Anthracycline cardiotoxicity: the heart an infant keeps for sixty years
also called daunorubicin cardiotoxicity infant leukemia · mitoxantrone cardiotoxicity Interfant · cardiomyopathy after infant leukemia therapy · anthracycline-sparing argument
The drugs that cure infant leukemia include ones that permanently damage heart muscle, and a baby's heart is more vulnerable than an older child's. The damage does not show up at the end of treatment; it shows up twenty and thirty years later, which is why a drug that avoids these agents would be worth a great deal even if it only matched current cure rates.
The same thing, for a scientist
Interfant-06 delivers cumulative daunorubicin of 180 mg/m2 on the standard arm (60 mg/m2 in induction plus 120 mg/m2 in OCTADAD) and 210 mg/m2 plus mitoxantrone 36 mg/m2 on the AML-style experimental arm, all before the child's first birthday and with body-surface-area doses reduced to two-thirds under 6 months and three-quarters from 6-12 months [E1]. Applying the empirically derived late-cardiomyopathy equivalence ratios of 0.6 for daunorubicin and 10.5 for mitoxantrone [E3] puts the standard arm at roughly 108 mg/m2 doxorubicin-equivalent and the experimental arm at roughly 504 mg/m2 - a four- to five-fold difference in predicted lifetime cardiomyopathy exposure between two randomised arms that produced indistinguishable six-year disease-free survival (39.3% vs 36.8%, log-rank P = 0.47) [E2]. Age at exposure is an independent risk factor separate from cumulative dose: in 115 children treated with doxorubicin for ALL, age under four years at treatment independently predicted increased left ventricular afterload (P = 0.003) after adjustment for cumulative dose, and 57% of the cohort had abnormal afterload or contractility a median of 6.4 years later [E4]. Dose-response begins low: relative to no exposure, cardiomyopathy odds ratios rise from 1.65 at 1-100 mg/m2 to 3.85 at 101-150, 7.23 at 201-250, 23.47 at 251-300 and 27.59 above 300 mg/m2 [E5]. This is the single strongest quantitative argument for a mechanism-directed agent that removes anthracycline from the infant backbone.
What Interfant-06 actually gives, in mg/m21 deeper
From the Interfant-06 protocol document, the anthracycline and anthraquinone schedule is 1:
| Course | Agent | Dose per administration | Administrations | Course total |
|---|---|---|---|---|
| Induction (all patients) | daunorubicin | 30 mg/m² IV over 60 min, days 8 and 9 | 2 | 60 mg/m² |
| ADE (experimental arm) | daunorubicin | 50 mg/m² over 1 h, days 1, 3, 5 | 3 | 150 mg/m² |
| MAE (experimental arm) | mitoxantrone | 12 mg/m² over 1 h, days 1, 3, 5 | 3 | 36 mg/m² |
| OCTADAD (standard arm) | daunorubicin | 30 mg/m² over 60 min, days 1, 8, 15, 22 | 4 | 120 mg/m² |
| OCTADA (experimental arm) | daunorubicin | omitted by design | 0 | 0 |
Cumulative totals per arm:
| Arm | Daunorubicin | Mitoxantrone | Etoposide |
|---|---|---|---|
| Standard (induction + IB + MARMA + OCTADAD) | 180 mg/m² | 0 | 0 |
| Experimental (induction + ADE + MAE + MARMA + OCTADA) | 210 mg/m² | 36 mg/m² | 1,000 mg/m² |
The protocol explicitly omits daunorubicin from OCTADA in the experimental arm "because of cumulative dose of anthracyclines", which shows the designers were already at their internal ceiling 1. Cardiac monitoring is built in: echocardiography before the first daunorubicin dose and after each subsequent 60 mg/m² (not corrected for age), with daunorubicin omitted for the course if left ventricular shortening fraction is repeatedly below 27% 1.
Two design details matter enormously and both cut in the same direction. First, doses are reduced for the youngest: two-thirds of the body-surface-area-calculated dose under 6 months and three-quarters from 6 through 12 months 1. Second, the AML-style courses ADE and MAE were imported wholesale from paediatric AML therapy - so the experimental arm of Interfant-06 is the cleanest available window onto what AML-intensity anthracycline exposure looks like when delivered to an infant.
