This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Armitage, J. O. Articles by Kroll, S. Search for Related Content PubMed PubMed Citation Articles by Armitage, J. O. Articles by Kroll, S.

Treatment-Related Myelodysplasia and Acute Leukemia in Non-Hodgkin’s Lymphoma Patients

From the University of Nebraska College of Medicine, Omaha, NE; University of Wisconsin Medical School, Madison, WI; University of Southern California Norris Cancer Hospital, Los Angeles, CA; University of Rochester Cancer Center, Rochester, NY; Corixa Corporation, Seattle WA; and British Columbia Cancer Agency Vancouver Clinic, Vancouver, British Columbia, Canada.

Address reprint requests to James O. Armitage, MD, University of Nebraska College of Medicine, 600 S 42nd St, Nebraska Medical Center Box 98332, Omaha, NE 68198-3332; email: joarmita{at}unmc.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
Purpose: Standard therapies for non-Hodgkin’s lymphoma (NHL) are associated with an increased risk of developing treatment-related myelodysplastic syndrome or acute myelogenous leukemia (tMDS/AML). However, there is considerable debate over the incidence or risk of tMDS/AML in NHL patients treated with any particular modality and the factors that contribute to malignant transformation.

Design: Conclusions were based on thorough analysis of data reported in the peer-reviewed literature and careful examination of the statistical methodology and methods for identifying cases of tMDS/AML. Unless noted, data are reported only for NHL patients, excluding Hodgkin’s disease patients.

Results: Despite differences in methods used to identify cases and to estimate the cumulative incidence over time (actuarial v cumulative calculations), up to 10% of NHL patients treated with either conventional-dose chemotherapy or high-dose therapy and autologous stem-cell transplantation may develop tMDS/AML within 10 years of primary therapy. Kaplan-Meier estimates of the actuarial incidence, which are based on censoring of patients who died without developing tMDS/AML, can lead to artificially high estimates with large confidence intervals at later time points. Although there is much debate about the cause(s) of tMDS/AML, there is compelling evidence that alkylating agents, certain other leukemogenic agents, and total-body irradiation (TBI) cause chromosomal damage that can lead to tMDS/AML.

Conclusion: Limiting exposure to alkylating agents and eliminating TBI from transplantation conditioning regimens may reduce the relative risk of tMDS/AML.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
STANDARD THERAPIES for non-Hodgkin’s lymphoma (NHL) are associated with an increased risk of developing treatment-related myelodysplasia (tMDS) or acute myelogenous leukemia (tAML). These secondary malignancies usually arise within 10 years (median, 5 to 7 years) after exposure^1,2 and are associated with a poor prognosis. A number of risk factors have been identified, including age, prolonged exposure to alkylating agents, and exposure to total-body irradiation (TBI). However, numerous fundamental questions remain unanswered with regard to the actual (as opposed to the actuarial) incidence of tMDS/AML, the etiology of the disease, and the role of additional patient or treatment-related risk factors. Despite a wealth of information from single-institution studies and analyses of large databases and transplant registries, many of these questions remain unanswered because of the variables among these studies that cannot be reconciled. These variables include differences in patient populations, treatment regimens, and methods used to identify patients with tMDS, as well as ambiguities with respect to the statistical methods used to estimate the cumulative incidence over time. Nevertheless, some consensus has begun to emerge and is the basis for this review.

	Clinical and Cytogenetic Characteristics of tMDS/AML

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
tMDS and tAML are clinically and cytogenetically distinct from de novo cases; tMDS is associated with a high risk of evolution to frank AML, increased blasts (5% to 29%) in the bone marrow, and a complex karyotype.^3,4 tMDS typically transforms to AML within approximately 6 months, and patients generally have a poor prognosis, with a median survival of less than 1 year from diagnosis.^3,4 Response to antileukemia therapy is usually poor, and subsequent bone marrow failure often leads to fatal bleeding and infectious complications. Patients with tMDS/AML may present with symptomatic anemia and thrombocytopenia and may be diagnosed clinically, but tMDS/AML is most often diagnosed when decreased blood cell counts lead to discovery of bone marrow dysplasia or characteristic cytogenetic abnormalities during routine follow-up.

The majority of patients with tMDS/AML have complex karyotypes, and the observed cytogenetic changes are often characteristic of chemotherapy-induced chromosomal damage.^5,6 Alkylating agents commonly used in the treatment of NHL, including carmustine, chlorambucil, cyclophosphamide, and melphalan, are associated with complete or partial deletions of chromosomes 5q and 7q,⁴ which can lead to myelodysplasia with a latency of 5 to 7 years. Complete or partial deletions of chromosomes 5 and/or 7 have been reported in approximately 90% of tMDS/AML patients.^4,5 The most common abnormalities are monosomy 5 or 7 and deletion of 5q. Topoisomerase II inhibitors, including etoposide and doxorubicin, are typically associated with balanced translocations involving chromosome bands 11q23 and 21q22.^7,8 Translocations at 11q23 frequently involve the MLL gene, which encodes a transcriptional regulator of genes associated with hematopoiesis.^9,10 These translocations generally result in a distinctly different pattern of secondary malignancy compared with that seen after treatment with alkylating agents. In these cases, tAML typically develops without a preleukemic phase and with a shorter latency period of 6 months to 5 years.⁴ This is clinically important because etoposide is widely used in the priming of autologous stem cells for subsequent transplantation and because etoposide-containing regimens are often used as salvage therapy for NHL. Cytogenetics has also been shown to correlate with response to intensive chemotherapy and survival.^5,11 Patients presenting with complete or partial deletions of chromosomes 5q and 7q respond poorly and have a poor prognosis, whereas patients who present with overt tAML and balanced chromosome translocations often respond well to intensive chemotherapy, with some cytogenetic subgroups achieving long-term complete remissions.¹²

	INCIDENCE OF tMDS/AML

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
There is considerable debate over the actual incidence or risk of tMDS/AML in NHL patients treated with any particular modality. The answer to this question is complicated by the fact that the incidence of tMDS/AML is expected to increase over time with continued follow-up, because exposure to leukemogenic chemotherapy and radiation increases with time from diagnosis and because of the known latency period between bone marrow injury and the development of myelodysplasia or leukemia. Most cases have been reported to occur between 2 and 10 years after initial therapy, with a median of approximately 6 years.^1,2,5,6,13 Because the incidence increases with continued follow-up, and because median follow-up varies between studies, crude incidence (ie, number of patients diagnosed with tMDS/AML divided by the total number of patients) is inadequate to accurately assess the true incidence of this complication over time. Generally, the crude incidence underestimates the true incidence, because with additional follow-up additional cases are diagnosed, and the number of patients at risk decreases as patients die. Therefore, investigators have relied on statistical methods to arrive at an actuarial estimate of the incidence of tMDS/AML. However, actuarial estimates can overestimate the true incidence if they do not account for competing risks.

