New findings suggest that patients with cancer who have non-O blood types (types A, B, and AB) have a higher risk of developing VTEs at 3 months or more after their initial cancer diagnosis.
However, this increased risk was seen only in that time period; in the first 3 months after diagnosis, there was no association between non-O blood type and VTE risk.
Also, the association between non-O blood type and increased VTE risk was seen only in patients who had intermediate and low thrombotic risk tumor types, but not in patients with tumors considered to be very high risk, such as pancreatic, gastroesophageal, and brain cancer.
The study was published online April 13 in Blood Advances.
The authors suggest that the new findings could help to identify patients with cancer who are at increased risk of developing VTE.
“We’ve known tumor type helps determine the baseline risk for VTE. But we continue to see that these risk assessments fail to capture all cancer patients who develop these blood clots,” explained study author Cornelia Englisch, an MD-PhD student at the Medical University of Vienna, Austria, in a statement. “By solely assessing tumor type, we miss up to 50% of people who develop VTE.”
Approached for comment, Gary H. Lyman, MD, MPH, professor in the Public Health Sciences and Clinical Research Divisions at the Fred Hutchinson Cancer Research Center, Seattle, Washington, said further studies are needed before these findings are integrated into clinical practice.
“We need to confirm the risk associated with non-O blood types and whether it is an independent risk factor beyond those that have already been identified and validated,” he told Medscape Medical News. “Nevertheless, if further confirmed as an independent risk factor, it likely will be incorporated into risk assessment and patient discussions moving forward.”
Blood Type Significant After 3 Months
For their study, Englisch and colleagues conducted an analysis of individuals who were participating in the Vienna Cancer and Thrombosis Study (CATS). This is a prospective cohort study that includes patients with newly diagnosed or recurrent cancer who are being observed for the development of VTE.
A total of 1708 patients were included in the analysis. They were followed prospectively for 2 years and information concerning their clinical course was obtained about every 3 months.
The cohort was grouped by blood type and also divided into groups depending on the thrombosis risk profile of their respective tumor type. Pancreatic, gastroesophageal and brain malignancies were considered very high VTE risk tumor types, and the remaining cancer types considered to be intermediate or low risk.
The most common cancer types within this cohort were lung (19%), breast (16%) and brain (14%) cancer and 32% of patients with solid tumors presented with metastatic disease at baseline.
At a median follow up of 24 months, 151 patients (8.8%) developed a VTE. The cumulative VTE incidence was 3.6% at 3 months, 6.2% at 6 months, 7.8% at 12 months, and 9.2% at 24 months. The 2-year mortality was 38%, with 649 patients dying during the 2-year observation period.
Blood type did not vary in patients with different tumor types and tumor risk categories and blood group non-O vs O was not associated with an increased likelihood of having stage IV disease. There was also no association between ABO blood group type and survival.
During the first 3 months of follow-up, patients with blood type non-O and O had a similar risk of VTE (subdistribution hazard ratio [SHR], 1.00; P = .992). But between 3 and 24 months, blood type non-O had a significantly increased association with a higher risk of VTE (SHR, 1.79; P = .015). This association remained after adjusting for multivariables, including age, sex, disease stage (stage IV vs others) and body mass index (adjusted SHR, 1.77; P = .017).
A second multivariable analysis showed that the association of non-O blood type and increased VTE risk beyond 3 months follow-up remained independent after adjusting for platelet count, sP-selectin levels, and D-Dimer (adjusted SHR, 1.75; P = .026).
The authors also looked at the association of von Willebrand factor (vWF) and factor VIII (FVIII) with blood type. Levels of vWF-antigen (ag) and FVIII activity in patients with blood type non-O were compared to blood type O patients; those with type non-O had higher levels of vWF-Ag and FVIII activity. After adjusting for continuous levels of FVIII in a multivariable model, the association of non-O blood type with VTE risk was weakened in both strength and magnitude for the ≥3-month follow-up period (adjusted SHR, 1.47; P = .109) and in the subgroup of patients with low-to-intermediate risk tumor types (adjusted SHR, 1.59; P = .057).
Validation and Confirmation Needed
Lyman noted that an increased risk for VTE with non-O blood type has been reported previously in patients without cancer and in a few small studies in patients with cancer. “The report here extends the evidence, albeit still exploratory, that patients with non-ABO blood groups who are otherwise at low or intermediate risk for VTE are actually at increased risk over time, while there appears to be little additional risk in those already at high risk for VTE,” he said.
Lyman explained that his group had previously developed the most commonly utilized VTE risk score for patients receiving cancer chemotherapy (Blood 2008;111(10):4902-7). “Risk with this score is based on cancer type, hemoglobin, white blood count, platelet count and body mass index, all of which are available to clinicians at the time of evaluation and treatment,” he said. “This risk score has been validated in more than a dozen studies and has been incorporated into guidelines and clinical practice.”
However, the risk model is not perfect and they are always looking for ways to improve the accuracy of the model. “Although preliminary, the results reported here appear encouraging, in that considering the patients ABO blood group may improve prediction in otherwise low- or intermediate-risk patients with cancer beyond 3 months,” he said. “If this is validated in other studies as an independent predictive factor in low- and intermediate-risk patients, it is likely to subsequently be incorporated into guidelines and clinical practice to identify patients who may be candidates for thromboprophylaxis with anticoagulation.”
Another expert also weighed in on the study.
“Prior studies have indeed shown an association of blood group types with risk of VTE, although most such studies have been conducted in a noncancer population,” said Alok Khorana, MD, professor of medicine, Cleveland Clinic Lerner College of Medicine, Case Western Reserve University, Ohio. “The findings of this paper are very interesting and thought-provoking, particularly in terms of mechanistically how this association takes place.”
It is important to note, Khorana emphasized, that these findings only apply to intermediate- and low-risk cancers and there was no association within the first 3 months of follow-up, which is when most VTE events occur. “Adjustment for FVIII also weakened this association, and due to these caveats, I would not recommend using it as a clinical risk factor just yet, but the authors have identified some very interesting associations that are deserving of further study and analyses,” he said.
“In the meantime, cancer-associated VTE remains an important contributor to morbidity and mortality in people with cancer,” he emphasized.
Korana also pointed out that recent revisions of clinical guidelines suggest or recommend consideration of primary prevention in high-risk patients and “public health efforts to reduce the burden of VTE in cancer should remain focused on improving compliance with primary prevention.”
The authors have disclosed no relevant financial relationships. Khorana has no disclosures. Lyman reports r esearch grant support from Amgen to institution, and consultancy for Beyond Spring, G1 Therapeutics, Sandoz, Samsung Bioepis, Merck, and Fresenius Kabi outside the current work.
Blood Adv. Published online April 13, 2022. Abstract