The protocol's own cardiotoxicity assumption was wrong
The Interfant-06 protocol states, in its drug-specific guidance, that "Mitoxantrone is less cardiotoxic than daunorubicin" 1. This reflected the standard 2006 assumption that haematologic potency tracks cardiotoxicity.
It is now contradicted. Pooling 28,423 five-year survivors from the Childhood Cancer Survivor Study, the Dutch DCOG-LATER study and the St Jude Lifetime study, with 399 cardiomyopathy cases over a median 20 years of follow-up, the late-cardiomyopathy equivalence ratios relative to doxorubicin were 0.6 (95% CI 0.4-1.0) for daunorubicin and 10.5 (95% CI 6.2-19.1) for mitoxantrone 3. The authors' own conclusion was that the conventional 4:1 haematologic equivalency "appeared to significantly underestimate" mitoxantrone's long-term cardiomyopathy risk 3.
Per milligram, mitoxantrone is therefore roughly seventeen times more cardiotoxic than daunorubicin for late cardiomyopathy. The 36 mg/m² of mitoxantrone in MAE is not a rounding error on the anthracycline budget; on these ratios it is the largest single item in it.
The arithmetic that makes the case for an anthracycline-sparing agent2 deeper
Combining the protocol doses 1 with the empirical equivalence ratios 3 gives doxorubicin-equivalent cumulative exposure by arm. This calculation is the author's arithmetic, not a published figure, and inherits the wide confidence interval on the mitoxantrone ratio.
| Arm | Daunorubicin x 0.6 | Mitoxantrone x 10.5 | Total doxorubicin-equivalent | Range using the mitoxantrone 95% CI (6.2-19.1) |
|---|---|---|---|---|
| Standard | 180 x 0.6 = 108 mg/m² | - | ~108 mg/m² | 108 |
| Experimental (AML-style) | 210 x 0.6 = 126 mg/m² | 36 x 10.5 = 378 mg/m² | ~504 mg/m² | ~349-814 mg/m² |
Now place those on the measured dose-response curve. Relative to no anthracycline exposure, the odds ratios for cardiomyopathy in a matched case-control study of 170 childhood cancer survivors with cardiomyopathy versus 317 without were 1.65 at 1-100 mg/m², 3.85 at 101-150, 3.69 at 151-200, 7.23 at 201-250, 23.47 at 251-300 and 27.59 above 300 mg/m² (P for trend < 0.001) 5. The standard arm sits around the 101-150 band; the experimental arm sits above 300.
And here is the finding that turns this from a toxicity note into a strategic argument: the two arms produced the same outcome. Six-year disease-free survival was 39.3% (SE 4.0; n = 169) with ADE+MAE and 36.8% (SE 3.9; n = 161) with protocol IB, log-rank P = 0.47, and overall Interfant-06 did not improve on Interfant-99 2. Infants in the experimental arm accepted a several-fold increase in predicted lifetime cardiomyopathy risk and received, on the trial's own primary comparison, nothing in return.
Why a baby is not a small child
Age at exposure is an independent risk factor, not merely a proxy for higher weight-adjusted dosing. In 115 children in continuous remission after doxorubicin for ALL, evaluated a median of 6.4 years after treatment, 57% had abnormalities of left ventricular afterload or contractility. Cumulative dose was the strongest single predictor (P < 0.002). But in the multivariate analysis restricted to patients who had received at least 228 mg/m², the only two significant predictors were higher cumulative dose (P = 0.01, predicting decreased contractility) and age under four years at treatment (P = 0.003, predicting increased afterload) 4.
The mechanism the authors proposed explains why infancy is the worst time: doxorubicin impairs myocardial growth. Increased afterload came from reduced ventricular wall thickness rather than from hypertension or dilatation, and afterload rose progressively in 24 of 34 patients followed serially (71%) 4. A heart that loses myocytes at eight months old never builds the muscle mass it was going to build, and the deficit is then amortised over an adult body for six decades. Eleven patients in that cohort had congestive heart failure within a year of treatment; five had recurrent failure 3.7-10.3 years after completing doxorubicin and two required heart transplantation 4.
