AML, a hematological malignancy characterized by clonal proliferation of myeloblasts, is the most common acute leukemia in adults [1]. Although advances in therapy have improved survival, AML continues to be difficult to treat and is associated with high mortality. In a phase III Southwest Oncology Group (SWOG/ECOG) intergroup study (n = 609) that included patients who received induction chemotherapy, 5 year survival rates for those with favorable, intermediate, and adverse risk cytogenetics were 55 %, 38 %, and 11 %, respectively [2]. Induction therapy is also associated with significant toxicity and complications, which contributes directly to morbidity and mortality [3].
During standard dose cytarabine induction, a day 14 nadir bone marrow biopsy (NBMB) is used as the first measure of response [4]. An early morphologic response, with blast clearance observed on the NBMB, has been cited as a major predictor for CR and OS [5]. Obtaining a CR, whether after one or two cycles of induction, is important for long-term survival [6]. Data is lacking to quantify the sensitivity and specificity of the NBMB in determining the presence of significant residual disease (RD) [7]. There are patients with NBMB findings suspicious for residual disease (sRD) and it is difficult to determine in which cases re-induction is beneficial. Additional re-induction (RI) chemotherapy in patients with residual disease has been associated significant morbidity and mortality [8,9]. With lack of efficacy data and significant toxicity associated with RI, the primary purpose of this retrospective study is to determine the utility of RI in patients with sRD.
Our study included patients with newly diagnosed of AML who are ≥18 years of age, treated with standard dose cytarabine induction chemotherapy (ICT), with a nadir bone marrow biopsy. Two hundred seventy adult patients were identified that met those criteria at Wake Forest Baptist Medical Center between January 2002 and December 2009. Patients included in the study received ICT with the backbone of cytarabine and anthracycline given in the standard 7 + 3 format. Patients were excluded if they did not have a 14-day (± 3 days) NBMB (n = 28) or if their care was transferred to another hospital prior to obtaining a recovery BM biopsy (n = 2) or if an alternative induction chemotherapy regimen was administered (n = 7). Patients transferred to hospice following a NBMB but before a recovery marrow were included and were classified as failing to achieve a CR. Of the 270 patients screened, 233 patients were included in the analysis. The primary objective of the study was to compare complete response rates between patients who received re-induction (RI) and patients who did not. Secondary objective of the study was to determine overall survival and adjusted overall survival rate differences between patients who received RI and patients who did not.
NBMB were classified as negative (<20 % cellularity and <5 % blasts); suspicious for residual disease, (sRD, ≥5 % and ≤20 % blasts (sRDa), or <5 % blasts but with morphological suspicion for residual disease as determined by the hematopathologist (sRDb); and positive (>20 % blasts). CR was defined as recovery BM with <5 % blasts, platelets ≥ 100,000/μL, and ANC ≥ 1000/μL. CRi was achieved if the BM had < 5 % blasts but there was incomplete count recovery of platelets (i.e. < 100,000/μL) or neutrophils (i.e. < 1000/μL). Patients who expired after a nadir BM biopsy was obtained but prior to obtaining a recovery BM biopsy were classified as having failed to achieve a CR. Of note, no patient in this analysis received myeloid growth factors.
Of 233 patients, 90 (39 %) had a negative NBMB, 37 (15 %) had a positive NBMB and 106 (46 %) had NBMB findings of sRD. Of those with sRD, 66 (62.3 %) underwent re-induction and 40 (37.7 %) did not undergo re-induction. Demographics of patients with NBMB findings of sRD are shown in Table 1.
Table 1.
Demographics of patients with nadir bone marrow findings suspicious for residual disease.
| sRD (n = 106) | RI (n = 66) | No RI (n = 40) | p-value* |
|---|---|---|---|
| Age | 0.42 | ||
| <60 (44.3 %) | 27 (40.9 %) | 20 (50.0 %) | |
| Median Age | 45 | 46.5 | 1.00 |
| ≥60 (55.7 %) | 39 (59.1 %) | 20 (50.0 %) | |
| Median Age | 67 | 75 | 0.0013 |
| Gender | 0.43 | ||
| Male (57.5 %) | 40 (60.6 %) | 21 (52.5 %) | |
| Female (42.5 %) | 26 (39.4 %) | 19 (47.5 %) | |
| Cytogenetic risk ** | 0.86 | ||
| Favorable (16 %) | 11 (16.7 %) | 6 (15.0 %) | |
| Intermediate (71.7 %) | 46 (69.7 %) | 30 (75.0 %) | |
| Poor (12.3 %) | 9 (13.6 %) | 4 (10.0 %) | |
| NBMB Finding | |||
| sRDa | 29 | 26 | |
| sRDb | 37 | 14 | |
| Median Blast Percentage | |||
| Overall | 5 | 2.5 | |
| sRDa | 1 | 0 | |
| sRDb | 10.5 | 9 | |
| Median Cellularity | |||
| Overall | 12.5 | 10 | |
| sRDa | 12.5 | 10 | |
| sRDb | 10 | 10 |
p-value compares < 60 vs ≥60, p-values are from Fisher’s exact test (age (<60 vs ≥60, gender, cytogenetic risk) or Kruskal-Wallis test (median age), p value of <0.05 was considered significant.
