A new staging system identifies one in four newly diagnosed myeloma patients as high risk but appears less useful in frail and transplant-ineligible patients.
A quarter of newly diagnosed multiple myeloma patients could be flagged as high risk using a new genomic staging system, Australian and New Zealand researchers say.
But real-world data suggests the tool may be least useful in the frail and transplant-ineligible populations where prognosis is often hardest to predict.
The researchers analysed data from 567 patients in the Australia and New Zealand Myeloma and Related Diseases Registry to evaluate the real-world performance of the recently proposed International Myeloma Society/International Myeloma Working Group Consensus Genomic Staging (CGS) system for high-risk multiple myeloma.
The CGS criteria were originally developed using clinical trial datasets to standardise risk stratification in future studies, but their utility in broader patient populations has remained uncertain.
The model classifies patients as high risk if they have at least one of several specific genomic abnormalities, including del(17p), high-risk immunoglobulin heavy chain translocations combined with 1q21 gain or del(1p32), or elevated beta-2 microglobulin levels in the setting of normal renal function.
However, because the criteria were largely derived from clinical trial populations, questions have remained about their applicability in real-world settings, particularly among patients who are older, frail or have significant comorbidities and are often underrepresented in trials.
To assess the model’s performance in clinical practice, investigators analysed data from the Australia and New Zealand Myeloma and Related Diseases Registry, which collects outcomes data from 21 sites across both countries.
Results have been published as a research letter in the British Journal of Haematology.
Of 832 patients diagnosed between 2012 and 2025, 567 had sufficient information to determine risk status using a modified version of the CGS criteria, which excluded genomic markers not routinely assessed in standard care, such as TP53 mutations detected by next-generation sequencing.
Among patients with complete genomic staging data, 22% were classified as high risk. Most met only a single high-risk criterion, while 13% fulfilled two or more criteria.
High-risk patients were significantly more likely to have advanced disease according to existing staging systems, including ISS, R-ISS and R2-ISS.
After a median follow-up of 44 months, high-risk patients had a median overall survival of 45.7 months compared with 85.5 months among standard-risk patients. High-risk status was associated with a 90% increase in mortality risk.
Median progression-free survival on first-line therapy was also significantly shorter, at 23 months versus 38.9 months.
Patients meeting multiple high-risk criteria experienced particularly poor outcomes, with median overall survival of just 37 months and progression-free survival of 15.6 months.
Subgroup analyses revealed marked differences in the model’s predictive performance.
Among transplant-eligible patients, high-risk status was associated with a 2.7-fold increase in mortality risk. In contrast, the association was not statistically significant in transplant-ineligible patients.
Similarly, high-risk genomic features strongly predicted mortality in non-frail patients but not in those classified as frail. A comparable pattern was observed for renal function, with prognostic discrimination strongest among patients with normal creatinine levels.
Related
The researchers suggested that in older, frailer or medically complex patients, competing clinical risk factors may have a greater influence on outcomes than tumour biology alone.
These findings raised questions about the use of genomic staging as the sole basis for treatment intensification strategies in these populations, they said.
The study also examined whether the new genomic model could identify patients with so-called functional high-risk disease, defined as progression within 18 months of commencing first-line therapy.
Although patients classified as high risk had more than double the odds of early progression, the CGS system did not outperform established staging tools.
The revised international staging system demonstrated the highest positive predictive value for early relapse, while the R2-ISS showed the strongest overall discriminative performance.
The researchers noted that despite advances in genomic profiling, many patients who experienced aggressive early relapse remained difficult to identify at diagnosis.
They said that dynamic measures of treatment response would remain essential alongside genomic risk stratification to better identify patients at greatest risk of poor outcomes.
“In summary, IMS/IMWG CGS-HR identifies approximately a quarter of newly diagnosed MM patients as high-risk in a real-world setting,” the researchers concluded.
“IMS/IMWG CGS-HR status predicts significantly inferior survival in patients who were transplant eligible, non-frail or had preserved renal function.
“However, its prognostic impact in transplant-ineligible, frail and renally impaired populations needs further study.
“Despite advances in genomic risk stratification, a substantial proportion of patients with FHR disease remain unrecognised at diagnosis, reinforcing the importance of incorporating early treatment response metrics alongside genomic risk models.”