Subclinical now, clinical later
The prevalence figures depend entirely on whether the question is 'does the survivor have symptoms' or 'is the echocardiogram abnormal'.
- Subclinical, early: 57% abnormal afterload or contractility at a median 6.4 years 4; 65% of those exposed to >= 228 mg/m² had increased afterload, decreased contractility or both 4.
- Subclinical, progressive: in a longitudinal series of 499 echocardiograms from 115 doxorubicin-treated ALL survivors (median cumulative dose 352 mg/m², median follow-up 11.8 years), fractional-shortening z-score transiently improved and then fell to -2.76 more than 12 years after diagnosis, with deficits present even after low cumulative doses (see counter-evidence on 4).
- Clinical, decades out: in 14,358 five-year survivors versus 3,899 siblings, the hazard ratio for congestive heart failure was 5.9 (95% CI 3.4-9.6), exposure to >= 250 mg/m² of anthracycline raised the hazard of heart failure, pericardial disease and valvular abnormality two- to five-fold, and cumulative incidence was still rising 30 years after diagnosis (see counter-evidence on 5).
- Clinically ascertained lifetime: estimated cumulative prevalence of cardiomyopathy by age 50 among at-risk survivors of childhood cancer was 21.6% (95% CI 19.3-23.9) (see counter-evidence on 5).
None of these cohorts is infant-specific. Every one of them is dominated by children treated at older ages, and each therefore understates rather than overstates the exposure-adjusted risk for a survivor treated in infancy, since age at exposure is an independent risk factor 4.
Mitigation: dexrazoxane and lifelong surveillance1 deeper
Dexrazoxane is the only pharmacological cardioprotectant with randomised long-term paediatric evidence. In DFCI 95-01, children with high-risk ALL were randomised to ten doses of doxorubicin 30 mg/m² (300 mg/m² cumulative) with or without dexrazoxane 300 mg/m² before each dose. Five years after completing doxorubicin, mean left ventricular fractional-shortening z-score was significantly worse than normal in the doxorubicin-alone group (-0.82, 95% CI -1.31 to -0.33) but not in the dexrazoxane group (-0.41, 95% CI -0.88 to 0.06), and dexrazoxane significantly protected left ventricular wall thickness and thickness-to-dimension ratio. Event-free survival was 77% versus 76% (P = 0.99) at a median 8.7 years, so oncological efficacy was not compromised (see counter-evidence on 1 for the full citation). The genuine weakness: the protective effect on fractional shortening at five years was significant in girls (1.17, 95% CI 0.24-2.11) but not in boys (-0.10, 95% CI -0.87 to 0.68).
Dexrazoxane has never been tested in a randomised infant-leukemia trial. Bounded 2026-09-19: PubMed dexrazoxane[Title/Abstract] AND (infant OR infants)[Title/Abstract] returns 3 records, and none is a leukaemia trial - a doxorubicin-extravasation case report (PMID 40306310), a bioflavonoid topoisomerase-II mechanism paper (PMID 32087851) and a proposed congenital-heart-surgery trial (PMID 23050818). The randomised dexrazoxane evidence in children comes from older-child ALL and Hodgkin protocols, not from infants. Its use in Interfant-06 is not part of the protocol backbone 1.
Surveillance is the other half of mitigation and it is lifelong. The International Late Effects of Childhood Cancer Guideline Harmonization Group produced harmonised cardiomyopathy surveillance recommendations precisely because congestive heart failure in this population is a progressive disorder with a variable asymptomatic period preceding symptoms (see counter-evidence on 5). For a child treated at six months old, 'lifelong' means approximately eighty years of echocardiographic follow-up begun before the patient can remember the exposure.
