Clinical Trials May Misrepresent True Thromboembolic Risks From Cancer Drugs

Researchers have identified that the reporting of venous and arterial thrombotic events in cancer clinical trials is inconsistent and potentially inaccurate, according to a comments article published in the Journal of Clinical Oncology.¹

“Mandatory and uniform reporting of all [venous and arterial thromboembolism events] in cancer trials is needed to ethically guide both the initial risk assessment and informed patient consenting of drug-associated [venous and arterial thromboembolism] in cancer-directed therapy,” the study authors stressed in their published commentary.

Thromboembolic events are possible to prevent, but prophylactic intervention is unlikely to be given when clinicians are unaware of the true extent of the risk. “Primary prevention of thromboembolic events is possible; however, it requires timely implementation of prophylactic strategies, as well as the accurate quantification of risk and mitigation of risk factors,” the authors added, led by corresponding author David C. Calverley, MD, Section of Hematology/Oncology, Division of Hospital and Specialty Medicine, Portland VA Medical Center in Oregon, and Division of Medical Oncology, University of British Columbia Faculty of Medicine, Vancouver, British Columbia.

Background

Thromboembolism has previously been indicated as a top non–cancer-related cause of mortality among patients with cancer who are receiving outpatient chemotherapy, tied with infections.² Cardiovascular adverse events, often caused by venous and arterial thromboembolism, are also a significant driver of non–cancer-related morbidity in patients with cancer.

A real-world analysis of venous thromboembolism in patients with cancer receiving chemotherapy found that 13.5% of patients experienced venous thromboembolism 12 months after starting chemotherapy, with ranges varying from 9.8% to 21.3% by the location of the cancer in the body.³ Thrombotic events are also more likely to occur within the first few months of a cancer diagnosis and decrease over time. This makes it even more important to understand the true burden of thromboembolism as thrombosis-inducing therapies are moved further up in treatment algorithms.

Thromboembolic adverse events can be prevented when patients are given prophylaxis, such as with rivaroxaban. A large randomized, controlled trial of primary prophylaxis for thromboembolic events with rivaroxaban at 10 mg once daily in ambulatory patients with cancer who were initiating systemic therapy showed that prophylaxis led to a 60% reduction in the composite risk of venous thromboembolism or venous thromboembolic–related death vs placebo. It also reduced the risk of arterial and visceral thromboembolic events by 38% compared with placebo.⁴

Inaccurate Reporting

The study authors pointed out specific instances of underreported or misreported venous thromboembolic adverse events from clinical trials of cancer therapeutics in patients with gastrointestinal tumors.

One such example was the study of chemotherapy schedules in patients with advanced colorectal cancer. The clinical trial reported a 1% rate of venous thromboembolism among participants in the study, yet further detailed chart review revealed that at least 10.2% of the study participants had venous thromboembolic events, making it the most common grade 3/4 adverse event in the study.⁵

Additionally, a systemic review of randomized clinical trials of patients with pancreatic cancer found that the reported rates of venous thromboembolism were much lower among the 13 trials in which the primary endpoint was focused on the efficacy of the chemotherapy treatment vs the nine trials focused on the efficacy of thromboprophylaxis (5.9 vs 16.5%; P < .001).⁶ Rates of venous thromboembolism from the thromboprophylaxis studies more closely matched those of real-world incidences of venous thromboembolism.⁷

Possible Reasons for Discrepancies

Calverley et al suggested that there are several possible factors contributing to the reporting discrepancies for thromboembolic events from randomized clinical trials of patients with cancer:

• The Common Terminology Criteria for Adverse Events poorly categorizes thromboembolic events
• There are no formal criteria or biomarkers for the assessment of thromboembolic events as treatment-related or not
• Asymptomatic venous thromboembolisms are important but often not reported due to a lack of awareness of recurrence risks with deep venous thrombosis, pulmonary embolus, etc.
• Trials ended too early
• Patients could be discontinued from treatment if the thromboembolism event is reported
• Lack of awareness of the clinical significance of certain thromboembolic events
• Drug company proprietary data

Based on these factors, Calverley et al also recommended a number of possible solutions to these issues:

• Incorporate International Society on Thrombosis and Haemostasis (ISTH) criteria for the reporting of thromboembolic events in phase III trials
• Report all thromboembolic events consistently, without causal bias
• Document serious and nonserious events separately
• Consider reporting on venous and arterial thromboembolisms as part of prespecified secondary endpoints
• Make trial follow-ups longer to match the duration of administration
• Harmonize definitions of venous/arterial thromboembolic events that require reporting as serious vs nonserious events (eg, catheter-associated, splanchnic, and superficial venous thrombosis) separately from each other
• Include descriptors of location and extent of the thromboembolic event

The researchers noted that “physicians vary their approach regarding [venous thromboembolism] attribution to the anticancer therapy under investigation. Some providers report all [venous thromboembolism] events, whereas others fail to report [venous thromboembolism] or [arterial thromboembolism] events based on subjective drug attribution judgments.”

