Impact of Diagnosis-to-Treatment Interval on the Outcome of Patients with Acute Myeloid Leukemia

doi:10.21203/rs.3.rs-8147567/v1

Download PDF

Short Report

Impact of Diagnosis-to-Treatment Interval on the Outcome of Patients with Acute Myeloid Leukemia

https://doi.org/10.21203/rs.3.rs-8147567/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background:Acute myeloid leukemia (AML) is considered an oncologic emergency, yet the optimal timing for treatment initiation remains uncertain.

Methods: We conducted a multi-institutional retrospective study of 698 adults with newly diagnosed AML presenting to four academic centers across the United States. Diagnosis-to-treatment intervals (DTI) were categorized as <5 days, 5–10 days, and >10 days. Outcomes were analyzed using multivariable models adjusting for age, treatment intensity, ELN 2017 risk classification, and white blood cell count.

Results: Among younger patients, DTI was not associated with differences in survival outcomes. In contrast, older patients demonstrated improved survival with delayed treatment (DTI >10 days), particularly those with lower white blood cell counts. No adverse effects from treatment delay were observed in younger cohorts.

Conclusions: This retrospective study has showed that prolonged DTI is associated with improved survival in older adults with newly diagnosed AML, challenging the traditional assumption that immediate therapy universally improves outcomes. These findings underscore the importance of individualized treatment timing in the era of precision oncology.

Acute myeloid leukemia

diagnosis-to-treatment interval

induction chemotherapy

molecular testing

risk stratification

targeted therapy

precision medicine

treatment outcomes

The diagnosis of acute myeloid leukemia (AML) is an oncologic emergency that often necessitates prompt initiation of therapy. The optimal timing of therapy initiation has long been debated, and current guidelines do not provide specific recommendations regarding the diagnosis-to-treatment interval (DTI) in patients with AML.

Historically, prompt initiation of induction chemotherapy was considered crucial to mitigate disease-related morbidity and mortality, particularly among younger patients with proliferative AML.¹ However, with the 2017 and 2022 revisions of the World Health Organization (WHO) and International Consensus Classification (ICC) systems, molecular characterization has become increasingly integral to AML diagnosis and risk stratification.^2,3 Moreover, the expanding molecular landscape of AML also informs therapeutic decision-making and disease monitoring.⁴ Over the past decade, the introduction of targeted therapies for actionable mutations has transformed AML management, underscoring the growing importance of genomic profiling in guiding treatment decisions and improving clinical outcomes.^4,5,6,7,8

The BEAT AML Master clinical trial, the first prospective precision medicine trial in hematologic malignancies, demonstrated superior outcomes when patients with actionable mutations received targeted therapy rather than standard of care. In addition, delaying therapy to await molecular testing results was found to be safe in patients without evidence of hyperleukocytosis.⁹

In routine practice, however, the turnaround time for comprehensive molecular profiling can be prolonged, typically requiring 7 to 14 days for completion. Although enrollment in clinical trials may expedite access to results and targeted agents, most patients are treated in real-world settings with standard of care regimens.¹⁰

To address the clinical uncertainty surrounding the timing of therapy initiation, we conducted a multi-institutional retrospective study across four U.S. academic centers to address the impact of DTI on outcomes in newly diagnosed AML. Our study aimed to define the optimal treatment window in the contemporary era of precision oncology, where molecular testing increasingly guides therapeutic selection and clinical decision-making.

We retrospectively collected data on 698 adult patients (aged ≥ 18 years) with newly diagnosed AML treated at four academic cancer centers between 2010 to 2021: the GW Cancer Center, Georgetown University Hospital, University of Virginia, and University of Washington/Fred Hutchinson Cancer Center. Patients with acute promyelocytic leukemia and those who elected to receive treatment at other institutions were excluded. Baseline demographic and clinical characteristics were obtained through detailed chart review and included age, sex, race, smoking status, comorbidities, and disease characteristics such as presenting white blood cell (WBC) count, bone marrow morphology, cytogenetic and fluorescence in situ hybridization (FISH) results, and next-generation sequencing (NGS) findings when available. Treatment-related variables included regimen intensity (low vs. high) and time from diagnosis to therapy initiation. Low-intensity therapy was defined as treatment with a hypomethylating agent (HMA) alone, HMA combined with venetoclax, or low-dose cytarabine (LoDAC) with or without venetoclax. High-intensity therapy was defined as conventional intensive induction chemotherapy (i.e., “7 + 3” or a high-dose cytarabine containing regimen). Institutional review board approval and data-sharing agreements were obtained at all participating sites.

The primary outcome was overall survival (OS), defined as the time from AML diagnosis to death from any cause or last follow-up for censored patients. Patients were categorized into three DTI groups: less than five days (< 1–5), six to ten days (6–10), and more than ten days (> 10).