Actuarial Estimates
Censored data methods are available for estimating the incidence of tMDS/AML. Two methods have been used in the literature, one that ignores competing risks and one that incorporates competing risks. Both actuarial methods can account for ongoing observation by censoring the patients at the time of their last assessment. However, competing risks analyses also account for the fact that individuals may fail from more than one cause (eg, diagnosis with tMDS/AML v death from other causes before a diagnosis of tMDS/AML).

Kaplan-Meier estimates. The Kaplan-Meier method is commonly used to arrive at an actuarial estimate, but this method ignores the presence of competing risks.¹⁴ Given that most statistical software packages do not include programs that incorporate competing risks, the Kaplan-Meier estimate is the most frequently reported estimate. In the classic Kaplan-Meier estimate, patients who were not diagnosed with tMDS/AML at their last follow-up are censored at the time of their last assessment or at the time of death. Because patients are censored at death, the estimate does not take into account competing causes of death. In other words, this method assumes that the risk of death from other causes has been completely removed. This assumption is difficult to justify given the well-known nonidentifiability problem associated with competing risks.^15,16 Consequently, the Kaplan-Meier estimates can give unexpected results. For example, if a patient with the longest follow-up is diagnosed with tMDS/AML, the estimate increases to 100%. SEs for the Kaplan-Meier estimate are usually calculated using Greenwood’s formula.¹⁷ The SE increases with each event (eg, each new diagnosis of tMDS/AML) and increases as the number of patients at risk decreases. The SE of the Kaplan-Meier estimate is especially problematic when an event occurs with a risk set of five or fewer patients. It is important that SEs and/or confidence intervals accompany the reported Kaplan-Meier estimates.

Cumulative incidence. In contrast, the cumulative incidence estimates the percentage of patients who will be diagnosed with tMDS/AML in a certain time interval in the presence of competing risks. It differs from the Kaplan-Meier estimate in that patients must both survive competing risks and be diagnosed with tMDS/AML. For example, a 5-year cumulative incidence estimate of 5% is interpreted to mean that if a group of patients receives therapy today and we wait 5 years, it is estimated that 5% of the patients will have been diagnosed with tMDS/AML before death from any other cause. This type of analysis has been described as a competing risks analysis,¹⁸ absolute cause-specific risk analysis,¹⁹ and cumulative incidence.²⁰ Andersen et al²⁰ provide a nonparametric estimate and associated SE. Similar to the Kaplan-Meier estimate, the SE for the cumulative incidence depends on the number of events and the size of the risk set at each event. SEs and/or confidence intervals should accompany the reported cumulative incidence.

Comparisons of Kaplan-Meier and cumulative incidence. A hypothetical example is provided in Table 1 to illustrate the practical differences between the Kaplan-Meier and cumulative incidence estimates. Suppose a cohort of 100 patients is followed-up for 10 years. One patient is diagnosed with tMDS/AML 0.5 years after treatment and one additional patient is diagnosed every subsequent year until 9.5 years after treatment. Thus, a total of 10 patients are diagnosed with tMDS/AML. Assume that nine patients die every year before a diagnosis of tMDS/AML (ie, 90 patients die from competing risks), and all patients are followed-up until death or diagnosis of tMDS/AML. The number of patients at risk and the number diagnosed with tMDS/AML are listed in Table 1.

View larger version (19K): [in this window] [in a new window]   Fig 1. Kaplan-Meier estimate of actuarial incidence compared with cumulative incidence calculated using Andersen’s competing risks analysis based on hypothetical data from Table 1.

Given the potential for overestimation of the true incidence of tMDS/AML using the Kaplan-Meier method, it is important to interpret the estimates reported in the literature in light of the statistical method used. However, there is a degree of ambiguity with regard to the statistical methodology used in the literature because of confusion over the nomenclature. It seems that most published studies have used the Kaplan-Meier method to arrive at an actuarial estimate, but the results are often presented as the cumulative incidence. Among all the studies presented here, only Darrington et al²¹ have explicitly calculated a cumulative incidence, and Stone et al²² explicitly stated that they used the actuarial method. Unfortunately, for most studies presented in this review, the authors do not explain their methods in sufficient detail to discern whether the estimate is cumulative or actuarial.^1,2,23–35 Therefore, unless otherwise noted, the estimates are assumed to be an actuarial incidence; 95% confidence intervals and SEs are provided with the estimate when available. The reliability of the statistical estimate of incidence must be considered when evaluating the relative risk of secondary myeloproliferative disease associated with any particular treatment modality.

Incidence in Patients Treated With Conventional-Dose Chemotherapy and Radiotherapy
Some of the earliest studies investigated the risk of tMDS/AML in NHL patients after conventional-dose chemotherapy and radiotherapy in the nontransplantation setting (Table 2).^{1,2,23–26,36} In the majority of these studies, patients were treated with chemotherapy, typically alkylating agents, with or without local radiotherapy. These studies clearly demonstrate the risk associated with conventional-dose chemotherapy; however, the risk associated with local radiotherapy is less certain. Two studies have reported no cases of tMDS/AML among 92 patients treated with local radiotherapy alone.^23,36 However, two published series of tMDS/AML cases have reported several cases of myelodysplasia among patients who received only radiotherapy, although in the majority of these patients radiotherapy was extensive, including a major portion of the pelvic bone marrow or total lymph node irradiation.^5,6

These results were confirmed by a study of 686 NHL patients treated at Duke University Medical Center between 1970 and 1981, which identified nine patients with tMDS/AML after a median follow-up of 5.5 years (1.3%).²³ None of the 72 patients who received local radiotherapy alone developed tMDS/AML, whereas five (1.6%) of 322 patients treated with chemotherapy alone and four (1.4%) of 292 patients treated with chemotherapy plus local radiotherapy developed tMDS/AML. All patients treated with chemotherapy received cyclophosphamide either alone or in combination with other agents. The conclusion from this study, which provides the only direct comparison of the relative risk in patients treated with local radiotherapy or chemotherapy alone versus combined-modality therapy, is that local radiotherapy does not increase the risk of secondary myelodysplasia or leukemia above that associated with alkylating agents alone. The estimated actuarial incidence of tMDS/AML among patients treated with chemotherapy plus local radiotherapy was somewhat lower in this study (4.5% at 10 years) than that reported in the NCI and Copenhagen studies. This may reflect differences in the statistical methods or a real difference in the observed incidence. However, the crude incidence was similar in these three studies with similar lengths of follow-up. Taken together, these three trials suggest that the actuarial incidence of tMDS/AML in patients treated with conventional-dose chemotherapy with or without local radiotherapy is in the range of 5% to 8% at 10 years.

