Viral Load Tests Can Help Predict Mpox Severity

Mpox is a viral infection that causes flu-like symptoms and a characteristic rash, which evolves significantly over time and varies between patients. The disease spreads mainly through direct contact with skin lesions, and patients remain contagious until the lesions fully heal. However, predicting how severe a case will be—or how long a patient may remain infectious—has been a major challenge. Two distinct variants of the virus exist: clade I, which has a higher mortality rate of up to 10%, and clade II, associated with lower mortality. A new study has identified that measuring the level of virus in the blood at the time skin lesions first appear could help predict whether a patient will experience a mild or severe case, offering an important clinical tool for early intervention.

This study was led by researchers at Nagoya University (Nagoya, Japan), in collaboration with other institutions, and focused on developing a predictive model for disease severity based on early viral loads. The research team analyzed medical records from 2007 to 2011 of patients infected with clade Ia mpox in the Democratic Republic of the Congo, the country most affected by the disease. They measured viral load in blood samples taken at the onset of visible skin lesions and used mathematical modelling and machine learning to interpret the data. Patterns in lesion development and symptom severity were used to determine how long patients took to recover and how sick they became. The model identified a viral load threshold of approximately 40,000 copies/mL—above which patients were significantly more likely to suffer severe, long-lasting symptoms and potentially remain contagious for longer.

The study, published in Science Translational Medicine, demonstrated that patients naturally fall into two groups—those with mild symptoms and quick recovery, and those with more severe, prolonged illness. These findings are especially timely as clade I mpox spreads across Africa, prompting renewed global health concern. By enabling early prediction of disease severity, this method could help doctors allocate resources more efficiently, identify high-risk cases faster, and provide more intensive treatment where needed. The team now plans to test the model on the clade Ib variant, which is part of the ongoing outbreak, and explore the use of this tool in guiding real-time clinical decisions in affected regions.

“If this method can be applied to current circulating mpox strains, we can move toward more personalized, data-driven medicine,” said co-lead author Shingo Iwami, professor at the Nagoya University Graduate School of Science. “For patients and their families, this could provide clearer expectations about recovery timelines and reassurance through more precise medical predictions after a frightening diagnosis.”

Related Links:
Nagoya University