They further noted that venous thromboembolic incidence in oncology studies should be reported to account for the competing risk of death, and that a failure to do so can artificially lower estimates for venous thromboembolism rates.

Specific Challenges and Events

Central venous catheter–related thrombosis can occur asymptomatically in about 14% to 18% of patients, with concerning symptoms in about 5%. The risk of recurrent venous thromboembolism is even higher in the setting of central venous catheter–related thrombosis, which led to various organizations creating guidelines to recommend the use of therapeutic-dose anticoagulants for at least 3 months and for the duration of the presence of the catheter.⁸

Even incidental events are important to be aware of for patients with cancer, as they can impact the patient’s prognosis. Incidental venous thromboembolic events discovered on CT imaging can lead to poorer clinical outcome risks similar to those from symptomatic events.⁹

Rates and risks for arterial thromboembolism in patients with cancer are less commonly reported, and studies of chemotherapy and targeted therapy agents often exclude patients who have had recent arterial thromboembolic events. A systematic review and meta-analysis of arterial thromboembolic events from clinical trials of vascular endothelial growth factor–targeted tyrosine kinase inhibitors showed an incidence rate of 1.4%,¹⁰ yet real-world data showed a rate of 3.99% when patients with a cardiovascular history and risk factors are included.¹¹

Reports of thromboembolic events were especially uncommon in the early phase III randomized clinical trials that led to the approvals of immune checkpoint inhibitors. Since then, an analysis of 18 real-world studies for patients receiving immune checkpoint inhibitors showed a cumulative incidence of 5% to 8% for venous thromboembolism at 6 months and more than 10% at 1 year.¹²

Real-world rates for arterial thromboembolic events with immune checkpoint inhibitors were 1% to 5% at 12 months compared with 1.1% at 3 years among RCTs with several of the latter trials not reporting any ATE event rates.^12,13Further, one study showed a possible delayed temporal effect of immune checkpoint inhibitor–related arterial thromboembolism, with a 41% risk of arterial thromboembolism at 1 year and a 97% risk at 4 years.¹⁴ 

Call to Action

Dr. Calverley and his colleagues made a number of recommendations to oncology professionals to try to improve the accuracy of reporting on venous and arterial thromboembolic events in clinical trials:

• Venous/arterial thromboembolic events should be included as prespecified secondary or safety endpoints in clinical trials to enable the most accurate toxicity reporting
• Reporting of all events with an appropriate categorization and regardless of drug causality should be mandatory
• The reporting of these events should be added to the CONSORT Harms reporting metrics that is required of all randomized clinical trials submitted for publication¹⁵
• Standardizing real-world quality assessments is recommended for toxicities to determine how events impact patients with comorbidities
• Reliable thrombosis risk assessment models should be developed for targeted and immunotherapy agents based on quality real-world data
• Follow-up plans for clinical trials should address the underestimation and undervaluing of such significant adverse events, and drug companies should be encouraged to conduct quality real-world trials postapproval to address toxicities
• Definitions of serious vs nonserious venous and arterial thromboembolic events should be updated and harmonized
• Individual participant data from randomized and other pivotal clinical trials should be publicly accessible as a requirement for publication in any journal

Calverley et al believe that these recommendations would provide ethical guidance for risk assessments and practical guidance for the timely administration of prophylactic treatments as needed for the greatest possible mitigation of risk and improvement of patient outcomes and quality of life.

“Care in a multidisciplinary setting (ie, primary care, preventive cardiology, and thrombosis specialists) is paramount to providing the best available care to this growing patient population but is reliant on the accuracy of data available to inform practice and change,” the authors concluded.

DISCLOSURE: Dr. Calverley reported no competing financial interests.