The argument stated in one paragraph
An infant treated on Interfant-06 receives 180-210 mg/m² of daunorubicin and, on the AML-style arm, 36 mg/m² of mitoxantrone, before the age of one, into a myocardium that is still growing, with age under four years an independent predictor of abnormal ventricular geometry 1 4. The measured dose-response says cardiomyopathy odds are already elevated in the 101-150 mg/m² band and rise 27-fold above 300 mg/m² 5. The randomised comparison of the higher-anthracycline arm against the lower one showed no disease-free survival benefit 2. A targeted agent that replaces even the OCTADAD daunorubicin block would remove 120 mg/m² of daunorubicin - two-thirds of the standard-arm total - and if it replaced the AML-style courses it would remove the mitoxantrone, which on current equivalence ratios is the single largest contributor to predicted lifetime cardiomyopathy risk in the whole protocol 3. That is a first-class, quantified, non-survival reason to develop the drug.
| kind | toxicity |
|---|---|
| severity | Severe and irreversible; expressed as progressive left ventricular dysfunction over decades rather than as acute toxicity. Estimated cumulative prevalence of cardiomyopathy by age 50 among at-risk childhood cancer survivors is 21.6% (95% CI 19.3-23.9); cardiomyopathy odds rise 27.6-fold above 300 mg/m2 cumulative anthracycline; subclinical afterload or contractility abnormalities are present in 57% of doxorubicin-treated childhood ALL survivors a median 6.4 years after treatment. |
| mitigation | Dose ceilings and arm design (Interfant-06 omits daunorubicin from OCTADA in the experimental arm explicitly because of cumulative anthracycline dose); age-based body-surface-area dose reduction (two-thirds under 6 months, three-quarters at 6-12 months); mandatory echocardiography before the first daunorubicin dose and after each subsequent 60 mg/m2 with omission if shortening fraction is repeatedly < 27%; dexrazoxane 300 mg/m2 before each doxorubicin dose (randomised evidence of long-term protection in children with high-risk ALL without loss of efficacy, but significant in girls only and never randomised in infants); lifelong risk-stratified cardiomyopathy surveillance per the International Late Effects Guideline Harmonization Group; and, prospectively, replacement of the anthracycline-containing blocks by a mechanism-directed agent. |
affects
[ "drug-daunorubicin", "drug-mitoxantrone", "drug-doxorubicin", "trial-interfant-06" ]
interfant 06 doses
{
"source": "Interfant-06 protocol, SKION version 01, April 2006",
"induction_daunorubicin_mg_per_m2_per_dose": 30,
"induction_daunorubicin_doses": 2,
"induction_daunorubicin_total_mg_per_m2": 60,
"ade_daunorubicin_mg_per_m2_per_dose": 50,
"ade_daunorubicin_doses": 3,
"ade_daunorubicin_total_mg_per_m2": 150,
"mae_mitoxantrone_mg_per_m2_per_dose": 12,
"mae_mitoxantrone_doses": 3,
"mae_mitoxantrone_total_mg_per_m2": 36,
"octadad_daunorubicin_mg_per_m2_per_dose": 30,
"octadad_daunorubicin_doses": 4,
"octadad_daunorubicin_total_mg_per_m2": 120,
"octada_daunorubicin": "omitted by design because of cumulative anthracycline dose",
"ade_mae_etoposide_total_mg_per_m2": 1000,
"cumulative_daunorubicin_standard_arm_mg_per_m2": 180,
"cumulative_daunorubicin_experimental_arm_mg_per_m2": 210,
"cumulative_mitoxantrone_experimental_arm_mg_per_m2": 36,
"bsa_dose_reduction_under_6_months": "2/3 of calculated dose",
"bsa_dose_reduction_6_to_12_months": "3/4 of calculated dose",
"cardiac_stopping_rule": "echo before first daunorubicin and after each 60 mg/m2; omit daunorubicin for the course if LVSF repeatedly < 27%"
}doxorubicin equivalent estimate
{
"method": "author's arithmetic: Interfant-06 planned cumulative doses multiplied by the late-cardiomyopathy equivalence ratios of Feijen et al. 2019 (daunorubicin 0.6, mitoxantrone 10.5)",
"standard_arm_mg_per_m2": 108,
"experimental_arm_mg_per_m2": 504,
"experimental_arm_range_using_mitoxantrone_ci": [
349,
814
],
"fold_difference_between_arms": 4.7,
"caveat": "Not a published figure. Mitoxantrone ratio 95% CI spans 6.2-19.1 and the mitoxantrone-doxorubicin relationship is better fitted by a linear-exponential model, so the point estimate is indicative only."