In contrast to local radiotherapy, treatment with low-dose TBI or total lymph node irradiation seems to be associated with an increased risk of tMDS/AML, particularly when combined with chemotherapy. At a median follow-up of 3.6 years, Gomez et al³⁶ reported one patient with tMDS/AML among 41 patients (2.4%) who received total lymph node irradiation alone and four patients with tMDS/AML among 40 patients (10%) who received total lymph node irradiation plus alkylating agents. Travis et al²⁴ reported a similarly high incidence of tMDS/AML among 41 patients who received TBI plus salvage therapy with alkylating agents. Five (12%) of 41 patients treated with this combination developed tMDS/AML at a mean follow-up of 9.7 years. The majority of cases reported in this cohort (four of five) occurred in the subset of patients who received a median radiation dose to the bone marrow of 8.8 Gy (range, 1.1 to 20.7 Gy). Mendenhall et al²⁶ also reported four patients with tMDS/AML among 44 patients (9%) treated with low-dose TBI alone at 10 years follow-up; all four patients received a cumulative radiation dose greater than 2 Gy. These studies suggest that low-dose TBI is leukemogenic and that the combination of low-dose TBI plus alkylating agents may have synergistic leukemogenic effects.

Incidence in Patients Treated With High-Dose Therapy and Autologous Transplantation
The dramatic increase during the last 10 to 15 years in the use of high-dose therapy followed by autologous stem-cell transplantation (ASCT) to treat patients with relapsed or refractory NHL or Hodgkin’s disease, together with reports of the high incidence of tMDS/AML in these patients, has provided the impetus for a host of studies to examine this problem (Table 3).^{21,22,27–35,37,38} Several reviews and editorials during the last 6 or 7 years have sought to draw some conclusions from these reports.^39–42 However, the actuarial or cumulative incidence of tMDS/AML reported in these studies, typically calculated from the time of transplantation, has varied widely, perhaps owing to differences in statistical methods or because of differences in patient populations and conditioning regimens.

Rohatiner et al²⁸ observed four patients with tMDS/AML among 64 patients (6.3%) with follicular lymphoma treated at St Bartholomew’s Hospital at a median follow-up of 3.5 years. All four patients had received extensive prior therapy with chlorambucil and/or cyclophosphamide, and the conditioning regimen consisted of cyclophosphamide plus TBI. A follow-up report on the St Bartholomew experience published recently by Micallef et al³⁵ with a median follow-up of 6 years documented 27 patients with tMDS/AML among 230 NHL patients (12%) who received cyclophosphamide plus TBI. The median time to development of MDS was 4.4 years (range, 0.9 to 8.8 years) from transplantation and 9.1 years (range, 2.7 to 21.6 years) from diagnosis of lymphoma. The median time from diagnosis to transplantation was 4 years among patients who developed tMDS/AML. This is the highest crude incidence reported in any study, but it seems consistent with the initial report, given the longer duration of follow-up.

Two reports from the Dana-Farber Cancer Institute found a similar incidence of tMDS/AML to that reported by Rohatiner et al, despite differences in median follow-up, in NHL patients treated with cyclophosphamide plus TBI followed by autologous bone marrow transplantation (ABMT). The first report by Stone et al²² in 1994 was based on analysis of 262 patients. In this cohort, 20 patients (7.6%) were diagnosed with tMDS/AML at a median follow-up of 2.6 years. The median time to development of MDS was 2.6 years (range, 0.8 to 8.4 years) from transplantation and 5.8 years (range, 2.3 to 11.8 years) from initial treatment for lymphoma. As in the other studies, all patients who developed tMDS/AML had been extensively pretreated with alkylating agents. The subsequent report by Friedberg et al³¹ documented 41 cases among 552 patients (7.4%) at a median follow-up of 6.3 years. Therefore, despite nearly twice the length of follow-up, there did not seem to be an increased crude incidence.

Two studies conducted at the City of Hope National Medical Center have demonstrated a relatively low incidence of tMDS/AML among patients who underwent ABMT. In the initial report by Traweek et al²⁹ in 1994, clonal chromosomal abnormalities consistent with tMDS/AML were detected in six (3.6%) of 167 NHL patients after they had undergone ABMT. All patients had morphologically and cytogenetically normal bone marrow at the time of stem-cell collection. The conditioning regimens used in this study consisted of etoposide and cyclophosphamide, with or without TBI, depending on the patient’s prior exposure to local radiotherapy. More recently Krishnan et al³⁴ reported the results of a retrospective case study using a nested case-control study design. This analysis identified only 11 patients with tMDS/AML among 394 NHL patients (2.8%) treated between 1986 and 1998 at the City of Hope National Medical Center. The median time to diagnosis of tMDS/AML was 2.4 years from transplantation and 3.8 years from primary diagnosis; however, these patients underwent transplantation early in the course of their disease.

Finally, Darrington et al²¹ reported a low overall incidence of tMDS among NHL patients treated at the University of Nebraska; only six (2.3%) of 262 patients developed tMDS with a median time of 3.7 years from transplantation and 5.7 years (range, 2.5 to 8.6 years) from initial diagnosis of lymphoma. Interestingly, all six of these patients received a conditioning regimen that included TBI, whereas none of 151 patients conditioned with chemotherapy alone (melphalan or cyclophosphamide) developed tMDS.

If one considers only the crude incidence reported in these studies, one might reach the conclusion that the incidence of tMDS/AML in patients treated with ASCT is as much as two- to three-fold higher than that reported for patients treated with conventional-dose chemotherapy, with or without local radiotherapy, as described in Table 2. There are several possible explanations for this, none of which are mutually exclusive. First, the fact that patients undergoing transplantation tend to receive more detailed follow-up, including pathologic and cytogenetic analysis of the bone marrow, might result in improved or earlier diagnosis of tMDS/AML. Second, the higher incidence may reflect the longer follow-up time relative to the initial diagnosis and start of primary therapy. Lastly, the conditioning regimen or some aspect of the transplantation itself may cause tMDS/AML or accelerate progression to MDS or leukemia.