REFERENCES

1. Calverley DC, Leader A, Cheong MA, et al: Inconsistent and inaccurate cancer clinical trial reporting of venous and arterial thrombotic events: An urgent call to action. J Clin Oncol 43:28515-2855, 2025.
2. Khorana AA, Francis CW, Culakova E, et al: Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. J Thromb Haemost 5:632-634, 2007.
3. Lyman GH, Eckert L, Wang Y, et al: Venous thromboembolism risk in patients with cancer receiving chemotherapy: A real‐world analysis. Oncologist 18:1321-1329, 2013.
4. Khorana AA, Soff GA, Kakkar AK, et al: Rivaroxaban for thromboprophylaxis in high-risk ambulatory patients with cancer. N Engl J Med 380:720-728, 2019.
5. Mandalà M, Barni S, Floriani I, et al: Incidence and clinical implications of venous thromboembolism in advanced colorectal cancer patients: the 'GISCAD-alternating schedule' study findings. Eur J Cancer 45:65-73, 2009.
6. Chiasakul T, Patell R, Maraveyas A, et al: Discordant reporting of VTE in pancreatic cancer: A systematic review and meta-analysis of thromboprophylaxis versus chemotherapeutic trials. J Thromb Haemost 19:489-501, 2021.
7. Chew HK, Wun T, Harvey D, et al: Incidence of venous thromboembolism and its effect on survival among patients with common cancers. Arch Intern Med 166:458-464, 2006.
8. Lee AY, Kamphuisen PW: Epidemiology and prevention of catheter-related thrombosis in patients with cancer. J Thromb Haemost 10:1491-1499, 2012.
9. Lyman GH, Carrier M, Ay C, et al: American Society of Hematology 2021 guidelines for management of venous thromboembolism: Prevention and treatment in patients with cancer. Blood Adv 5:927-974, 2021.
10. Choueiri TK, Schutz FA, Je Y, et al: Risk of arterial thromboembolic events with sunitinib and sorafenib: A systematic review and meta-analysis of clinical trials. J Clin Oncol 28:2280-2285, 2010.
11. Vallerio P, Orenti A, Tosi F, et al: Major adverse cardiovascular events associated with VEGF-targeted anticancer tyrosine kinase inhibitors: A real-life study and proposed algorithm for proactive management. ESMO Open 7:100338, 2022.
12. Wang TF, Khorana AA, Carrier M: Thrombotic complications associated with immune checkpoint inhibitors. Cancers 13:4606, 2021.
13. Solinas C, Saba L, Sganzerla P, et al: Venous and arterial thromboembolic events with immune checkpoint inhibitors: A systematic review. Thromb Res 196:444-453, 2020.
14. Zhu J, Chen Y, Zhang Y, et al: Association of immune checkpoint inhibitors therapy with arterial thromboembolic events in cancer patients: A retrospective cohort study. Cancer Med 12:18531-18541, 2023.
15. Junqueira DR, Zorzela L, Golder S, et al: CONSORT Harms 2022 statement, explanation, and elaboration: Updated guideline for the reporting of harms in randomised trials. BMJ 381:e073725, 2023

EXPERT POINT OF VIEW

Alok Khorana, MD, Professor of Medicine, Cleveland Clinic Lerner College of Medicine, Case Western Reserve University, the Sondra and Stephen Hardis Chair in Oncology Research, Vice-Chair for Clinical Services of the Taussig Cancer Institute, and Director of the Gastrointestinal Malignancies Program at the Cleveland Clinic, urges physicians and industry members not to think of venous and arterial thromboembolic events as prohibitive but as a toxicity that can be managed with proper understanding and prophylactic management.

“Venous and arterial thromboembolism are well-known complications of both cancer and treatments for cancer. Because the association of cancer with thrombosis is so well known, the occurrence of such events is often assumed to be related to the underlying malignancy. However, risk of thromboembolic events in cancer can be increased substantially by a variety of treatments, ranging from traditional chemotherapy to immune checkpoint inhibitors to mutation-targeting agents to supportive care drugs such as erythropoietin. Because of this multifactorial etiology, assigning causality can be challenging, and risk with emerging drugs can often be missed.”

“Calverley and colleagues have issued an important call to action to improve the accuracy of reporting of both venous and arterial thromboembolic events in the setting of new drug development in malignancy. Assigning causality should not be the priority, rather mandatory reporting of events will allow a better understanding of the risk and whether it is elevated with a specific drug or not. The type, timing, and seriousness of events should also be harmonized with how such events are evaluated in practice rather than shoehorning events into artificial categories of common toxicity criteria.”

“It is important to note that a variety of agents for thromboprophylaxis are available, effective, and mostly safe. An elevated risk of thromboembolism should therefore not be viewed as an event that could derail drug development but rather as a complication whose risk can be mitigated with appropriate information.”

DISCLOSURE: Dr.Khorana received consulting honoraria from Regeneron, Anthos, Novartis, Pfizer, BMS, Sanofi, WebMD, Alnylam, and Sirius.