Therapy-related acute myelogenous leukemia and myelodysplastic syndrome in patients with acute lymphoblastic leukemia treated with the hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone regimens

Authors


Abstract

BACKGROUND:

Secondary malignancies including myeloid neoplasms occur infrequently in acute lymphoblastic leukemia (ALL) and to the authors' knowledge have not been as well documented in adults as in children.

METHODS:

A total of 641 patients with de novo ALL who were treated with the hyper-CVAD (hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone) regimen or its variants were analyzed.

RESULTS:

Sixteen patients (2.49%) developed secondary acute myelogenous leukemia (AML) (6 patients) or myelodysplastic syndrome (MDS) (10 patients). At the time of ALL diagnosis, the median age was 53 years; cytogenetics were normal in 11 patients, pseudo-diploidy with del(2) in 1 patient, t(9;22) in 1 patient, and unavailable in 3 patients. Frontline therapy included hyper-CVAD in 7 patients, hyper-CVAD with rituximab in 8 patients, and hyper-CVAD with imatinib in 1 patient. Karyotype at time of AML/MDS diagnosis was −5, −7 in 9 patients, normal in 1 patient, complex in 1 patient, inv(11) in 1 patient, t(4;11) in 1 patient, del(20) in 1 patient, and unavailable in 2 patients. Secondary AML/MDS developed at a median of 32 months after ALL diagnosis. Cytarabine plus anthracycline–based treatment was given to 12 patients with AML and high-risk MDS. One patient with MDS received arsenic trioxide, 1 received clofarabine, and 2 received decitabine. Response to treatment was complete remission in 3 patients, partial remission in 6 patients, and no response in 6 patients; 1 patient was untreated. Eight patients (1 with AML and 7 with MDS) underwent allogeneic stem cell transplantation, and all but 2 died at a median of 3 months (range, 0.5–11 months) after transplantation. The median overall survival after a diagnosis of secondary AML and MDS was 9.25 months (range, 1+ to 26+ months).

CONCLUSIONS:

Secondary AML and MDS occur infrequently in adult patients with de novo ALL treated with the hyper-CVAD regimens, and response to therapy is poor. Cancer 2009. © 2008 American Cancer Society.

Intensive combination chemotherapy regimens for acute lymphoblastic leukemia (ALL) have greatly improved the outcome of younger patients, with long-term survival rates of > 80% reported in children.1 However, adults with ALL have 5-year survival rates ranging from 30% to 60%, depending on the risk features present at the time of diagnosis. Long-term survivors of cancer are at risk of developing secondary malignancies because of the effects of radiotherapy and/or chemotherapy, and potentially because of an underlying genetic predisposition to malignancies. The cumulative risk of secondary malignancies after successful treatment of ALL has been estimated to be 6.27% at 30 years of follow-up for pediatric patients.2-8 In 1 report in adult patients, the risk was 9.4% at 15 years.9 The risk of secondary hematologic malignancy was 2.36% at 30 years in children and 3.3% at 18 years in adults.2, 9 Secondary (or therapy-related) acute myelogenous leukemia (AML) and myelodysplastic syndrome (MDS) have been reported after chemotherapy with alkylating agents and/or topoisomerase inhibitors for the treatment of Hodgkin disease, non-Hodgkin lymphoma, breast cancer, ovarian carcinoma, or testicular carcinoma.10-13

The hyper-CVAD regimen (hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with high-dose methotrexate and cytarabine followed by maintenance therapy with mercaptopurine, methotrexate, vincristine, and prednisone) is an effective frontline therapy for adults with de novo ALL, with a median survival of 32 months and a 5-year survival rate of 38%.14 As the survival rates for adult patients with ALL treated with various therapeutic regimens improve, we may encounter long-term consequences of therapy. One study of 125 patients treated with the hyper-CVAD regimen reported 2 patients (3.2%) who developed MDS/AML at 4 months, 21 months, 24 months, and 37 months, respectively, after therapy.15 In another report of 1494 adult patients with ALL treated with daunorubicin, rubidazone, idarubicin, vincristine, mitoxantrone, cytarabine, L-asparaginase, cyclophosphamide, methotrexate, and doxorubicin, 23 patients (1.5%) developed secondary neoplasms, 6 patients had AML, and 3 had MDS.9 In these patients, AML and MDS developed at a median of 1.3 and 2.5 years, respectively, after diagnosis of ALL.