}dose response odds ratios
{
"source": "Blanco et al. 2012, Children's Oncology Group case-control (170 cases / 317 controls)",
"reference": "0 mg/m2",
"bands_mg_per_m2": {
"1-100": 1.65,
"101-150": 3.85,
"151-200": 3.69,
"201-250": 7.23,
"251-300": 23.47,
">300": 27.59
},
"p_trend": "<0.001"
}age at exposure
{
"finding": "age under 4 years at treatment independently predicted increased left ventricular afterload after adjustment for cumulative dose",
"p_value": 0.003,
"cohort": "115 doxorubicin-treated childhood ALL survivors, Lipshultz 1991",
"mechanism": "impaired myocardial growth and inadequate ventricular mass rather than acute myocyte loss alone"
}long term prevalence
{
"subclinical_abnormal_afterload_or_contractility_pct": 57,
"subclinical_timepoint": "median 6.4 years after doxorubicin",
"fractional_shortening_z_score_beyond_12_years": -2.76,
"chf_hazard_ratio_vs_siblings": 5.9,
"chf_hr_ci95": [
3.4,
9.6
],
"anthracycline_threshold_ccss_mg_per_m2": 250,
"cumulative_cardiomyopathy_prevalence_by_age_50_pct": 21.6,
"cumulative_cardiomyopathy_ci95": [
19.3,
23.9
]
}dexrazoxane
{
"trial": "DFCI 95-01",
"randomised": [
100,
105
],
"analysed": [
66,
68
],
"doxorubicin_cumulative_mg_per_m2": 300,
"dexrazoxane_mg_per_m2_per_dose": 300,
"lvfs_z_score_at_5y_doxorubicin_alone": -0.82,
"lvfs_z_score_at_5y_doxorubicin_alone_ci95": [
-1.31,
-0.33
],
"lvfs_z_score_at_5y_with_dexrazoxane": -0.41,
"lvfs_z_score_at_5y_with_dexrazoxane_ci95": [
-0.88,
0.06
],
"efs_pct": [
77,
76
],
"efs_p_value": 0.99,
"sex_interaction": "significant protection in girls, not in boys",
"tested_in_infants": false
}randomised outcome of the higher anthracycline arm
{
"trial": "Interfant-06",
"arms": [
"ADE + MAE (AML-style)",
"protocol IB (lymphoid)"
],
"n": [
169,
161
],
"six_year_dfs_pct": [
39.3,
36.8
],
"se": [
4,
3.9
],
"log_rank_p": 0.47
}bounded searches
[
{
"query": "dexrazoxane[Title/Abstract] AND (infant OR infants)[Title/Abstract]",
"records": 3,
"relevant": 0,
"date": "2026-09-19",
"source": "PubMed eutils esearch",
"pmids_returned": [
"40306310",
"32087851",
"23050818"
]
}
]- Anthracyclines may be doing real work in KMT2A-rearranged leukemia beyond generic cytotoxicity, so removing them could cost efficacy in ways the Interfant-06 null randomisation does not exclude: that trial compared ADE+MAE against IB, not anthracycline against no anthracycline, and every arm retained the 60 mg/m2 induction daunorubicin.
- If a targeted agent must be given for years rather than months to hold remission, the chronic-exposure toxicity profile of the new drug may simply replace the cardiac profile of the old one, with the added disadvantage that the new risk is unquantified.
- A regimen that drops anthracycline but raises glucocorticoid, alkylator or intrathecal exposure to compensate moves the late-effect burden rather than reducing it.
- Cardiac surveillance is itself a lifelong burden - eighty years of echocardiograms for a child treated at six months - and a drug that avoids anthracycline only escapes that obligation if it can be shown not to have its own late cardiac signal, which requires decades of follow-up nobody has yet.
- What is the actual delivered cumulative anthracycline dose per patient in Interfant-06, and what is the measured rate of subclinical and clinical cardiac dysfunction in those survivors? Neither has been published, so the whole quantitative case here rests on planned protocol doses and borrowed age-at-exposure effects.