A landmark analysis was performed to evaluate the association between DTI category and OS and to mitigate immortal time bias. Patients who died or were censored within the first 10 days of diagnosis were excluded. For the remaining cohort, survival time was calculated beginning on day 10 post-diagnosis (landmark time).

Cox proportional hazards regression models were used to evaluate the association between DTI and OS. The < 1–5 day category served as the reference group. Multivariable models were adjusted for clinically relevant covariates, including age (≤ 65 vs. >65 years), baseline WBC count (≤ 100 × 10⁹/L vs. >100 × 10⁹/L), 2017 European Leukemia Net (ELN) risk category (favorable, intermediate, or adverse), and treatment intensity (high vs. low). Interaction terms between DTI and both age and WBC categories were included in to assess whether the effect of DTI on OS differed across subgroups.

Hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) were estimated for all comparisons. Linear contrasts of model coefficients were constructed to derive HRs and 95% CIs for specific subgroup analyses (DTI effects within age or WBC strata and pairwise comparisons between non-reference DTI categories. Confidence intervals were calculated using the Delta method, with log-transformed limits subsequently exponentiated. The proportional hazards assumption was verified by examining the Schoenfeld residuals and associated plots, with no significant violations observed. For analyses involving multiple contrasts, p-values were adjusted using the single-step method based on the multivariate normal distribution of contrasts to control the family-wise error rate. All analyses were performed using the R Statistical Software (version v4.4.2; R Core Team, 2024). Survival analyses were conducted using the survival package, and multiple comparison adjustments were implemented using the multcomp package. A two-sided p-value < 0.05 was considered statistically significant.

A total of 698 patients across four academic centers were included in the analysis assessing the impact of the DTI on outcomes in AML. Patients were categorized into three DTI groups: <1–5 days (n = 324), 6–10 days (n = 117), and > 10 days (n = 257). Baseline characteristics are summarized in Table 1. Patients with delayed therapy (> 10 days) were older (median 66.0 years, interquartile range [IQR] 57–74) compared with those treated within 1–5 days (median 60.5 years, IQR 49–70) and presented with lower white blood cell (WBC) counts (mean 15.0 vs. 52.9 ×10³/µL). Elevated WBC counts (> 100 ×10³/µL) were most common among those treated within five days (19.1%). Adverse ELN risk was more frequent in the > 10-day group (46.7%) than in the < 1–5-day group (38.3%, Fisher's exact test, p = 0.005). High-intensity induction therapy was more often administered to patients with early DTI (77.5% for < 1–5 days vs 57.2% for > 10 days; Fisher's exact test, p < 0.001).

Kaplan–Meier survival curves demonstrated substantial overlap across DTI categories, with no significant difference by log-rank test (p = 0.982; WBC ≤ 100 × 10³/µL subgroup p = 0.949) (Fig. 1). Similarly, unadjusted Cox analyses revealed no significant association between DTI and OS, with HRs near 1 for both the 6–10 day and > 10-day groups relative to the < 1–5 day reference.

After multivariable adjustment for age, WBC count, ELN risk, and treatment intensity (Table 2), a prolonged DTI (> 10 days) was associated with improved survival, corresponding to a 32% reduction in mortality risk compared with early treatment (HR 0.68, 95% CI: 0.53–0.87, p = 0.003). Subgroup analyses (Table 3) revelated that this benefit was most pronounced among older patients (> 65 years) with WBC ≤ 100 ×10³/µL, in whom a DTI > 10 days conferred a 38% lower mortality risk compared with treatment within five days (HR 0.62, 95% CI: 0.39–0.99, adjusted p = 0.040). Within this same subgroup, outcomes were also superior for patients treated after > 10 days compared with those treated within 6–10 days (HR 0.53, 95% CI: 0.30–0.95, adjusted p = 0.024). No significant survival differences by DTI were observed among younger patients (≤ 65 years) or among those with WBC > 100 ×10³/µL after adjustment for covariates.

Independent of DTI, several factors were significantly associated with survival. Adverse predictors included older age (per year; HR 1.03, 95% CI 1.02–1.04), higher WBC count (per 10³/µL; HR 1.002, 95% CI 1.001–1.004), intermediate ELN risk (HR 1.70, 95% CI 1.22–2.37), adverse ELN risk (HR 2.12, 95% CI 1.54–2.94), and low-intensity therapy vs high-intensity (HR 1.64, 95% CI 1.27–2.13) (Table 2). These associations were similar in interaction models, with older age (> 65 vs ≤ 65) (HR 2.14, 95% CI 1.54–2.98), intermediate ELN risk (HR 1.76, 95% CI 1.26–2.46), adverse ELN risk (HR 2.20, 95% CI 1.59–3.05), and low-intensity therapy (HR 1.81, 95% CI 1.41–2.31) continuing to predict inferior outcomes (Table 4). The only factor independently associated with improved OS was a prolonged DTI > 10 days, an effect restricted to older patients with lower WBC counts, as noted above.

