high. Patients falling into the low- and intermediate-risk categories could be managed on an outpatient basis with close monitoring, while those in the high-risk category would require hospitalization [12].

Six variables considered in the EPIPHANY index are based on a modified HESTIA score. These variables include systolic blood pressure less than 100 mmHg, oxygen saturation <90%, a respiratory rate ≥30 breaths per minute, a pulse >110 beats per minute, sudden or progressive dyspnea, and high hemorrhagic risk, indicated by platelet counts <50,000/ml [12].

In addition to these variables, the EPIPHANY index also accounts for the patient’s ECOG performance status, the specific symptoms of PE, and whether there is an ongoing or recent resection of the primary tumor [12].

It’s worth noting that this study included a substantial number of patients, with 73.6% experiencing cancer progression and 53.6% undergoing chemotherapy. Incidental PE was found in 53% of patients, and all patients received anticoagulant therapy, primarily in the form of low-molecular-weight heparin [12].

Of the 1,075 patients enrolled in the study, 208 developed serious complications within 15 days of their PE diagnosis. The mortality rate within 15 days after diagnosis was 10.1% [12].

Data analysis revealed that the risk groups discriminated by the EPIPHANY index effectively distributed the probability of serious complications and death, demonstrating statistically significant differences. This validation was conducted externally, confirming the utility of the EPIPHANY model [10].

The EPIPHANY model is a valuable tool for standardized decision-making when lacking high-quality evidence. It effectively manages patients with previous incidental or symptomatic PE events related to cancer and contributes to a better understanding of clinical and epidemiological patterns among these patients [12].

To facilitate clinical decision-making, a calculator was developed that categorizes patients into low, intermediate, and high-risk groups, providing severe complications and mortality rates with statistically significant differences. This information guides treatment decisions, allowing for outpatient management, outpatient monitoring, or hospitalization, as appropriate.

Comments on the factors to consider when assessing the need for hospitalization

However, when dealing with low-risk PE cases, it’s essential to consider the clinical severity of the embolism. Factors such as the patient being asymptomatic, having only minor symptoms, or the PE being suspected or incidental may indicate suitability for outpatient treatment. Hospitalization should be reserved for cases with acute bleeding events, a high risk of bleeding, thrombocytopenia (platelet count below 50,000/ml), or contraindications for anticoagulation therapy. Careful attention to these factors helps differentiate and manage low-risk PE patients effectively.

It is also crucial to consider cancer-related conditions when determining whether a patient should be hospitalized. Hospitalization may be warranted if the patient has uncontrolled cancer-related symptoms, if there is a need for an initial cancer evaluation involving invasive tests, if concomitant cancer-related complications are present, or if there is evidence of cancer progression requiring additional diagnostic tests. In such cases, hospitalization is advisable. However, outpatient treatment may be appropriate for patients without these specific conditions.

Additionally, hospitalization may be necessary for patients with other comorbidities, especially when concurrent medical complications require intensive care. For example, pregnancy is a situation that may necessitate hospitalization.

Furthermore, the decision to hospitalize should consider the patient’s ability to receive adequate monitoring for PE after discharge. Additionally, psychosocial factors and patient preferences should be considered. This preference should be considered if a patient prefers to avoid hospitalization but requires long-term care.

Conclusion

In the era of precision medicine, it is imperative to tailor our decisions, even for cancer patients with PE, to minimize unnecessary hospital admissions, optimize the allocation of healthcare resources, and enhance patient care.

Moving forward, further research is needed in this area. There is limited data, particularly regarding the risk-benefit analysis of fibrinolysis in cases of unsuspected or incidental PE and the use of inferior vena cava filter placement. Additionally, there is a need to develop models specific to palliative care in this context.

Watch António Araújo’s lecture:

References

1. Freund Y, Cohen-Aubart F, Bloom B. Acute pulmonary embolism: a review. JAMA. 2022;328(13):1336-1345. doi:10.1001/jama.2022.16815
2. Hua X, Han SH, Wei SZ, Wu Y, Sha J, Zhu XL. Clinical features of pulmonary embolism in patients with lung cancer: A meta-analysis. PLoS One. 2019;14(9):e0223230. doi:10.1371/journal.pone.0223230
3. Font C, Carmona-Bayonas A, Fernández-Martinez A, et al. Outpatient management of pulmonary embolism in cancer: data on a prospective cohort of 138 consecutive patients. J Natl Compr Canc Netw. 2014;12(3):365-373. doi:10.6004/jnccn.2014.0038
4. Font C, Carmona-Bayonas A, Beato C, et al. Clinical features and short-term outcomes of cancer patients with suspected and unsuspected pulmonary embolism: the EPIPHANY study. Eur Respir J. 2017;49(1):1600282. doi:10.1183/13993003.00282-2016
5. Svendsen E, Karwinski B. Prevalence of pulmonary embolism at necropsy in patients with cancer. J Clin Pathol. 1989;42(8):805-9. doi: 10.1136/jcp.42.8.805.
6. Sindet-Pedersen C, El-Chouli M, Nourhavesh N, et al. High risk of rehospitalization within one year following a pulmonary embolism-Insights from the Danish nationwide registries from 2000-2020. Eur Heart J Qual Care Clin Outcomes. 2023. doi:10.1093/ehjqcco/qcad046. Epub ahead of print.
7. Muñoz-Guglielmetti D, Cooksley T, Ahn S, et al. Risk stratification for clinical severity of pulmonary embolism in patients with cancer: a narrative review and MASCC clinical guidance for daily care. Support Care Cancer. 2022;30(10):8527-8538. doi:10.1007/s00520-022-07131-1
8. Li X, Hu Y, Lin P, et al. Comparison of different clinical prognostic scores in patients with pulmonary embolism and active cancer. Thromb Haemost. 2021;121(6):834-844. doi:10.1055/a-1355-3549
9. Weeda ER, Caranfa JT, Zeichner SB, Coleman CI, Nguyen E, Kohn CG. External validation of generic and cancer-specific risk stratification tools in patients with pulmonary embolism and active cancer. J Natl Compr Canc Netw. 2017;15(12):1476-1482. doi:10.6004/jnccn.2017.7004
10. Ahn S, Cooksley T, Banala S, Buffardi L, Rice TW. Validation of the EPIPHANY index for predicting risk of serious complications in cancer patients with incidental pulmonary embolism. Support Care Cancer. 2018;26(10):3601-3607. doi:10.1007/s00520-018-4235-9
11. Sánchez-Cánovas M, Jimenez-Fonseca P, Fernández Garay D, et al. Prediction of serious complications in patients with pulmonary thromboembolism and solid cancer: Validation of the EPIPHANY Index in a prospective cohort of patients from the PERSEO study. PLoS One. 2023;18(5):e0266305. doi:10.1371/journal.pone.0266305
12. Carmona-Bayonas A, Jiménez-Fonseca P, Font C, et al. Predicting serious complications in patients with cancer and pulmonary embolism using decision tree modelling: the EPIPHANY Index. Br J Cancer. 2017;116(8):994-1001. doi:10.1038/bjc.2017.48