- Does dexrazoxane protect infants, and is it safe to give alongside the mitoxantrone in MAE? It has never been randomised in an infant leukemia trial, yet infants are the age group in which age at exposure is most adverse.
- Is the mitoxantrone-to-doxorubicin equivalence ratio of 10.5 correct in infants? It was derived from only 265 mitoxantrone-exposed survivors of all ages and carries a three-fold confidence interval, and everything in the AML-style-arm calculation depends on it.
- If a menin inhibitor or similar agent replaced the OCTADAD anthracycline block, would leukemia control be preserved in the low-risk and medium-risk strata where that block currently sits?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Interfant-06 delivers daunorubicin 30 mg/m2 on days 8 and 9 of induction (60 mg/m2), daunorubicin 50 mg/m2 x 3 in ADE (150 mg/m2), mitoxantrone 12 mg/m2 x 3 in MAE (36 mg/m2) and daunorubicin 30 mg/m2 x 4 in OCTADAD (120 mg/m2), with daunorubicin omitted from OCTADA in the experimental arm because of cumulative anthracycline dose; doses are reduced to two-thirds of the body-surface-area calculation under 6 months and three-quarters from 6-12 months; echocardiography is required before the first daunorubicin dose and after each subsequent 60 mg/m2, with the drug omitted if shortening fraction is repeatedly below 27%.
Interfant-06 treatment protocol for infants under 1 year of age with acute lymphoblastic leukaemia (SKION version 01, April 2006) - sections 5.1-5.6 and drug-specific guidance · Interfant-06 study group / SKION (Dutch Childhood Oncology Group) · Interfant-06 study protocol document 2006What this evidence cannot show
This is a trial protocol document, not a peer-reviewed publication, and it is the SKION (Dutch) national version of April 2006; later amendments exist and the protocol was amended during the trial (for example asparaginase changes in MARMA and OCTADAD). Doses given are planned doses; actual delivered cumulative dose per patient will be lower wherever courses were omitted, delayed, or curtailed by the built-in shortening-fraction stopping rule, and the trial report does not publish delivered cumulative anthracycline per patient. Doses are per m2 of a reduced body-surface-area calculation, so mg/kg exposure in an infant is not directly comparable to an older child's.
Evidence that argues against this (1)
- The anthracycline load is not immovable: a randomised trial in children with high-risk ALL showed that dexrazoxane 300 mg/m2 before each of ten 30 mg/m2 doxorubicin doses preserved left ventricular fractional shortening and wall thickness at 5 years without compromising event-free survival (77% vs 76%, P = 0.99) - though the protection was significant in girls and not in boys, and dexrazoxane has never been randomised in infants. Lipshultz SE et al. 2010
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In Interfant-06, 651 infants achieved 6-year event-free survival of 46.1% and overall survival of 58.2%, and the randomised comparison of AML-style myeloid consolidation (ADE + MAE) against the lymphoid course IB showed no difference in 6-year disease-free survival (39.3%, SE 4.0, n = 169 vs 36.8%, SE 3.9, n = 161; log-rank P = 0.47); outcome did not improve compared with Interfant-99.
Outcome of Infants Younger Than 1 Year With Acute Lymphoblastic Leukemia Treated With the Interfant-06 Protocol: Results From an International Phase III Randomized Study · Pieters R, De Lorenzo P, Ancliffe P, Aversa LA, Brethon B, Biondi A, Campbell M, Escherich G, Ferster A, Gardner RA, Kotecha RS, Lausen B, Li CK, Locatelli F, Attarbaschi A, Peters C, Rubnitz JE, Silverman LB, Stary J, Szczepanski T, Vora A, Schrappe M, Valsecchi MG · Journal of Clinical Oncology 2019What this evidence cannot show
The randomised comparison was powered for disease-free survival, not for cardiac endpoints, and the report contains no cardiac toxicity or cumulative anthracycline data. Only medium- and high-risk KMT2A-rearranged infants were randomised (330 of 651), so the null result applies to that subgroup. 46% of high-risk patients intended for transplant never received it because of early events, which complicates interpretation of the high-risk stratum. The absence of a DFS benefit does not by itself prove the extra anthracycline was harmful - it proves it was not beneficial for leukemia control.