Alternatively, the true cumulative incidence of tMDS/AML may be quite similar for patients treated with ASCT compared with conventional-dose chemotherapy. Darrington et al²¹ have explicitly reported a cumulative incidence, which was 8% at 5 years for patients conditioned with chemotherapy plus TBI. The highest cumulative incidence was 14.5% at 10 years for patients treated at the Dana-Farber Cancer Institute with chemotherapy plus TBI.⁴³ However, the estimates derived by several other studies are consistent with a cumulative incidence of 5% to 10% at 5 to 10 years after transplantation (see Rohatiner et al, Califaretti et al, and Krishnan et al in Table 3).^28,32,34 If one keeps in mind that these estimates are based on time from transplantation, which typically ranges from 2 to 5 years after the initial diagnosis of lymphoma, the cumulative incidence of tMDS/AML among patients receiving ASCT does not seem to be substantially greater than for patients receiving conventional-dose chemotherapy. This conclusion is also borne out by the European Bone Marrow Transplantation (EBMT) Lymphoma Registry study,³³ which estimated a 5.7% actuarial incidence at 10 years (95% confidence interval, 3.4% to 9.4%) on the basis of an analysis of 3,205 NHL patients. Although this estimate may be somewhat low because the study used questionnaires to identify cases, it may be more reflective of the actual incidence than the actuarial estimates from smaller studies.

A number of studies have, however, reported much higher actuarial estimates, ranging from 14% at 5 years to 36.5% at 10 years posttransplantation. The 18% (± 9%) 6-year estimate reported by Stone et al²² is explicitly an actuarial calculation that is based on censoring of patients at death. Likewise, the 14% (± 14.7%) 5-year estimate reported by Miller et al,²⁷ the 19.8% 10-year estimate reported by Friedberg et al,³¹ and the 36.5% 10-year estimate reported by Micallef et al³⁵ seem to be actuarial estimates that are similarly confounded by patient censoring. Therefore, these results should be interpreted with caution. In reply to a letter to the Journal of Clinical Oncology by Mounier and Gisselbrecht, Friedberg et al⁴³ reanalyzed the data from their series of 552 patients treated with chemotherapy plus TBI at the Dana-Farber Cancer Institute and reported a cumulative incidence of 14.5%, which is lower than the originally reported actuarial estimate of 19.8%, but the authors commented that this is still unacceptably high.

The possibility that the cumulative incidence of tMDS/AML in patients treated with ASCT is similar to that observed in patients treated with conventional-dose chemotherapy is supported by a number of studies.^38,40–42 This is particularly true if the estimated incidence is calculated from the time of primary therapy rather than from the time of transplantation. Unfortunately, no meta-analysis has yet been conducted to examine this possibility. Nonetheless, researchers in Copenhagen reached the conclusion that the incidence of tMDS/AML was similar between patients treated with conventional-dose chemotherapy and ASCT after comparing the actuarial incidence between a cohort of 76 lymphoma patients treated with high-dose therapy and ASCT, and seven different cohorts of patients with lymphomas or solid tumors who were treated at their institution with conventional-dose chemotherapy and who received the same rigorous follow-up.³⁸

	PREDISPOSING FACTORS FOR tMDS/AML

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
Despite more than 20 years of investigation, the etiology of tMDS/AML remains controversial. Alkylating agents and TBI are certainly the prime suspects; however, it remains unclear what their relative contributions are, particularly in the transplantation setting. A variety of factors that seem to be correlated with an increased risk of developing tMDS/AML have been identified by univariate and multivariate analyses, particularly in more recent studies involving ASCT.⁴⁰ The only patient-specific risk factor seems to be age. Several studies have identified age greater than 38 or 40 years,^21,22 or age as a continuous variable,^33,35 as a significant risk factor for tMDS/AML in NHL patients. With respect to treatment-related factors, a number of measures that reflect the extent of chemotherapy have been identified as significant risk factors, including duration of chemotherapy, number of prior relapses, and interval from diagnosis of lymphoma to ASCT. Other markers reflective of damage to the bone marrow, including cumulative radiation dose to the bone marrow and platelet count, have also been correlated with the risk of tMDS/AML.

One of the earliest attempts to draw a correlation between treatment-related factors and risk of tMDS/AML was reported by Greene et al¹ in their analysis of nontransplantation patients in the NCI database. They found that duration of chemotherapy was poorly correlated with risk, whereas increasing cumulative radiation dose to the bone marrow in patients who received TBI or hemibody irradiation showed a significant trend toward increased risk. However, a more detailed analysis of the mean cumulative dose of cyclophosphamide demonstrated that mean doses were two- to three-fold higher in patients who developed tMDS/AML than in patients who did not.

In the transplantation setting, indicators of the extent of previous chemotherapy have been found to be correlated with the risk of tMDS/AML. Several studies conducted at the Dana-Farber Cancer Institute have identified prolonged interval from diagnosis to ASCT, longer duration of prior chemotherapy, increased exposure to alkylating agents, number of prior relapses, and low platelet count as significant independent risk factors by univariate analyses.^22,30,41 Similarly, a multivariate analysis of the EBMT Lymphoma Registry demonstrated that the interval from diagnosis to first transplant and the number of transplants were prognostic factors for the development of tMDS/AML.³³ Micallef et al³⁵ have also reported that prior fludarabine therapy and time from diagnosis to ASCT were significant risk factors by a multivariate analysis. This study, and the analysis of the EBMT Lymphoma Registry,³³ further demonstrated a significant correlation between histologic subtype and the risk of tMDS/AML. Thus, patients with low-grade histology, who typically undergo transplantation later and receive more chemotherapy before transplantation, were at significantly greater risk compared with patients with high-grade lymphoma.

Prior radiotherapy has also been identified as a risk factor in several transplantation studies.^22,30,41 This likely reflects exposure to low-dose TBI rather than local radiotherapy. The available data suggest that cumulative bone marrow exposures less than 2 Gy are generally not associated with an increased risk.

The Role of Chemotherapy
Patients with Hodgkin’s disease and NHL typically receive multiple chemotherapy agents, particularly alkylating agents, that are known to be associated with an increased risk of tMDS/AML. Alkylating agents have long been suspected to be an important factor in the origin of secondary leukemias. Most notably, carmustine, melphalan, and cyclophosphamide are mutagenic in vitro and have leukemogenic potential in vivo, as documented in patients with a variety of primary malignancies, including ovarian cancer, gastrointestinal malignancies, and lymphoma.^2,44–47 Long-term treatment with alkylating agents results in cumulative chromosomal damage to bone marrow stem cells and a dose-dependent increase in the risk of developing treatment-related leukemia and preleukemia.^1,13,48,49 The risk of tMDS/AML after long-term treatment with alkylating agents has been estimated at 1% to 1.5% per year from 2 to 10 years after the start of primary chemotherapy,² and this could account for the observed incidence of tMDS/AML both in the nontransplantation and transplantation setting. Moreover, alkylating agents and topoisomerase II inhibitors produce characteristic abnormal karyotypes that are observed at a high frequency in patients with tMDS/AML.^{4–6,27,29,34,42}