We analyzed the medical records of adult patients with ALL who were treated with the hyper-CVAD regimen at the University of Texas M. D. Anderson Cancer Center (UTMDACC) to determine the incidence of secondary myeloid malignancies (AML/MDS) and to discern the characteristics of these patients.

MATERIALS AND METHODS

Patients

Adult patients with ALL, treated at the UTMDACC from 1992 to 2007 with the hyper-CVAD regimen or its variants and who were enrolled in clinical trials were included in this analysis. The studies were approved by the institutional review board, and all patients signed informed consent according to institutional guidelines. The eligibility criteria, treatment, and follow-up methods have been described earlier.14 Patients were regularly observed during their initial treatment and subsequent remission.

Evaluation of Patients and Statistical Methods

Patients were evaluated for response to treatment and for the development of secondary myeloid neoplasms. The duration of complete remission (CR) was calculated from the time of achieving CR until there was evidence of leukemia recurrence (eg, ≥10% of lymphoblasts in the bone marrow, or central nervous system or extramedullary disease recurrence). Time to the development of AML/MDS was noted from the time of ALL diagnosis. Survival after the therapy-related AML/MDS was calculated from the time of diagnosis of the AML/MDS to death from any cause or until last follow-up. The Kaplan–Meier method was used to determine survival.

RESULTS

We identified 641 adult patients with ALL who had undergone treatment with the hyper-CVAD-based regimens. The median age of the patients was 40 years (range, 15-92 years). Of these, 16 (2.49%) patients developed therapy-related AML/MDS; 1 other patient developed a squamous cell cancer of the skin. These were the only other secondary malignancies observed in the entire cohort of patients. Among the patients with secondary AML/MDS, male and female distribution was equal. The different subtypes of ALL diagnosed in these patients were pre-B ALL in 10 patients, mature-B ALL in 2 patients, and T-cell ALL in 4 patients. The median follow-up of the patients from the time of ALL diagnosis was 41.75 months (range, 21-90.5 months). The characteristics of these patients at ALL diagnosis are shown in Table 1. The treatment of ALL was varied and included hyper-CVAD–based regimens as shown in Table 2. The response to the treatment is also reported in Table 2. As depicted, all 16 patients achieved CR with the hyper-CVAD treatment. The cumulative doses of the chemotherapy drugs administered are listed in Table 3.

Table 1. Characteristics of Patients With an Acute Lymphoblastic Leukemia Diagnosis (n=16)
CharacteristicNo. (%)Median (Range)
  1. ALL indicates acute lymphoblastic leukemia.

Age at diagnosis of ALL, y 53 (23-66)
Cytogenetics
 Diploid11 (69) 
 Pseudo–diploid with del(2)1 (6) 
 t(9;22)1 (6) 
 Unavailable3 (19) 
ALL morphology:
 Pre-B cell10 (63) 
 T-cell type4 (25) 
 Mature B cell2 (12) 
Table 2. Response to Treatment of Acute Lymphoblastic Leukemia (n=16)
 No. (%)Median (Range)
  1. ALL indicates acute lymphoblastic leukemia; hyper-CVAD, hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone; CR, complete remission.

Therapy for ALL
Hyper-CVAD7 (44) 
Hyper-CVAD with rituximab8 (50) 
Hyper-CVAD with imatinib1 (6) 
Response to treatmentCR in all cases 
Remission duration, mo 30.2 (10.5–73.5)
Table 3. Cumulative Doses of the Chemotherapeutic Drugs and Radiation in the Hyperfractionated Cyclophosphamide, Vincristine, Doxorubicin, and Dexamethasone Regimen*
 Dose
  • iv indicates intravenously; po, orally; CNS, central nervous system; Gy, grays.

  • *

    There were some variations in the regimen depending on the particular protocol used.