Evidence that argues against this (1)
- The only therapy addition in recent years that appears to have moved outcomes in this population was immunotherapeutic rather than cytotoxic: adding one post-induction course of blinatumomab to the Interfant-06 backbone in 30 newly diagnosed infants produced no protocol-defined toxic effects and a 2-year disease-free survival of 81.6% (95% CI 60.8-92.0) against 49.4% (95% CI 42.5-56.0) in Interfant-06 historical controls, which is direct evidence that the anthracycline intensification route is not the only or best lever. van der Sluis IM et al. 2023
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a pooled analysis of 28,423 five-year survivors of childhood cancer from the Childhood Cancer Survivor Study, the DCOG-LATER study and the St Jude Lifetime study, with 399 cardiomyopathy cases over a median 20 years of follow-up, late-cardiomyopathy dose equivalence ratios relative to doxorubicin were 0.6 (95% CI 0.4-1.0) for daunorubicin, 0.8 (0.5-2.8) for epirubicin and 10.5 (95% CI 6.2-19.1) for mitoxantrone, meaning the conventional 4:1 haematologic equivalency substantially underestimates mitoxantrone's long-term cardiomyopathy risk.
Derivation of Anthracycline and Anthraquinone Equivalence Ratios to Doxorubicin for Late-Onset Cardiotoxicity · Feijen EAM, Leisenring WM, Stratton KL, Ness KK, van der Pal HJH, van Dalen EC, Armstrong GT, Aune GJ, Green DM, Hudson MM, Loonen J, Oeffinger KC, Robison LL, Yasui Y, Kremer LCM, Chow EJ · JAMA Oncology 2019What this evidence cannot show
Only 265 of 28,423 survivors received mitoxantrone, which is why the ratio's confidence interval spans 6.2 to 19.1 - a three-fold uncertainty that propagates directly into any doxorubicin-equivalent calculation for the Interfant-06 experimental arm. The authors note the mitoxantrone-doxorubicin relationship was better described by a linear-exponential than a linear model, so a single multiplicative ratio is an approximation. Cohorts were treated between 1962 and 2005 and the endpoint is severe, life-threatening or fatal cardiomyopathy by age 40, not subclinical dysfunction. Median age at cancer diagnosis was 6.1 years; infants are a small minority, so the ratios are not age-at-exposure-specific.
Evidence that argues against this (1)
- The Interfant-06 protocol itself asserts the opposite, stating in its drug guidance that 'Mitoxantrone is less cardiotoxic than daunorubicin' - the 2006 consensus position, based on haematologic rather than cardiac equivalence. The disagreement is real and is resolved in favour of the pooled survivor data, but it means that protocols written before 2019 were designed on the wrong ratio. source 2006
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Among 115 children in continuous remission a median of 6.4 years after doxorubicin therapy for ALL (18 receiving a single 45 mg/m2 dose, 97 receiving 228-550 mg/m2, median 360), 57% had abnormalities of left ventricular afterload or contractility; cumulative dose was the strongest predictor (P < 0.002), and in multivariate analysis restricted to those receiving >= 228 mg/m2 the only significant predictors were higher cumulative dose (P = 0.01) and age under four years at treatment (P = 0.003, predicting increased afterload); afterload increased progressively in 24 of 34 serially evaluated patients.
Late cardiac effects of doxorubicin therapy for acute lymphoblastic leukemia in childhood · Lipshultz SE, Colan SD, Gelber RD, Perez-Atayde AR, Sallan SE, Sanders SP · New England Journal of Medicine 1991What this evidence cannot show
Cross-sectional with a serial subgroup of only 34; single institution; treatment era 1976-1990 with doxorubicin doses (median 360 mg/m2) far above what any infant receives today. The age threshold is 'under four years', not 'under one year', so infancy specifically is not isolated. Endpoints are echocardiographic indices of afterload and contractility, which are surrogates for clinical heart failure and depend on the normative z-score model used. Reported symptoms correlated poorly with measured ventricular function, so the clinical meaning of the 57% figure is not a 57% rate of heart failure.