As per SWOG2.
Amongst patients in the sRD group, 52 of 66 of people (78.8 %) who underwent re-induction achieved CR/CRi and 32 of 40 (80.0 %) of patients who did not undergo re-induction achieved CR/CRi. Induction death rate for sRD patients who did receive re-induction was 16.7 % (11 of 66 patients) and 17.5 % (7 of 40) for sRD patients who did not receive re-induction. Not unexpectedly, patients with a positive NBMB had a CR/CRi rate of 59 % if they received re-induction treatment (20 out of 34 patients) or 0 % if no re-induction chemotherapy was given (0 out of 3 patients). Of the patients with a negative NBMB, 89 % achieved CR/CRi. Interestingly, median survival rates and survival estimates at 1, 2, and 3 years of patients with NBMB findings of sRD were not statistically different (p = 0.7484, Fig. 1A) in patients who did and did not receive re-induction. Median survival was 16.9 months in patients who did and 20.1 months in patients who did not receive re-induction in the sRD group. This was not statistically significant in subgroup analysis of these patients with favorable (p = 0.4415, Fig. 1B), intermediate (p = 0.5018, Fig. 1C) or poor (p = 0.4839, Fig. 1D) cytogenetics (see Table 2) by ELN2017 criteria [14]. This remained true for all patients with sRD even after adjusting for age (continuous), gender (M/F), cytogenetics (favorable/intermediate/poor), comorbidities (yes/no), transplant (yes/no) with Hazard ratio of 1.05 (RI vs no RI) and a 95 % CI (0.64, 1.70); p = 0.86 (Fig. 2). Cytogenetic risk was the only significant predictor in the adjusted model (p = 0.0005).
Fig. 1.
Kaplan-Meier plot of overall survival in the suspicious group. Median survival rates and survival estimates at 1, 2, and 3 years of patients with NBMB findings of sRD were not statistically different (p = 0.7484, Fig. 1A) in patients who did and did not receive re-induction. This was not statistically significant in subgroup analysis of these patients with favorable (p = 0.4415, Fig. 1B), intermediate (p = 0.5018, Fig. 1C) or poor (p = 0.4839, Fig. 1D) cytogenetics.
Table 2.
Median survival and survival estimates at 1, 2, and 3 years of patients with nadir bone marrow findings suspicious for residual disease.
| Suspicious Group (sRD) | N | Deaths | Median Survival (months) |
KM Survival Estimate (%) at: |
p-value* | |||
|---|---|---|---|---|---|---|---|---|
| Estimate | 95 % CI | 1 year | 2 year | 3 year | ||||
| Overall | 106 | |||||||
| No re-induction | 40 | 30 | 20.1 | 9.7, 27.0 | 62.5 | 40.0 | 32.5 | 0.7484 |
| Re-induction | 66 | 51 | 16.9 | 9.5, 23.8 | 57.6 | 37.9 | 30.3 | |
| Cytogenetic Groups | ||||||||
| Favorable | ||||||||
| No re-induction | 6 | 3 | NA | NA | 100 | 83.3 | 66.7 | 0.4415 |
| Re-induction | 11 | 3 | NA | NA | 90.9 | 90.9 | 90.9 | |
| Intermediate | ||||||||
| No re-induction | 30 | 23 | 20.1 | 8.6, 27.0 | 60.0 | 36.7 | 30.0 | 0.5018 |
| Re-induction | 46 | 39 | 16.9 | 8.8, 23.1 | 60.9 | 32.6 | 21.7 | |
| Poor | ||||||||
| No re-induction | 4 | 4 | 5.2 | 0.7, 17.8 | 25.0 | 0 | 0 | 0.4839 |
| Re-induction | 9 | 9 | 6.4 | 1.5, 9.5 | 0 | 0 | 0 | |
p-values are from Fisher’s exact test, a value of <0.05 was considered significant.
Fig. 2.
Estimated survival based on proportional hazards Cox model. Median survival rates and survival estimates at 1, 2, and 3 years of patients with NBMB findings of sRD were not statistically different even after adjusting for age (continuous), gender (M/F), cytogenetics (favorable/intermediate/poor), comorbidities (yes/no), transplant (yes/no) with Hazard ratio of 1.05 (RI vs no RI) and a 95 % CI (0.64, 1.70); p = 0.86. Cytogenetic risk was the only significant predictor in the adjusted model (p = 0.0005).
In terms of the primary objective, our retrospective analysis of 233 patients indicated improved rates of CR/CRi with early blast clearance. This is consistent with previous studies [5,11,13]. However, while the nadir BM biopsy was sensitive (92.0 %) in predicting a CR/CRi in our study, it was much less specific (38.1 %) in predicting which patients would fail to obtain a CR/CRi without early re-induction chemotherapy. As for the secondary objective, the overall survival rates of patients with NBMB suspicious for residual disease, there was no statistical difference in OS or survival at 1, 2, and 3 years for patients who did or did not receive re-induction chemotherapy. These data show that the NBMB is useful in identifying patients with 20 % or more blasts who require additional therapy but suggest a threshold of 5 % blasts for consideration of RI therapy is too stringent. Our study raises questions on the utility of re-induction in adult patients with AML with ≤20 % blasts on NBMB. This finding warrants prospective trials randomizing patients with less than 20 % but greater than 5 % blasts on NBMB to RI or observation. Such a trial would better define the patient population who benefits from re-induction therapy.
