Higher-risk MDS care in England carries a heavy NHS cost before and after treatment failure

A literature-based micro-cost analysis has put a sharp price tag on the routine care pathway for higher-risk myelodysplastic syndromes in England. The model, built from published sources and national tariff data, estimated a total direct NHS provider cost of £146,921 per patient for an illustrative non-curative pathway that included diagnosis, 12 cycles of azacitidine, post-hypomethylating agent supportive care, and terminal care.

The burden was not spread evenly across the disease course. Costs rose steeply during active treatment, remained substantial after hypomethylating agent failure, and climbed again at the end of life, where inpatient care became dominant.

Azacitidine treatment drives the largest monthly spend

During active azacitidine therapy, the per-patient-per-month cost reached £8721.58. In practical terms, this is the period when drug acquisition, administration, transfusions, and routine clinical monitoring stack together into a sustained monthly expense. In a disease like HR-MDS, where treatment is rarely curative, the economic pressure does not come from a single event but from repeated cycles of outpatient delivery and support care layered over months.

The base-case pathway assumed 12 cycles of azacitidine, each given for 7 days within a 28-day cycle. That is a long enough treatment period for cumulative costs to matter, even before considering complications, repeated transfusion needs, or admission rates seen in routine practice. Hematologic malignancies often create this kind of layered resource use: the drug itself is only one part of the bill, while blood product support and clinical contact consume a major share of system resources.

Failure of hypomethylating therapy leaves a costly supportive-care phase

Once hypomethylating agent therapy failed, the monthly cost remained high at £5399.58. This phase is clinically important because it reflects a transition from disease-directed treatment to supportive management, when transfusion dependence and hospital contact often intensify. Patients no longer receive the same structured treatment rhythm, but they do not become inexpensive to care for. The opposite is usually true.

The study’s design excluded patients who went on to allogeneic haematopoietic stem cell transplantation or transformed to acute myeloid leukemia, since those pathways follow different diagnostic and therapeutic trajectories. Even with those exclusions, the economic burden stayed substantial. For health-system planners, this is the point at which real-world resource use starts to diverge from simplified treatment algorithms: once first-line therapy no longer controls the disease, admissions and transfusion support can rapidly dominate service demand.

Terminal care shifts the cost burden toward hospital delivery

End-of-life care produced the highest monthly cost in the model at £12,554.58 for hospital-dominant terminal care. A separate scenario assuming hospice-dominant terminal care lowered the overall pathway estimate, but only to a range of approximately £136,840 to £140,990 depending on the hospice bed-day cost applied.

The difference is clinically intuitive. Hospital-based terminal care in a transfusion-dependent blood cancer typically carries high bed-day and acute-care costs, while hospice-based care shifts spending into a different setting with a different tariff structure. The final month of life in HR-MDS therefore becomes a major cost inflection point, not because the biology suddenly changes, but because the place of care does.

Diagnostic genomics adds a one-time cost before treatment begins

The initial diagnostic work-up, including genomics, was estimated at £2710 to £2810. In a disease where risk stratification shapes treatment intent and prognosis, that upfront expense is part of the pathway rather than an optional extra. Molecular and cytogenetic assessment helps define disease biology, yet it also increases early administrative and laboratory costs before the first cycle of therapy starts.

In the base-case model, that diagnostic expense was included as a one-off cost at month 0. An alternative pathway excluding genomic testing reduced the total cost, but the overall burden still remained high because most spending was concentrated later in the disease course. Even without the genomic component, the pathway did not become inexpensive; it simply shifted a modest share of costs out of the diagnostic phase.

What the £146,921 figure means for NHS planning

The total estimated direct NHS provider cost of £146,921 per patient is not a population prevalence estimate and should not be read as the cost of all HR-MDS care in England. It is a modelled cost for a specific illustrative pathway in routine practice. Even so, the figure is useful because it captures how quickly a non-curative hematologic malignancy can accumulate expense across sequential phases of care.

By design, the analysis used published literature and publicly available price lists rather than individual patient data. That approach makes the estimate transparent and reproducible, but it also means the model depends on the quality and applicability of the underlying studies. Cycle-level or phase-level transfusion and admission rates were used when available in the literature for adults with HR-MDS receiving first-line azacitidine in routine practice.

Cost concentration matters more than the average alone

The monthly cost pattern is the real signal here. Active azacitidine therapy, post-failure supportive care, and terminal care each sit at different points on the care continuum, yet all three phases create heavy pressure on NHS resources. The total cost is meaningful, but the phase-specific distribution is what service planners can actually use when building budgets, estimating transfusion demand, or modelling future pathways.

A single number can obscure the clinical work behind it. This model shows the opposite: HR-MDS spending concentrates where disease control becomes difficult and where care shifts from outpatient treatment to repeated support or inpatient management. In blood cancers like this, the economic trajectory often mirrors the clinical one. As marrow failure worsens, transfusion dependence grows, admissions rise, and the end-of-life setting becomes a major determinant of cost.

The analysis also placed its findings in the context of a National Institute for Health and Care Excellence technology appraisal for azacitidine, uprated to £124,848 for 2024/2025. The comparison is informative but limited by differences in population, model structure, treatment duration, and price year. It offers context, not validation.

What remains most actionable is the phase-specific estimate itself. The per-patient-per-month costs can support UK budget-impact analyses, service planning, and future economic models, provided local validation and scenario testing are done before the numbers are used for decision-making.


Reporting by Dr. Revathy Mohan V.R., MBBS for Medicine World Foundation, with additional data provided by BMJ Open.