Evidence that argues against this (1)
- Longitudinal follow-up of the same population showed the trajectory is not monotonic: fractional-shortening z-scores transiently IMPROVED before falling to -2.76 beyond 12 years, so any single cross-sectional prevalence - including the 57% here - can substantially misrepresent risk depending on when the echocardiogram was done. Lipshultz SE et al. 2005
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a matched case-control study of 170 childhood cancer survivors with cardiomyopathy versus 317 without, cardiomyopathy odds ratios relative to no anthracycline exposure were 1.65 at 1-100 mg/m2, 3.85 at 101-150, 3.69 at 151-200, 7.23 at 201-250, 23.47 at 251-300 and 27.59 above 300 mg/m2 (P for trend < 0.001), demonstrating increased risk at doses as low as 101-150 mg/m2, with no apparently safe dose for survivors homozygous for the CBR3 V244M G allele.
Anthracycline-related cardiomyopathy after childhood cancer: role of polymorphisms in carbonyl reductase genes - a report from the Children's Oncology Group · Blanco JG, Sun CL, Landier W, Chen L, Esparza-Duran D, Leisenring W, Mays A, Friedman DL, Ginsberg JP, Hudson MM, Neglia JP, Oeffinger KC, Ritchey AK, Villaluna D, Relling MV, Bhatia S · Journal of Clinical Oncology 2012What this evidence cannot show
Case-control design with cases ascertained as survivors who already had cardiomyopathy, so odds ratios are not absolute risks and cannot be read as 'a 27% chance'. Matching was on diagnosis, year of diagnosis, follow-up length and race but not on age at exposure, which is the variable this node cares most about. Anthracycline exposure is expressed as a pooled cumulative dose across agents, so the bands do not distinguish daunorubicin from doxorubicin from mitoxantrone - which, given equivalence ratios spanning 0.6 to 10.5, is a material limitation when mapping Interfant-06 arms onto the curve. All cancer types pooled; no infant-specific stratum.
Evidence that argues against this (3)
- A much larger cohort study of 14,358 five-year survivors versus 3,899 siblings placed the anthracycline threshold higher, at >= 250 mg/m2, for a two- to five-fold increase in congestive heart failure, pericardial disease and valvular abnormality - so the claim of meaningful risk at 101-150 mg/m2 is not universally reproduced and may partly reflect case-control ascertainment. That cohort also found cumulative incidence still rising 30 years after diagnosis (CHF hazard ratio 5.9 vs siblings). Mulrooney DA et al. 2009
- Clinically ascertained lifetime burden puts cumulative cardiomyopathy prevalence by age 50 at 21.6% (95% CI 19.3-23.9) among at-risk survivors, and heart valve disorder at 83.5% - figures that are far higher than questionnaire-based estimates and that set the true scale of the surveillance obligation. Hudson MM et al. 2013
- Because asymptomatic cardiomyopathy precedes symptomatic heart failure by a variable interval, international guideline harmonisation was required to define who should be surveilled and how often - implying that neither a single dose threshold nor a single prevalence figure is adequate for clinical decision-making. Armenian SH et al. 2015
time-limited combination is motivated by lifetime toxicity in infants
names this drug in data.affects; late-cardiomyopathy equivalence ratio 0.6 vs doxorubicin
names this drug in data.affects; equivalence ratio 10.5 vs doxorubicin against the protocol's claim that it is less cardiotoxic than daunorubicin
names this drug in data.affects; doxorubicin is the reference agent for the equivalence ratios
- ← supportsNine survivors in ten carry a long-term complication, and it is worse the younger they were
4 of 67 survivors with cardiomyopathy at 150-225 mg/m2, two requiring heart transplant
THREE NAMED DRUGS, NO ENTRIES: daunorubicin, doxorubicin and mitoxantrone are named in data.affects and none is written.
the one late effect adult cost data partially capture
- ← analogous toThe same pocket suppresses JUND: menin inhibition's unmeasured endocrine liability in infants
same shape of problem: an acceptable acute profile in a population defined by how long it will live afterwards
a late effect that sets the bar any new therapy must beat
- ← causesInterfant-06 — 651 infants, and the definitive answer that myeloid-style consolidation does not help
ADE+MAE delivers daunorubicin plus mitoxantrone to infants for no survival benefit