Chemotherapy before transplantation. The role of prior exposure to alkylating agents in the etiology of tMDS/AML in patients who receive subsequent high-dose therapy and ASCT is well documented. In nearly every study, the majority of patients who have developed secondary MDS have been heavily pretreated with alkylating agents. In contrast, patients who undergo transplantation in first complete remission (CR) seem to have a much lower risk. For example, Taylor et al³⁷ reported no patients with tMDS/AML among 62 NHL patients, the majority of whom (n = 48) had high-grade lymphoma and underwent transplantation in first CR. However, unlike most other studies, patients in this study received autologous bone marrow that was not cryopreserved before transplantation. Given that the majority of patients who develop tMDS/AML after ASCT are older, have a history of extensive exposure to alkylating agents, and often have low platelet counts, these patients are likely to have suffered extensive damage to bone marrow before stem-cell collection. This conclusion is supported by cytogenetic data, which are based on a sensitive fluorescence in situ hybridization assay, that demonstrate the presence of chromosomal damage before high-dose therapy and ASCT in nine of 12 patients who developed tMDS/AML.⁵⁰ Nevertheless, there have been reports of patients who did not have extensive exposure to alkylating agents and who did not undergo transplantation in first CR but developed tMDS/AML. For instance, eight (6%) of 127 patients at the Dana-Farber Cancer Institute who received only six to eight cycles of standard cyclophosphamide, doxorubicin, vincristine, and prednisone and underwent transplantation in first CR developed tMDS/AML at a median follow-up of 6.3 years.³¹

The purine nucleoside analog fludarabine also seems to be leukemogenic, particularly when combined with an alkylating agent. Morrison et al⁵¹ recently reported one patient with tMDS/AML among 188 patients with chronic lymphocytic leukemia who were treated with fludarabine, and there were no patients with tMDS/AML among 191 patients treated with chlorambucil alone. However, five (3.5%) of 142 patients treated with fludarabine plus chlorambucil developed tMDS/AML at a median follow-up of 1.5 years. This represents a high crude incidence given the short length of follow-up. If this is confirmed in other studies, it suggests that this combination may result in accelerated leukemogenesis. Micallef et al³⁵ have also reported that prior fludarabine therapy was significantly associated with an increased risk of developing tMDS/AML posttransplantation in a multivariate analysis.

The Role of Radiation
There is little evidence to suggest that local radiotherapy is associated with an appreciable risk of tMDS/AML. In contrast, several reports have implicated total lymph node irradiation and low-dose TBI with high cumulative doses to the bone marrow in the etiology of secondary MDS.^24,26,36 Moreover, the risk seems to be increased by the combination of low-dose TBI with alkylating agents.²⁴ Animal studies suggest that low-dose TBI may expand the pool of bone marrow stem cells that are susceptible to chromosomal damage by alkylating agents.⁵² In the Dana-Farber series, a univariate analysis also identified prior pelvic radiation as a significant risk factor for patients undergoing ASCT.²²

Radiation as part of the conditioning regimen for ASCT. TBI has clearly been implicated in the etiology of tMDS/AML when used in the preparative regimen for ASCT. The Nebraska study reported by Darrington et al²¹ provided the first suggestion that TBI may be a significant risk factor for development of secondary MDS in patients undergoing ASCT. Several subsequent studies have added support to this hypothesis. First, a follow-up study from the Dana-Farber Cancer Institute, reported by Wheeler et al,³⁰ demonstrated a very low incidence of tMDS/AML among patients conditioned with high-dose cyclophosphamide, carmustine, and etoposide (Table 3). Only one of 150 NHL patients developed tMDS/AML at a median follow-up of nearly 4 years. This is in stark contrast to the 7.6% incidence at a median follow-up of 2.6 years among patients conditioned with cyclophosphamide plus TBI at the same institution.²² Most recently, Micallef et al³⁵ have also reported a high incidence of tMDS/AML (12% at a median follow-up of 6 years) among lymphoma patients at St Bartholomew’s Hospital who received TBI as part of the conditioning regimen. Subsequently, the results of a large EBMT cooperative group study of 3,205 patients in the EBMT Lymphoma Registry demonstrated that TBI as part of the conditioning regimen was a significant and independent risk factor for development of tMDS/AML in both univariate and multivariate analyses.³³ Indeed, after the use of TBI as part of the conditioning regimen was discontinued at the St Bartholomew’s Hospital in 1996, no patient has since developed tMDS/AML.³⁵

These studies raise the possibility that the conditioning regimen, and more specifically TBI, can increase the risk of tMDS/AML. Albeit, tMDS/AML has also been reported after conditioning with chemotherapy alone. Two (4.1%) of 49 NHL patients treated in Copenhagen with carmustine, etoposide, cytarabine, and melphalan before ASCT developed tMDS/AML (Table 3).³⁸ However, on the basis of the short latency period after ASCT and the patients’ prior exposure to alkylating agents, the authors concluded that these cases were most likely related to prior chemotherapy rather than to the conditioning regimen or ASCT.

Risk Associated With Radioimmunotherapy
Radionuclide-conjugated anti-CD20 antibodies, including tositumomab and iodine-131 (¹³¹I) tositumomab (Bexxar therapy; Corixa Corp, San Francisco, CA, and Glaxo SmithKline, Philadelphia, PA) and yttrium-90 (⁹⁰Y)–conjugated ibritumomab tiuxetan (Zevalin; IDEC Pharmaceuticals Corp, San Diego, CA), have recently been investigated for the treatment of low-grade NHL. Radioimmunotherapy delivers high radiation doses to tumors with minimal toxicity to normal organs. The dose-limiting toxicity is determined by the absorbed dose to bone marrow. Several cases of tMDS/AML have recently been reported in heavily pretreated relapsed and refractory NHL patients who were treated with radioimmunotherapy. Kaminski et al⁵³ reported five patients with tMDS/AML among 59 patients treated with ¹³¹I tositumomab in a phase I/II trial with a median follow-up of 3.1 years. In this trial, the median time from diagnosis to study entry was 45 months, patients had been heavily pretreated with a median of three prior chemotherapy regimens, and 14 patients had undergone prior ASCT. Likewise, Witzig et al⁵⁴ reported four patients with tMDS/AML among 211 patients treated with ⁹⁰Y-conjugated ibritumomab tiuxetan in four clinical trials. Patients in these trials had received a median of two prior therapies (range, one to nine), and none had undergone prior ASCT. It is too early to draw any conclusions with respect to the possible contribution of radioimmunotherapy to the risk of developing secondary malignancies; however, these reports seem to be consistent with the observed incidence of tMDS/AML in heavily treated NHL patients. Notably, no patients with tMDS/AML have been identified among 76 newly diagnosed NHL patients treated with ¹³¹I tositumomab,⁵⁵ and none have been reported among relapsed patients treated with other biologic agents, including the unconjugated anti-CD20 antibody, rituximab.^56–58