The proportional hazards assumption was assessed with Schoenfeld residuals. Although the global test approached significance (p = 0.049), the assumption held for the DTI variable (p = 0.332), corroborated by residual plots showing no systematic deviation (Fig. 2).

Our findings reveal an unexpected association between the DTI and outcomes in older patients with AML. Specifically, prolonged DTI was associated with reduced mortality in older patients, whereas no survival benefit was observed in younger patients. Even among young patients with elevated WBC counts, treatment delay did not significantly influence outcomes. This observation differentiates our analysis from prior large registry studies, such as the German Study Alliance Leukemia–Acute Myeloid Leukemia (SAL-AML) and TriNetX registry or reports by Röllig et al. and Bertolli et al, which found no significant impact of DTI on survival.^11,12,13

Another large retrospective study of 1,317 patients from the Cleveland Clinic and the MD Anderson Cancer Center reported lower complete response (CR) and OS rates among young patients when DTI exceeded five days, while outcomes among older patients remained unaffected.⁶ Similarly, an ECOG study of 362 patients demonstrated lower CR rates with delayed therapy initiation, although this did not translate into an OS difference.¹³

Evaluating the effect of DTI is inherently complex due to selection bias inherent to retrospective study designs. To mitigate this bias, we stratified patients by WBC count. Interestingly, even in patients with hyperleukocytosis, prolonged DTI did not adversely influence outcomes. Other potential confounders, such as disseminated intravascular coagulopathy (DIC), active infection/ or sepsis, and transfusion requirements, may also have contributed to differences in early mortality and treatment tolerance.

To our knowledge, this is the first study to demonstrate a potential adverse impact of shorter DTI in older patients with AML. Several explanations may account for this finding. AML is a biologically and clinically heterogeneous disease, and older patients who received therapy earlier may have presented with greater clinical instability or more aggressive disease features, prompting expedited treatment initiation. Such patients may also have been less physiologically fit, predisposing them to treatment-related complications and early mortality. Our dataset did not capture clinical parameters such as infection or DIC, which limits our ability to test this hypothesis directly. Furthermore, our cohort largely predated the widespread use of venetoclax-based regimens, which may limit applicability to contemporary treatment paradigms.¹⁴

Given these considerations, our results should be regarded as hypothesis-generating rather than definitive evidence of causality. Notably, prior studies examining DTI in AML have employed heterogeneous definitions of “delay”, with intervals varying widely across analyses. Some modeled DTI as a continuous variable, while others used categorical thresholds, as in our study. A prior meta-analysis suggested that median DTI across studies typically ranged from four to eight days.¹⁵

Our study has several limitations. First, its retrospective design introduces potential selection bias and confounding by indication, as patients perceived to be more acutely ill were likely prioritized for earlier treatment. Second, the landmark analysis, while essential to minimize immortal time bias, inherently excluded patients who died within 10 days of diagnosis, thereby limiting generalizability to early survivors. Third, data granularity was constrained: precise treatment initiation dates were unavailable, necessitating calculating of OS from the date of diagnosis rather than from the start of therapy. This approach, although standard, reflects the full disease trajectory rather than isolating post-treatment survival.

Conclusion In this multi-institutional retrospective study, a prolonged diagnosis-to-treatment interval was associated with improved survival among older patients with newly diagnosed AML, particularly those with lower white blood cell counts, while no adverse effect of treatment delay was observed in younger patients. These findings challenge the traditional assumption that immediate initiation of therapy universally improves outcomes and underscore the importance of individualized treatment timing in the era of precision oncology. Prospective studies are warranted to validate these results and to determine whether a brief delay to obtain comprehensive molecular data and optimize therapeutic selection may improve outcomes in clinically stable patients.

Acknowledgments: The authors used an artificial intelligence-based tool for scholarly writing, specifically “Paperpal,” to assist with refining the English language and improving writing quality. No AI tools were used to summarize the content or any other form of writing. The authors take full responsibility for the content of this manuscript.

There was no funding associated with this paper.

Conflict of Interest:

FEC reports Consultant: SPD Oncology, Amgen, CTI BioPharma, AbbVie, MorphoSys, PharmaEssentia, BMS, Geron, Sobi, DAVA Oncology, Taiho Oncology, Daiichi Sankyo, Syndax, Novartis, Merck.

CL does consulting for AbbVie, Astellas, Arcellx, Syndax, Kura, Rigel, Stemline, Geron, Daiichi, and BMS, and has research funding from Jazz Pharma and BMS.

