Oncofusion

start / graph / cost-frontline-infant-leukemia-care

Explainermoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q8-impact

What it costs to treat a child through frontline leukemia therapy

also called cost of ALL induction · infant leukemia treatment cost

Just the first thirty days of leukemia treatment costs a US children's hospital somewhere between $32,000 and $47,000 per child, depending on how long that hospital keeps children in. Nobody has published what a full two-year infant leukemia course costs, which is a real gap.

The same thing, for a scientist

There is no published total cost of care for infant acute leukemia. What exists is the induction segment: across 30 US children's hospitals in the Pediatric Health Information System, median 30-day inpatient cost per patient from initial admission was $32,000 at hospitals with a short typical induction length of stay, $40,000 at medium and $47,000 at long, with median LOS varying from 5 to 31 days between institutions. Complications move this sharply: induction hyperglycemia requiring insulin was associated with longer stay, more ICU admission and higher cost. By contrast, the drugs themselves are cheap: the entire national supply of chemotherapy and chemoprotective agents to treat every child diagnosed with ALL in Thailand in 2017 was estimated at US$814,952, about US$2,563 per child.

Go deeper
Induction, the only well-measured segment2 deeper

Using the Pediatric Health Information System across US children's hospitals, ALL induction admissions from 2007-2013 showed median LOS varying from 5 to 31 days between institutions, and 30-day median inpatient cost per patient of $32,000 (short-LOS hospitals), $40,000 (medium) and $47,000 (long) 1.

The same study found that hospitals admitting new ALL patients for more than seven days had lower odds of PICU readmission (OR 0.68 medium, 0.31 long) and, for long-LOS hospitals, lower odds of any readmission (OR 0.44), at substantially higher induction cost 1. That is a genuine cost-versus-safety trade, not waste.

These are children of all ages with ALL; infants are a small minority and are systematically sicker, so the infant figure is almost certainly higher. No infant-specific figure was found.

Complications are the cost driver, not the protocol

Hyperglycemia requiring insulin during induction, which affects 10-50% of paediatric ALL patients, was associated in PHIS with increased mortality after induction, more subsequent infection, more need for transplant, more relapse, longer inpatient stay, higher cost of care and more ICU admission 2.

This matters for the platform argument in a direction people usually miss: if a targeted agent lets you de-intensify chemotherapy, the saving is not mainly in drug cost, it is in avoided complications and avoided intensive care. Any health-economic case for menin inhibition in infants should be built on complication avoidance and transplant avoidance, not on drug substitution.

The drugs are almost free; the care is not

A costing exercise for Thailand estimated that all ten anticancer medicines and four chemoprotective agents needed to give every child diagnosed with ALL in the country a full course of national-guideline chemotherapy in 2017 would cost US$814,952 in total, which is 0.005% of Thailand's total health expenditure; extending to undiagnosed children raised it to US$1,365,422. All fourteen agents are available as generics and thirteen are on the WHO essential medicines list for children 3.

Against 318 diagnosed children, that is roughly US$2,563 of drug per child for an entire multi-year ALL course. One month of adult revumenib at wholesale acquisition cost, approximately $79,000 (see cost-menin-inhibitor-course), buys about thirty-one of those complete courses.

That contrast is the honest economic framing of targeted therapy in this disease. It is not an argument against menin inhibitors, whose comparator is death rather than cheap chemotherapy. It is an argument that any global-access claim attached to the platform has to survive a three-order-of-magnitude price step.

What could not be found

A deliberate negative result, recorded so nobody repeats the search.

  • No published total cost of care for infant (<1 year) acute leukemia, in any health system, was located in PubMed as of 2026-09-18.
  • No cost analysis of the Interfant protocols. Interfant-06 and Interfant-21 report outcomes, not costs.
  • No cost-effectiveness analysis of a menin inhibitor in any population, paediatric or adult.

The first and third of these are tractable and would be genuinely publishable work for a group with PHIS or national claims access. The absence of a menin-inhibitor cost-effectiveness analysis in particular means that when payers begin to push back on a $474,000 minimum course, there is currently no analysis on the table to defend it.