In conclusion, although there is tremendous variability in the reported incidence of tMDS and tAML, this is clearly an important clinical problem. Despite the differences between studies in terms of the methods used to identify cases and to estimate the incidence over time (actuarial v cumulative incidence), it seems that up to 10% of patients treated for NHL using either conventional-dose chemotherapy or high-dose therapy and ASCT may develop this complication within 10 years of primary therapy. The incidence of the diagnosis of tMDS/AML is dependent on the type of follow-up and frequency of blood examinations leading to bone marrow biopsies and associated cytogenetic analyses. It can be expected that a number of patients will die from progressive NHL with undiagnosed tMDS/AML present. Therefore, reports on the rate of tMDS/AML should summarize the frequency and type of follow-up, as well as the criteria used for the diagnosis. It is also helpful if investigators clearly state the statistical methods used to estimate the incidence and avoid using the term cumulative incidence unless a competing risks analysis was performed.

Although there is much debate about the cause(s) of tMDS/AML, there is compelling evidence that alkylating agents and certain other leukemogenic agents, as well as TBI, can cause chromosomal damage, ultimately resulting in myelodysplasia and overt leukemia. The characteristic latency period and chromosomal damage observed in most patients with tMDS/AML suggest that prior chemotherapy was the primary cause of the chromosomal damage that resulted in myelodysplasia. However, use of TBI as part of the conditioning regimen before ASCT seems to increase the risk of chromosomal damage. Importantly, transplantation of collected bone marrow containing stem cells harboring occult chromosomal damage from previous exposure to leukemogenic chemotherapy may also lead to outgrowth of preleukemic clones. It is too early to determine whether radioimmunotherapy is associated with any increased risk, but to date the reported cases seem consistent with the observed incidence in heavily treated NHL patients.

Treatment decisions should be based on a rational risk-benefit analysis of the available data. Alternatives to alkylating agents might be considered if equally effective agents are available. Unfortunately, one of the newest agents for the treatment of lymphoma, namely fludarabine, also seems to be associated with an increased risk of tMDS/AML. In the transplantation setting, a variety of options are available. Many transplantation centers now use TBI much more selectively. Other approaches to minimize the risks include transplantation in first CR to reduce prior exposure to leukemogenic chemotherapy. Although all of these measures may ultimately reduce the risks, patients will likely continue to develop treatment-related MDS and leukemia, and efforts should continue to identify more effective treatment strategies for this serious complication. Currently, there is no approved therapy for MDS; therefore, the standard of care remains supportive care with growth factor support, transfusions, and antibiotics. However, the use of hypomethylating agents, such as 5-azacytidine and decitabine, seems promising.^59,60

	REFERENCES

TOP ABSTRACT INTRODUCTION Clinical and Cytogenetic... INCIDENCE OF tMDS/AML PREDISPOSING FACTORS FOR... REFERENCES

 
1. Greene MH, Young RC, Merrill JM, et al: Evidence of a treatment dose response in acute nonlymphocytic leukemias which occur after therapy of non-Hodgkin’s lymphoma. Cancer Res 43:1891–1898, 1983[Abstract/Free Full Text]

2. Pedersen-Bjergaard J, Ersboll J, Sorensen HM, et al: Risk of acute nonlymphocytic leukemia and preleukemia in patients treated with cyclophosphamide for non-Hodgkin’s lymphomas: Comparison with results obtained in patients treated for Hodgkin’s disease and ovarian carcinoma with other alkylating agents. Ann Intern Med 103:195–200, 1985[Medline]

3. Scheinberg DA, Maslak P, Weiss M: Myelodysplastic syndromes, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology (ed 5). Philadelphia, PA, Lippincott-Raven, 1997, pp 2388–2396

4. Thirman MJ, Larson RA: Therapy-related myeloid leukemia. Hematol Oncol Clin North Am 10:293–320, 1996[CrossRef][Medline]

5. Le Beau MM, Albain KS, Larson RA, et al: Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplastic syndromes and acute nonlymphocytic leukemia: Further evidence for characteristic abnormalities of chromosomes no. 5 and 7. J Clin Oncol 4:325–345, 1986[Abstract/Free Full Text]

6. Whang-Peng J, Young RC, Lee EC, et al: Cytogenetic studies in patients with secondary leukemia/dysmyelopoietic syndrome after different treatment modalities. Blood 71:403–414, 1988[Abstract/Free Full Text]

7. Pedersen-Bjergaard J, Philip P: Balanced translocations involving chromosome bands 11q23 and 21q22 are highly characteristic of myelodysplasia and leukemia following therapy with cytostatic agents targeting at DNA-topoisomerase II. Blood 78:1147–1148, 1991 (letter)[Free Full Text]

8. Larson RA, Le Beau MM, Ratain MJ, et al: Balanced translocations involving chromosome bands 11q23 and 21q22 in therapy-related leukemia. Blood 79:1892–1893, 1992[Free Full Text]

9. Rowley JD, Reshmi S, Sobulo O, et al: All patients with the T(11;16)(q23;p13.3) that involves MLL and CBP have treatment-related hematologic disorders. Blood 90:535–541, 1997[Abstract/Free Full Text]

10. Super HJ, McCabe NR, Thirman MJ, et al: Rearrangements of the MLL gene in therapy-related acute myeloid leukemia in patients previously treated with agents targeting DNA-topoisomerase II. Blood 82:3705–3711, 1993[Abstract/Free Full Text]

11. Pedersen-Bjergaard J, Philip P: Cytogenetic characteristics of therapy-related acute nonlymphocytic leukaemia, preleukaemia and acute myeloproliferative syndrome: Correlation with clinical data for 61 consecutive cases. Br J Haematol 66:199–207, 1987[Medline]

12. Andersen MK, Larson RA, Mauritzson N, et al: Therapy-related MDS or AML with INV (16) or T(15;17): Characteristics and relation to type of primary therapy—Preliminary results from the international workshop on leukemia karyotype and prior treatment. Blood 96:463a, 2000 (abstr 1995)

13. Pedersen-Bjergaard J, Specht L, Larsen SO, et al: Risk of therapy-related leukaemia and preleukaemia after Hodgkin’s disease: Relation to age, cumulative dose of alkylating agents, and time from chemotherapy. Lancet 2:83–88, 1987[Medline]

14. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457–481, 1958[CrossRef]

15. Cox DR: The analysis of exponentially distributed lifetime with two types of failure. J R Stat Soc B 21:411–421, 1959

16. Tsiatis A: A nonidentifiability aspect of the problem of competing risks. Proc Natl Acad Sci U S A 72:20–22, 1975[Abstract/Free Full Text]