Dose-intensive phase
 Cyclophosphamide, g/m2 iv7.2
 Vincristine, mg iv16
 Doxorubicin, mg/m2 iv200
 Dexamethasone, g1.28
 Methotrexate, g/m2 iv4
 Cytarabine, g/m2 iv48
Maintenance phase
 6-mercaptopurine, g po109
 Vincristine, mg iv48
 Methotrexate, g po2.08
 Prednisone, g po24
Early and late intensifications
 Methotrexate, mg/m2 iv800
 L-asparaginase, U iv160,000
 Cyclophosphamide, g/m2 iv3.6
 Vincristine, mg iv8
 Doxorubicin, mg/m2 iv100
 Dexamethasone, g iv0.64
Intrathecal treatment: 4 or 8 or 16, depending on CNS risk
 Methotrexate, mg on Day 212
 Cytarabine, mg on Day 8100
Radiotherapy, Gy
 Mediastinal disease at diagnosis30
 Cranial nerve root involvement20-30

The patients had been followed for a median of 30.2 months (range, 10.5-73.5 months) after achieving CR until the development of AML/MDS. Six patients of 16 (37%) developed AML, and 10 (63%) patients developed MDS. The characteristics of patients at the time of identification of AML/MDS are shown in Table 4. None of these 16 patients underwent an allogeneic stem cell transplantation, and only 2 had received radiotherapy as part of their therapy for ALL; both developed MDS. AML/MDS was identified at a median of 32 months (range, 11-74 months) after ALL diagnosis. The time to development of AML appeared to be longer, with a median of 35.5 months (range, 13-74 months) when compared with the time to the development of MDS (median, 31 months; range, 11-42 months), but the difference was not statistically significant (P = .38). The cytogenetic abnormalities at the time of detection of AML/MDS are shown in Table 4.

Table 4. Patient Characteristics at Time of Diagnosis of Therapy-related Acute Myelogenous Leukemia/Myelodysplastic Syndrome (n=16)
CharacteristicNo. (%)Median (Range)
  1. AML indicates acute myelogenous leukemia; MDS, myelodysplastic syndrome; ALL, acute lymphoblastic leukemia.

Age at diagnosis of AML/MDS, y 56 (26–70)
Time from ALL diagnosis to detection of AML/MDS, mo
 For all patients  (AML + MDS)16 (100)32 (11–74)
 For patients with  secondary AML6 (37)35.5 (13–74)
 For patients with  secondary MDS10 (63)31 (11–42)
Cytogenetics
 Diploid1 (6) 
 del(5) del(7)9 (57) 
 inv(11)1 (6) 
 t(4;11)1 (6) 
 Complex1 (6) 
 del(20)1 (6) 
 Unavailable2 (12) 

After the diagnosis of AML/MDS, the patients were treated with a variety of regimens that included the combination of cytarabine with an anthracycline in 11 of 16 (69%) patients, arsenic trioxide in 1 (6%) patient, clofarabine in 1 (6%) patient, and decitabine in 2 (12%) patients. The response to treatment is shown in Table 5. Only 3 patients (1 [6%] with AML and 2 [12%]) with MDS achieved a CR. Three (19%) patients with AML and 3 (19%) with MDS achieved a partial remission, whereas 2 (12%) patients with AML and 4 (25%) patients with MDS had no response. One patient with MDS had not been treated at the time of last follow-up, 1 month after the diagnosis.

Table 5. Response to Treatment of Acute Myelogenous Leukemia/Myelodysplastic Syndrome (n=16)
ResponseNo. (%)
 AMLMDSTotal
  1. AML indicates acute myelogenous leukemia; MDS, myelodysplastic syndrome; CR, complete remission; PR, partial remission.

CR1 (6)2 (12)3 (19)
PR3 (19)3 (19)6 (38)
No response2 (12)4 (25)6 (37)
Untreated 1 (6)1 (6)

Eight patients (1 with AML and 7 with MDS) underwent an allogeneic stem cell transplantation at a median of 5.5 months (range, 3.5–9 months) from the time of the diagnosis of AML/MDS. All but 2 of these 8 patients died at a median of 9.75 months (range, 6.5–15 months) after the detection of AML/MDS. The remaining 2 patients (both of whom had MDS) received an allogeneic stem cell transplantation (1 from a sibling and 1 from an unrelated donor) at 3.5 months and 5 months, respectively, after the diagnosis. They were alive 26 months and 10 months, respectively, after the diagnosis of MDS, and remained in CR at the time of last follow-up. Two other patients (both of whom had MDS) who did not undergo transplant were alive 1 month and 24 months, respectively, after the diagnosis of MDS. One patient had not been treated at the time of last follow-up, 1 month after the diagnosis, and the other patient received decitabine and achieved a CR, but developed disease recurrence after 11 months. The median overall survival after the detection of therapy-related AML/MDS is 9.25 months (range, 1+ to 26+ months) and is depicted in Table 6 and Figure 1. The median survival times for therapy-related AML and MDS are 8.25 months (range, 6-16 months) and 10.5 months (range, 1+ to 26+ months), respectively. The difference is not statistically significant (P = 1).