Facts and figures
thailand national chemo drug budget usd814952
thailand children diagnosed318
thailand drug cost per child usd2563
induction 30day inpatient cost usd
{
  "short_los_hospitals": 32000,
  "medium_los_hospitals": 40000,
  "long_los_hospitals": 47000,
  "los_range_days": [
    5,
    31
  ],
  "year_range": "2007-2013",
  "population": "paediatric ALL, all ages"
}
not found
[
  "total cost of care for infant (<1y) acute leukemia, any health system",
  "cost analysis of the Interfant protocols",
  "cost-effectiveness analysis of any menin inhibitor"
]
How this could bite us back
  • If the cost case for a targeted agent rests on avoided transplant and avoided intensive care, then any agent that adds its own toxicity burden (differentiation syndrome, QTc prolongation, prolonged cytopenias) erodes its own economic case as well as its clinical one.
  • Building a global-access narrative on a drug priced three orders of magnitude above the chemotherapy it supplements invites the response that the same money would save more children if spent on supportive care in low-resource settings. That objection deserves a real answer, not a slogan.
Still open
  1. What is the full episode cost of an Interfant-06 or Interfant-21 course for an infant, including transplant where it happens? PHIS or a national claims linkage could answer this directly.
  2. How much of the cost difference between infants and older children with ALL is driven by ICU use and infection rather than by protocol intensity?
  3. Would a menin inhibitor that avoided transplant in even a fifth of infants pay for itself at current list price? See `cost-allogeneic-hsct-paediatric-leukemia` for the comparator.
Every source behind this entry (3)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Across US children's hospitals in the Pediatric Health Information System (2007-2013), median induction length of stay for newly diagnosed ALL varied from 5 to 31 days between institutions, and 30-day median inpatient cost per patient was $32,000 at short-LOS, $40,000 at medium-LOS and $47,000 at long-LOS hospitals, with longer-LOS hospitals showing lower odds of PICU readmission and of any readmission.

    The Effects of Hospital Length of Stay on Readmissions for Children With Newly Diagnosed Acute Lymphoblastic Leukemia · Wedekind MF, Dennis R, Sturm M, et al. · Journal of Pediatric Hematology/Oncology 2016
    PMID 26989914DOI
    What this evidence cannot show

    Administrative charge data converted to costs, 2007-2013, so figures are a decade old and understate current dollars. Covers all paediatric ALL, not infants, and only the first 30 days. Hospital-level LOS categories are a crude exposure and residual confounding by case mix is likely.

    Evidence that argues against this (1)
    • Jurisdiction and cost year make this entry's headline range unusable without restatement. The $32,000-$47,000 figures are US children's-hospital costs derived from charges via cost-to-charge ratios over 2007-2013, and no inflation adjustment or payer-allowed conversion is applied. For a same-disease, same-country comparison a decade later, the 2025 JMCP budget-impact model of relapsed/refractory acute leukaemia uses cost inputs in 2024 US dollars covering drug acquisition, administration, grade 3 or greater adverse events, supportive care and monitoring, subsequent treatment and end-of-life care, and is explicit that wholesale acquisition cost from RED BOOK was used for drug costs. Any comparison between this entry's induction segment and a modern targeted-therapy course has to restate both to a common year and perspective; as written they are not comparable numbers. Full text of the 2025 model read at PMC12205252. Abraham I et al. 2025
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In the Pediatric Health Information System, paediatric ALL patients requiring insulin for hyperglycemia during induction had increased post-induction mortality, more infectious complications, more subsequent transplant and relapse, longer inpatient stay, higher cost of care and more ICU admission.

    PMID 32589365DOI
    What this evidence cannot show

    Retrospective administrative-database study using ICD-9 codes and medication charges; hyperglycemia is a marker of higher-risk disease and older age, so association is not causation and the cost increment is not quantified in dollars in the abstract.

    Evidence that argues against this (1)
    • Insulin-requiring hyperglycaemia during ALL induction is driven principally by corticosteroids and asparaginase, not by the leukaemia, so it is a cost of the REGIMEN rather than of the disease. It also comes from an administrative database where insulin administration is a billing event, which captures treatment intensity rather than glycaemic severity. Pediatric Blood & Cancer 2020
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Providing the full national-guideline course of ten anticancer and four chemoprotective agents to all 318 children diagnosed with ALL in Thailand in 2017 was estimated to cost US$814,952, 0.005% of national health expenditure, rising to US$1,365,422 including undiagnosed children; all fourteen agents are available as generics and thirteen are WHO essential medicines for children.

    PMID 33109678DOI
    What this evidence cannot show

    Drug acquisition cost only. It excludes hospitalisation, supportive care, transfusion, diagnostics, transplant and the workforce, which are the dominant costs everywhere. Thai generic prices are not transferable to high-income systems. The per-child figure of ~US$2,563 is this node's arithmetic on the published total and case count.

    Evidence that argues against this (1)
    • The entry's conclusion that 'the drugs themselves are cheap' holds only for generic cytotoxics and inverts once a targeted agent enters. Treating all 318 Thai children diagnosed with ALL in 2017 with the full national-guideline course of fourteen agents was estimated at US$814,952, about US$2,563 per child. The 2025 JMCP model enters revumenib at $36,867 per 28-day cycle (RED BOOK wholesale acquisition cost, 2024 USD, US). One 28-day cycle of one menin inhibitor at US list price is therefore about 4.5% of the entire annual national ALL chemotherapy budget for Thailand, and roughly 14 times the per-child cost of a complete conventional course. The two figures are in different jurisdictions and different years and cannot be netted, which is precisely the point: the 'drugs are cheap' claim is a statement about off-patent generics in a middle-income market and does not survive the arrival of a patented targeted agent. Both sources read (Thai abstract via PubMed, JMCP full text at PMC12205252). Abraham I et al. 2025
Connected to