17. Greenwood M: The natural duration of cancer. Rep Public Health Med Subj 33:1–26, 1926

18. Seal HL: Studies in the history of probability and statistics: XXXV. Multiple decrements or competing risks. Biometrika 64:429–439, 1977[Abstract/Free Full Text]

19. Benichou J, Gail MH: Estimates of absolute cause-specific risks in cohort studies. Biometrics 46:813–826, 1990[CrossRef][Medline]

20. Andersen PK, Boran G, Keiding N: Statistical Models Based on Counting Processes. New York, NY, Springer, 1993

21. Darrington DL, Vose JM, Anderson JR, et al: Incidence and characterization of secondary myelodysplastic syndrome and acute myelogenous leukemia following high-dose chemoradiotherapy and autologous stem-cell transplantation for lymphoid malignancies. J Clin Oncol 12:2527–2534, 1994[Abstract/Free Full Text]

22. Stone RM, Neuberg D, Soiffer R, et al: Myelodysplastic syndrome as a late complication following autologous bone marrow transplantation for non-Hodgkin’s lymphoma. J Clin Oncol 12:2535–2542, 1994[Abstract/Free Full Text]

23. Lavey RS, Eby NL, Prosnitz LR: Impact on second malignancy risk of the combined use of radiation and chemotherapy for lymphomas. Cancer 66:80–88, 1990[CrossRef][Medline]

24. Travis LB, Weeks J, Curtis RE, et al: Leukemia following low-dose total body irradiation and chemotherapy for non-Hodgkin’s lymphoma. J Clin Oncol 14:565–571, 1996[Abstract/Free Full Text]

25. Ingram L, Mott MG, Mann JR, et al: Second malignancies in children treated for non-Hodgkin’s lymphoma and T-cell leukaemia with the UKCCSG regimens. Br J Cancer 55:463–466, 1987[Medline]

26. Mendenhall NP, Noyes WD, Million RR: Total body irradiation for stage II–IV non-Hodgkin’s lymphoma: Ten-year follow-up. J Clin Oncol 7:67–74, 1989[Abstract]

27. Miller JS, Arthur DC, Litz CE, et al: Myelodysplastic syndrome after autologous bone marrow transplantation: An additional late complication of curative cancer therapy. Blood 83:3780–3786, 1994[Abstract/Free Full Text]

28. Rohatiner AZS, Johnson PWM, Price CGA, et al: Myeloablative therapy with autologous bone marrow transplantation as consolidation therapy for recurrent follicular lymphoma. J Clin Oncol 12:1177–1184, 1994[Abstract/Free Full Text]

29. Traweek ST, Slovak ML, Nademanee AP, et al: Clonal karyotypic hematopoietic cell abnormalities occurring after autologous bone marrow transplantation for Hodgkin’s disease and non-Hodgkin’s lymphoma. Blood 84:957–963, 1994[Abstract/Free Full Text]

30. Wheeler C, Khurshid A, Ibrahim J, et al: Low incidence of post-transplant myelodysplasia/acute leukemia (MDS/AML) in NHL patients autotransplanted after cyclophosphamide, carmustine and etoposide (CBV). Blood 90:385b, 1997 (abstr 4480)

31. Friedberg JW, Neuberg D, Stone RM, et al: Outcome in patients with myelodysplastic syndrome after autologous bone marrow transplantation for non-Hodgkin’s lymphoma. J Clin Oncol 17:3128–3135, 1999[Abstract/Free Full Text]

32. Califaretti N, Al-Homaidhi A, Panzarella T, et al: Secondary myelodysplasia (MDS) and acute myeloid leukemia (AML) after autologous blood or stem cell transplantation (ABMT) for Hodgkin’s disease (HD) and non-Hodgkin’s lymphoma (NHL). Blood 94:344a, 1999 (abstr 1536)

33. Milligan DW, Ruiz de Elvira MC, Kolb H-J, et al: Secondary leukaemia and myelodysplasia after autografting for lymphoma: Results from the EBMT—EBMT Lymphoma and Late Effects Working Parties. Br J Haematol 106:1020–1026, 1999[CrossRef][Medline]

34. Krishnan A, Bhatia S, Slovak ML, et al: Predictors of therapy-related leukemia and myelodysplasia following autologous transplantation for lymphoma: An assessment of risk factors. Blood 95:1588–1593, 2000[Abstract/Free Full Text]

35. Micallef INM, Lillington DM, Apostolidis J, et al: Therapy-related myelodysplasia and secondary acute myelogenous leukemia after high-dose therapy with autologous hematopoietic progenitor-cell support for lymphoid malignancies. J Clin Oncol 18:947–955, 2000[Abstract/Free Full Text]

36. Gomez GA, Aggarwal KK, Han T: Post-therapeutic acute malignant myeloproliferative syndrome and acute nonlymphocytic leukemia in non-Hodgkin’s lymphoma. Cancer 50:2285–2288, 1982[CrossRef][Medline]

37. Taylor PR, Jackson GH, Lennard AL, et al: Low incidence of myelodysplastic syndrome following transplantation using autologous non-cryopreserved bone marrow. Leukemia 11:1650–1653, 1997[CrossRef][Medline]

38. Pedersen-Bjergaard J, Pedersen M, Myhre J, et al: High risk of therapy-related leukemia after BEAM chemotherapy and autologous stem cell transplantation for previously treated lymphomas is mainly related to primary chemotherapy and not to the BEAM-transplantation procedure. Leukemia 11:1654–1660, 1997[CrossRef][Medline]

39. Armitage JO: Myelodysplasia and acute leukemia after autologous bone marrow transplantation. J Clin Oncol 18:945–946, 2000[Free Full Text]

40. Pedersen-Bjergaard J, Andersen MK, Christiansen DH: Therapy-related acute myeloid leukemia and myelodysplasia after high-dose chemotherapy and autologous stem cell transplantation. Blood 95:3273–3279, 2000[Abstract/Free Full Text]

41. Stone RM: Myelodysplastic syndrome after autologous transplantation for lymphoma: The price of progress. Blood 83:3437–3440, 1994[Free Full Text]

42. Traweek ST, Slovak ML, Nademanee AP, et al: Myelodysplasia and acute myeloid leukemia occurring after autologous bone marrow transplantation for lymphoma. Leuk Lymphoma 20:365–372, 1996[Medline]

43. Friedberg JN, Neuberg D, Freedman AS: In reply to Mounier N, Gisselbrecht C: Myelodysplasia after autotransplantation. J Clin Oncol 18:3446–3447, 2000 (letter)[Free Full Text]