Acknowledgements
This work was supported by the Frances P. Tutwiler Fund, the Doug Coley Foundation for Leukemia Research, The McKay Cancer Research Foundation, and the National Institute of Health (TSP is supported by NCI 1R01CA197991-01A1, SI is supported by NCI P30CA012197). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.
Footnotes
Ethics approval, consent to participate and consent for publication
Patient consent was not obtained as this is a retrospective review.
Availability of data and materials
The single institute date from this study is available from the corresponding author upon reasonable request.
Declaration of Competing Interest
The authors declare no competing financial interests.
Contributor Information
Kavya K. Kannan, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Paz Vellanki, University of Maryland Greenebaum Comprehensive Cancer Center, Baltimore, MD, USA.
Scott Isom, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA; Department of Biostatistics and Data Science, Wake Forest Baptist Health, USA.
Bernard Tawfik, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Allison Winter, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Heidi D. Klepin, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA
Leslie R. Ellis, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA
Rupali Roy Bhave, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Dianna Howard, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Megan Manuel, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Sarah Dralle, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Susan Lyerly, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
Bayard L. Powell, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA
Timothy S. Pardee, Department of Internal Medicine, Section on Hematology and Oncology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA; Department of Cancer Biology, Comprehensive Cancer Center of Wake Forest Baptist Health, USA.
References
- [1].Deschler B, Lubbert M, Acute myeloid leukemia: epidemiology and etiology, Cancer 107 (9) (2006) 2099–2107. [DOI] [PubMed] [Google Scholar]
- [2].Slovak ML, Kopecky KJ, Cassileth PA, et al. , Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study, Blood 96 (13) (2000) 4075–4083. [PubMed] [Google Scholar]
- [3].Schellongowski P, Staudinger T, Kundi M, et al. , Prognostic factors for intensive care unit admission, intensive care outcome, and post-intensive care survival in patients with de novo acute myeloid leukemia: a single center experience, Haematologica 96 (2) (2011) 231–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Mattison RJ, Luger SM, Lazarus HM, New strategies for the evaluation of the nadir bone marrow following induction in acute myeloid leukemia, Curr. Opin. Hematol 20 (2) (2013) 93–99. [DOI] [PubMed] [Google Scholar]
- [5].Kern W, Haferlach T, Schoch C, et al. , Early blast clearance by remission induction therapy is a major independent prognostic factor for both achievement of complete remission and long-term outcome in acute myeloid leukemia: data from the German AML Cooperative Group (AMLCG) 1992 Trial, Blood 101 (1) (2003) 64–70. [DOI] [PubMed] [Google Scholar]
- [6].Rowe JM, Kim HT, Cassileth PA, et al. , Adult patients with acute myeloid leukemia who achieve complete remission after 1 or 2 cycles of induction have a similar prognosis: a report on 1980 patients registered to 6 studies conducted by the Eastern Cooperative Oncology Group, Cancer 116 (21) (2010) 5012–5021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Morris TA, DeCastro CM, Diehl LF, et al. , Re-induction therapy decisions based on day 14 bone marrow biopsy in acute myeloid leukemia, Leuk. Res 37 (1) (2013) 28–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Cheson BD, Bennett JM, Kopecky KJ, et al. , Revised recommendations of the international working group for diagnosis, standardization of response criteria, treatment outcomes, and reporting standards for therapeutic trials in acute myeloid leukemia, J. Clin. Oncol 21 (24) (2003) 4642–4649. [DOI] [PubMed] [Google Scholar]
- [9].Hussein K, Jahagirdar B, Gupta P, Burns L, Larsen K, Weisdorf D, Day 14 bone marrow biopsy in predicting complete remission and survival in acute myeloid leukemia, Am. J. Hematol 83 (6) (2008) 446–450. [DOI] [PubMed] [Google Scholar]
- [11].Yanada M, Borthakur G, Ravandi F, Bueso-Ramos C, Kantarjian H, Estey E, Kinetics of bone marrow blasts during induction and achievement of complete remission in acute myeloid leukemia, Haematologica 93 (8) (2008) 1263–1265. [DOI] [PubMed] [Google Scholar]
- [13].Xiao Z, Xue H, Li R, Zhang L, Yu M, Hao Y, The prognostic significance of leukemic cells clearance kinetics evaluation during the initial course of induction therapy with HAD (homoharringtonine, cytosine arabinoside, daunorubicin) in patients with de novo acute myeloid leukemia, Am. J. Hematol 83 (3) (2008) 203–205. [DOI] [PubMed] [Google Scholar]
- [14].Dohner H, Estey E, Grimwade D, et al. , Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel, Blood 129 (4) (2017) 424–447. [DOI] [PMC free article] [PubMed] [Google Scholar]