Predicting Multiple Myeloma Risk: A New Era in Early Detection
Multiple myeloma (MM) is a cancer of plasma cells, a type of white blood cell. Early detection is crucial for improving treatment outcomes, but identifying individuals at high risk before symptoms appear has been a significant challenge. Recent research, though, is changing that landscape, offering promising tools for proactive risk assessment.
A study published in the British Journal of Haematology explored whether readily available clinical and laboratory markers could predict the progress of MM. Researchers at Clalit Health Services in Israel analyzed electronic health records (EHRs) of patients who were later diagnosed with MM, comparing their data to a control group who remained MM-free. the goal? To pinpoint patterns and variables associated with increased risk.
The initial analysis involved 4256 patients diagnosed with MM between 2002 and 2019, carefully matched with a larger control group (over 42,000 individuals) based on age, sex, and location. Investigators meticulously reviewed EHR data from five years prior to diagnosis, examining over 200 different clinical and lab parameters.
Initially, a complex machine learning model was developed. While accurate, it demanded considerable computational resources – a barrier to widespread implementation. Recognizing this limitation, the team developed a simplified model, designed for use by community physicians with standard resources.
This streamlined model focused on 20 key variables. patients who ultimately developed MM exhibited specific patterns: higher erythrocyte sedimentation rates (a marker of inflammation), lower hemoglobin levels, reduced absolute neutrophil counts, and decreased neutrophil/lymphocyte ratios. Elevated levels of globulins and ferritin were also observed. This simplified model achieved an area under the receiver operator characteristic (AUC) of 0.72, indicating a good level of predictive accuracy.
What does this mean for patients and clinicians?
The potential impact is significant.This model offers a pathway to earlier detection, potentially allowing for intervention before the disease progresses. This is particularly relevant given previous research demonstrating the benefits of early treatment. A study in the New England Journal of Medicine showed that high-risk smoldering MM patients treated with lenalidomide plus dexamethasone experienced a substantially longer time to disease progression compared to those under observation.
Though, implementing such a model isn’t without considerations. A critical decision involves setting a risk threshold. A lower threshold would increase detection rates but also lead to more false positives and increased testing costs.A higher threshold would reduce costs but risk missing early-stage cases. Furthermore,the authors acknowledge the need for external validation – testing the model’s accuracy in diverse populations – to ensure its reliability.
Despite these caveats, the researchers are optimistic. They beleive their findings provide a practical, actionable tool for clinicians, empowering them to proactively identify individuals at increased risk of developing multiple myeloma. This represents a potential paradigm shift, moving from reactive treatment to proactive prevention and early intervention.
References:
- Mittelman M, Israel A, Oster HS, et al. Can we identify individuals at risk to develop multiple myeloma? A machine learning-based predictive model. Br J Haematol. Published online June 16, 2025. doi:10.1111/bjh.20136
- Mateos MV, Hernández MT, Giraldo P, et al. Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma. N Engl J Med. 2013;369(5):438-447. doi:10.1056/NEJMoa1300439