44. Boice JD Jr, Greene MH, Killen JY Jr, et al: Leukemia and preleukemia after adjuvant treatment of gastrointestinal cancer with semustine (methyl-CCNU). N Engl J Med 309:1079–1084, 1983[Abstract]

45. Greene MH, Boice JD Jr, Greer BE, et al: Acute nonlymphocytic leukemia after therapy with alkylating agents for ovarian cancer: A study of five randomized clinical trials. N Engl J Med 307:1416–1421, 1982[Abstract]

46. Greene MH, Boice JD Jr, Strike TA: Carmustine as a cause of acute nonlymphocytic leukemia. N Engl J Med 313:579, 1985 (letter)[Medline]

47. Greene MH, Harris EL, Gershenson DM, et al: Melphalan may be a more potent leukemogen than cyclophosphamide. Ann Intern Med 105:360–367, 1986[Medline]

48. Bennett JM, Moloney WC, Greene MH, et al: Acute myeloid leukemia and other myelopathic disorders following treatment with alkylating agents. Hematol Pathol 1:99–104, 1987[Medline]

49. Cuzick J, Erskine S, Edelman D, et al: A comparison of the incidence of the myelodysplastic syndrome and acute myeloid leukaemia following melphalan and cyclophosphamide treatment for myelomatosis: A report to the Medical Research Council’s working party on leukaemia in adults. Br J Cancer 55:523–529, 1987[Medline]

50. Abruzzese E, Radford JE, Miller JS, et al: Detection of abnormal pretransplant clones in progenitor cells of patients who developed myelodysplasia after autologous transplantation. Blood 94:1814–1819, 1999[Abstract/Free Full Text]

51. Morrison VA, Rai KR, Peterson B, et al: Therapy-related myelodysplastic syndrome (t-MDS) or acute myeloid leukemia (t-AML) in patients with chronic lymphocytic leukemia (CLL) treated with chlorambucil (C), fludarabine (F), or fludarabine + chlorambucil (F + C): An Intergroup study (Cancer and Leukemia Group B (CALGB) 9011). Proc Am Soc Clin Oncol 18:9A, 1999 (abstr 29)

52. Rubin P, Constine LS III, Scarantino CW: The paradoxes in patterns and mechanism of bone marrow regeneration after irradiation: 2. Total body irradiation. Radiother Oncol 2:227–233, 1984[Medline]

53. Kaminski MS, Estes J, Zasadny KR, et al: Radioimmunotherapy with iodine ¹³¹I tositumomab for relapsed or refractory B-cell non-Hodgkin lymphoma: Updated results and long-term follow-up of the University of Michigan experience. Blood 96:1259–1266, 2000[Abstract/Free Full Text]

54. Witzig TE, Gordon LI, Wiseman GA, et al: Zevalin radioimmunotherapy is safe and well tolerated in patients with relapsed or refractory, B cell non-Hodgkin’s lymphoma (NHL). Blood 96:731a, 2000 (abstr 3160)

55. Kaminski MS, Estes J, Tuck M, et al: Iodine I 131 tositumomab therapy for previously untreated follicular lymphoma (FL). Proc Am Soc Clin Oncol 19:5a, 2000 (abstr 11)

56. McLaughlin P, Grillo-Lopez AJ, Link BK, et al: Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: Half of patients respond to a four-dose treatment program. J Clin Oncol 16:2825–2833, 1998[Abstract]

57. Piro LD, White CA, Grillo-Lopez AJ, et al: Extended rituximab (anti-CD20 monoclonal antibody) therapy for relapsed or refractory low-grade or follicular non-Hodgkin’s lymphoma. Ann Oncol 10:655–661, 1999[Abstract/Free Full Text]

58. Davis TA, Grillo-Lopez AJ, White CA, et al: Rituximab anti-CD20 monoclonal antibody therapy in non-Hodgkin’s lymphoma: Safety and efficacy of re-treatment. J Clin Oncol 18:3135–3143, 2000[Abstract/Free Full Text]

59. Silverman LR, Demakos EP, Peterson B, et al: A randomized controlled trial of subcutaneous azacitidine (AZA C) in patients with the myelodysplastic syndrome (MDS): A study of the cancer and leukemia group B (CALGB). Proc Am Clin Oncol Soc 17:14a, 1998 (abstr 53)

60. Wijermans P, Lubbert M, Verhoef G, et al: Low-dose 5-aza-2'-deoxycytidine, a DNA hypomethylating agent, for the treatment of high-risk myelodysplastic syndrome: A multicenter phase II study in elderly patients. J Clin Oncol 18:956–962, 2000[Abstract/Free Full Text]

Submitted July 24, 2001; accepted November 15, 2002.

This article has been cited by other articles:

				G. Beadle, P. Baade, and L. Fritschi Acute myeloid leukemia after breast cancer: a population-based comparison with hematological malignancies and other cancers Ann. Onc., July 22, 2008; (2008) mdn530v1. [Abstract] [Full Text] [PDF]

				C. Tarella, M. Zanni, M. Magni, F. Benedetti, C. Patti, T. Barbui, A. Pileri, M. Boccadoro, F. Ciceri, A. Gallamini, et al. Rituximab Improves the Efficacy of High-Dose Chemotherapy With Autograft for High-Risk Follicular and Diffuse Large B-Cell Lymphoma: A Multicenter Gruppo Italiano Terapie Innnovative nei Linfomi Survey J. Clin. Oncol., July 1, 2008; 26(19): 3166 - 3175. [Abstract] [Full Text] [PDF]

				J. W. Friedberg, P. Cohen, L. Chen, K. S. Robinson, A. Forero-Torres, A. S. La Casce, L. E. Fayad, A. Bessudo, E. S. Camacho, M. E. Williams, et al. Bendamustine in Patients With Rituximab-Refractory Indolent and Transformed Non-Hodgkin's Lymphoma: Results From a Phase II Multicenter, Single-Agent Study J. Clin. Oncol., January 10, 2008; 26(2): 204 - 210. [Abstract] [Full Text] [PDF]

				B. Barlogie, G. Tricot, J. Haessler, F. van Rhee, M. Cottler-Fox, E. Anaissie, J. Waldron, M. Pineda-Roman, R. Thertulien, M. Zangari, et al. Cytogenetically defined myelodysplasia after melphalan-based autotransplantation for multiple myeloma linked to poor hematopoietic stem-cell mobilization: the Arkansas experience in more than 3000 patients treated since 1989 Blood, January 1, 2008; 111(1): 94 - 100. [Abstract] [Full Text] [PDF]

				G. Leone, L. Pagano, D. Ben-Yehuda, and M. T. Voso Therapy-related leukemia and myelodysplasia: susceptibility and incidence Haematologica, October 1, 2007; 92(10): 1389 - 1398. [Abstract] [Full Text] [PDF]