Younger Age at Diagnosis Is Associated with an Increased Risk of Venous Thromboembolism in Multiple Myeloma

Introduction

Multiple Myeloma (MM) patients are at an approximately 9-fold increased risk of developing venous thromboembolism (VTE), with risk being highest in the first year following diagnosis¹. VTE is associated with significant morbidity and negatively influences survival in MM². Although the Khorana score has been shown to predict rate of thrombosis in solid tumors, the validity of this score in haematological malignancies has yet to be assessed. Given the elevated rates of VTE in these conditions, in particular MM, clinically relevant risk prediction scores are essential. Additionally, data from the MRC-XI trial indicates that standard thromboprophylaxis may not prevent VTEs in MM³. Therefore identification of risk factors for MM-VTE are required to improve our understanding of the pathophysiology of thrombosis and to develop risk-adapted clinical practice guidelines. Through interrogation of an extensive clinical database we sought to identify factors predictive for VTE in our MM population.

Methods

We performed a retrospective cohort study of all newly diagnosed MM patients at our centre from 2001-2017. Patient medical records were reviewed for clinical and laboratory data including FBC parameters, beta-2-microglobulin, paraprotein and serum free light chain on the day of diagnosis, to minimize steroid effect. All VTE events were recorded, along with MM treatment regimen and thromboprophylaxis at time of event. History of thrombosis was defined as occurring within 6 month prior to, or following a diagnosis of MM. Patients with MGUS or smoldering MM were excluded. Statistical analysis including logistic regression and cox proportional hazard modelling was performed using SPSS (IBM Analytics, USA). A comparison of mortality was also performed between age matched cases with VTE and controls without VTE.

Results

Over a period of 17 years, 266 patients were diagnosed with myeloma, of which 34 (12.7%) developed VTE following MM diagnosis or within the preceding six months. The mean age of the VTE cohort at MM diagnosis was younger than the mean age of the non-VTE cohort (62.5 years vs. 68.6 years). Pulmonary embolisms and deep vein thromboses were equally represented (44% and 56% respectively) and additional risk factors for thrombosis were present in 46% of patients, not related to MM therapy. Of the patients on immunomodulatory drugs or corticosteroids at time of VTE, all were receiving thromboprophylaxis with either low molecular weight heparin (LMWH) or aspirin at time of VTE.

The mortality odds ratio was 3.3 (95% CI 2.4-4.5) in patients who developed VTE in comparison to age matched controls with MM. Younger age at MM diagnosis (<64 years) predicted for VTE occurrence in logistic regression univariate (p-value=0.002) and multivariate analysis (p=0.004). Higher white cell count (WCC) at MM diagnosis showed a trend toward significance in univariate analysis (p-value=0.06) and, in combination with age, demonstrated an area under the curve of 0.72 on ROC analysis for prediction of VTE. Interestingly, the increased risk of VTE in younger patients was not related to longer duration of MM exposure or longer follow up as there was no statistically significant difference in time to VTE between all age groups (median 9 months). Other parameters incorporated in the Khorana score, such as haemoglobin and platelet count did not increase the risk of VTE (p-value=0.57, and 0.25 respectively).

Conclusions

Our data confirms that VTE is associated with an increased mortality in MM patients and estimates the risk of death to be 3.3 fold higher in these patients. As recently reported in a large cohort of MM patients, younger age is associated with an increased risk of VTE development⁴, our data support this finding and excludes longer duration of MM, and follow-up time, as confounding variables. Importantly, our data confirms, in unselected "real world" patients the signal that is now apparent from analysis of VTE in the MRC-XI trial³, that thromboprophylaxis with LMWH or aspirin is suboptimal for VTE prevention. This may point to alternative thrombotic mechanisms in MM-VTE and further data in larger MM cohorts is needed to develop risk adapted strategies for prevention strategies for these patients.

References

1. Kristinsson SY et el, Blood. 2008 Nov 1;112(9):3582-6.

2. Schoen MW et al. J Clin Oncol 36, 2018 (suppl; abstr 8051).

3. Bradbury CA et al, Blood 2017 130:553.

4. Sanfilippo KM et al. Blood 2016;128:4726.