Figure 1.

Survival curve is shown for patients with therapy-related acute myelogenous leukemia and myelodysplastic syndrome (AML/MDS) that developed during treatment with the hyper-CVAD (hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone) regimen.

Table 6. Survival (in Months) After the Development of Therapy-related Acute Myelogenous Leukemia/Myelodysplastic Syndrome
 No. of PatientsMedian (Range)
  1. AML indicates acute myelogenous leukemia; MDS, myelodysplastic syndrome; SCT, stem cell transplantation.

Survival
 Overall169.25 (1++ to 26++)
 AML68.25 (6-16)
 MDS1010.5 (1++ to 26++)
Survival with allogeneic SCT after AML/MDS
 AML15
 MDS73.5 (0.5-22.5++)

DISCUSSION

Therapy-related (secondary) cancers, particularly myeloid neoplasms (AML/MDS), have been well described after successful treatment of pediatric ALL, but to our knowledge are not commonly reported in the adult population.2-9 Secondary AML/MDS have been described after alkylating agents (with −5, −7 karyotype, 5 to 7 years after therapy) and after topoisomerase inhibitors (with 11q23− and t[4;11] karyotype) 2 to 3 years after therapy.10-13 Several factors, including genetic predisposition, the biologic characteristics of the primary cancer, and type of chemotherapy, may affect the development of secondary cancer.

Tavernier et al reported 9 adult patients (of 1494 [0.6%]) with ALL who developed secondary AML (6 patients) and MDS (3 patients).9 In these patients, AML and MDS developed at a median of 16 months and 30 months, respectively, after the diagnosis of ALL; the median overall survival for all patients was 5.7 months. Their cohort of patients had an age range of 15 to 60 years (median, 33 years). In the current study, the patients' ages ranged from15 to 92 years (median, 40 years), and secondary AML and MDS developed at a median of 35.5 months and 31 months, respectively, after the diagnosis and treatment of ALL. This difference in time to the development of AML and MDS was not found to be statistically significant (P = .07). The higher incidence of AML/MDS in our patients (2.49% vs 0.6% in report by Tavernier et al9) could be because of the older patient population or higher cumulative dose of cyclophosphamide in the hyper-CVAD regimen, or a combination of both. Gill et al reported on 4 patients with ALL who subsequently developed AML/MDS at 4 months, 21 months, 24 months, and 37 months, respectively, after therapy.15

We have found a relatively low incidence of therapy-related AML/MDS in adult patients with ALL. Our patients had been followed for a median of 32 months (range, 11–74 months) after the diagnosis of ALL before they developed AML or MDS. With the currently available regimens, survival for adult patients with ALL is limited (approximately 40% at 3 years). Therefore, a significant number of patients do not survive long enough to allow a secondary cancer to develop, and the risk of a second malignancy in adult ALL is likely to be underestimated. With continued improvement in the outcomes of patients with ALL, we may well encounter a higher incidence of therapy-related AML/MDS.

We conclude that secondary AML and MDS occur infrequently (2.49%) in adult patients with de novo ALL who are treated with the hyper-CVAD regimens, and response to their treatment is poor. Because of the small number of patients in the current study, we were unable to identify any specific risk factors that might predispose to the development of secondary leukemia. Furthermore, no specific cytogenetic abnormalities occurred more frequently than expected. It is interesting to note that the median age of our patients with therapy-related AML/MDS at ALL diagnosis was 53 years (range, 23–66 years). This was significantly older than the median age of 40 years (range, 15–92 years) for the entire population (P < .0001). We can hypothesize that the bone marrow stem cells of older patients are more susceptible to the DNA-damaging effects of the cytotoxic regimens used. Although an underlying genomic instability may contribute to the risk of the development of secondary AML and MDS, it is likely that the more significant culprit is the intensive cytotoxic regimens used to treat patients with ALL. In the future, a risk-adapted approach, in which the intensity of chemotherapy may be adjusted to the disease characteristics and the presence or absence of minimal residual leukemia, may allow for a reduction in exposure to the potentially causative cytotoxic agents. Furthermore, the development of rationally designed and potent target-specific agents such as the second-generation tyrosine kinase inhibitors may allow for the design of less intensive regimens with similar efficacy in specific subsets of adult patients with ALL.

Conflict of Interest Disclosures

The authors made no disclosures.

Ancillary