MEP has research funding to her institution from Abbvie, Ascentage, Astex, Oncoverity, and Pfizer

Authors contributions:

SN, FEC, IT, MEP and CL contributed to study conception and design. Material preparation, data collection and analysis were performed by Samah Nassereddine, Vanyal Aggarwal, Kimberly Doucette, Lacy Williams, Leah Wells, Ramy Elsarrag, Jordan Selep, Yuan Feng, and Shanshan Liu, Guoquing Diao. The first draft of the manuscript was written by Samah Nassereddine and all authors commented on previous versions of the manuscript.

Ethics and Consent to Participate declarations: not applicable

Sekeres MA, Elson P, Kalaycio ME et al (2009) Time from diagnosis to treatment initiation predicts survival in younger, but not older, acute myeloid leukemia patients. Blood 113(1):2836. 10.1182/blood200805157065
Arber DA, Orazi A, Hasserjian R et al (2022) International Consensus Classification of Myeloid Neoplasms and Acute Leukemia: integrating morphologic, clinical, and genomic data. Blood 140(11):1200–1228. 10.1182/blood.2022012382
Khoury JD, Solary E, Abla O et al (2022) The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia 36(7):1703–1719. 10.1038/s41375-022-01613-1
Döhner H, Wei AH, Appelbaum FR et al (2022) Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood 140(12):1345–1377. 10.1182/blood.2022016867
Stone RM, Mandrekar SJ, Sanford BL et al (2017) Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med 377(5):454–464. 10.1056/NEJMoa1614359
Erba HP, Montesinos P, Vrhovac R et al (2023) Quizartinib versus placebo in combination with chemotherapy in patients with FLT3-ITD–positive acute myeloid leukaemia (QuANTUM-First): a double-blind, randomised, placebo-controlled, phase 3 trial. Lancet 401(10389):1873–1885. 10.1016/S0140-6736(23)00441-4
DiNardo CD, Jonas BA, Pullarkat V et al (2020) Azacitidine and venetoclax in previously untreated acute myeloid leukemia. N Engl J Med 383(7):617–629. 10.1056/NEJMoa2012971
Montesinos P, Recher C, Vives S et al (2022) Ivosidenib and azacitidine in IDH1-mutated acute myeloid leukemia. N Engl J Med Apr 21(16):1519–1531. 10.1056/NEJMoa2117344
Burd A, Levine RL, Ruppert AS et al (2020) Precision medicine treatment in acute myeloid leukemia using prospective genomic profiling: feasibility and preliminary efficacy of the Beat AML Master Trial. Nat Med 26(12):1852–1858. 10.1038/s41591-020-1089-8
Franco S, Geng X, Korostyshevskiy V, Lai C, Karp JE (2023) Prognostic impact of time from diagnosis to treatment in patients with acute myeloid leukemia: systematic review and metaanalysis. Cancer 129(19):29752985. 10.1002/cncr.34894
Baden D, Zukunft S, Hernandez G et al (2024) Time from diagnosis to treatment has no impact on survival in newly diagnosed acute myeloid leukemia treated with venetoclax-based regimens. Haematologica 109(8):2469–2477. 10.3324/haematol.2024.285225
Röllig C, Kramer M, Schliemann C et al (2020) Does time from diagnosis to treatment affect the prognosis of patients with newly diagnosed acute myeloid leukemia? Blood 136(7):823–830. 10.1182/blood.2019004583
Bertoli S, Bérard E, Huguet F et al (2013) Time from diagnosis to intensive chemotherapy initiation does not adversely impact the outcome of patients with acute myeloid leukemia. Blood 121(14):2618–2626. 10.1182/blood-2012-09-454553
Rowe JM, Bennett JM, Fielding AK et al (2020) Acute myeloid leukaemia in adult patients: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 31(6):697–710. 10.1016/j.annonc.2020.03.283
Walter RB, Othus M, Borthakur G et al (2011) Prediction of early death after induction therapy for newly diagnosed acute myeloid leukemia with pretreatment risk scores: a novel paradigm for treatment assignment. J Clin Oncol 29(33):4417–4423. 10.1200/JCO.2011.35.7525

Tables 1 to 4 are available in the Supplementary Files section.

No competing interests reported.

Download PDF

Reviews received at journal
04 Jan, 2026
Reviews received at journal
01 Jan, 2026
Reviewers agreed at journal
09 Dec, 2025
Reviewers agreed at journal
06 Dec, 2025
Reviewers invited by journal
04 Dec, 2025
Editor assigned by journal
27 Nov, 2025
Submission checks completed at journal
27 Nov, 2025
First submitted to journal
18 Nov, 2025

You are reading this latest preprint version

Impact of Diagnosis-to-Treatment Interval on the Outcome of Patients with Acute Myeloid Leukemia

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1