|
|
 Previous Article | Table of Contents | Next Article 
Blood, 1 March 2003, Vol. 101, No. 5, pp. 1666-1667
Is the cure worse than the disease?
No, it isn't. The worrisome observations of the
delayed development of t-MDS/AML in some lymphoma patients
after HDC and AHSCT should not obscure the overall success of
this procedure in selected patients. That said, t-MDS/AML is
the serious nonrelapse problem in these patients, and
efforts to identify risk factors that would reliably identify patients
at high risk of developing t-MDS/AML would be beneficial, both in
focusing increased study of such patients and potentially, to develop
therapies to decrease the impact of this (currently) very lethal complication. The report of Metayer and colleagues in this issue (page 2015) adds to
a growing, if inconsistent, literature regarding risk factors. Using
detailed data from 12 institutions in the Autologous Blood and Marrow
Transplant Registry, they performed a case-control study using 56 patients who developed t-MDS/AML and 168 matched control patients from
a larger group of 2739 lymphoma patients receiving transplants
at these centers over 6 years. The cumulative incidence of t-MDS/AML
was 3.7% at 7 years. (For unclear reasons, this is lower than reported
in other published series, and it would be unrealistic to state that
the "true incidence" is known with certainty.) In multivariate
analysis, exposure to certain alkylating agents with a known leukemogen
pedigree (especially nitrogen mustard and chlorambucil, with notation
of a dose effect), as well as conditioning regimens that included TBI,
but only at doses of 1320 cGy, were associated with a statistically
significant increased risk of developing t-MDS/AML. The use of
reconstituting cells from either steady-state or mobilized blood was
associated with an increased risk as well, albeit a nonsignificant one.
Thus, both pretransplantation and transplantation-related
factors were implicated. The authors are to be commended for this effort, which clearly gives
additional direction to strategies or techniques that might decrease
this risk. That said, significant limitations of this study must be
acknowledged; to a large degree, these limitations are generic to
studies evaluating "late events" that require prolonged follow-up
for delineation in a field in which evolutionary therapy changes are
routine. For example, this analysis includes both Hodgkin and the
non-Hodgkin lymphomas, whereas other series focused on one or the
other, in the main. This obviously may have implications on the
specific therapy given and the contribution to the MDS/AML. Also, the
routine use of the nitrogen mustard, chlorambucil, and perhaps other
suspect drugs has been greatly curtailed in the past decade,
as the CHOP and ABVD regimens have become standards of care for many
non-Hodgkin and Hodgkin lymphoma patients, respectively. In the same
context, the use of local radiotherapy (a potential contributor
exonerated in this analysis) as either primary therapy or as an adjunct
to the above or similar regimens appears to be declining. Also, the use
of TBI is not considered critical and may also be declining in use.
Thus, the warning about these specific agents arrives as the danger is receding. In contrast, the issue of potential increased risk from the use of
reconstituting cells obtained from the blood is more complex; this
technique has almost fully supplanted marrow harvest in
autotransplantation and has been the standard of care for nearly a
decade. Since the main advantages of using mobilized blood
reconstituting cells are usually not reflected onto major outcome
parameters, further delineation of this point is critical to ensure we
are not increasing t-MDS/AML by the use of this technique. What lessons can be learned from this experience? First, a prolonged
period of observation of patients is necessary to delineate late
complications such as t-MDS/AML. Next, before we breathe a sigh of
relief at the passing of the routine use of agents such as the nitrogen
mustard and chlorambucil, we may get another chance to revisit this
situation as long-term effects of increased use of topoisomerase-II
inhibitors and possibly radioimmunotherapies are assessed. In addition,
we will need to become more facile at being able to identify high-risk
patients on an individual rather than a population basis;
conventional cytogenetics are clearly inadequate, but other techniques
(perhaps including FISH, SKY, or selective gene array
studies) may be more suitable. Finally, however, we should again take note of the potency of the
HDC/AHSCT technique in lymphoma patients: thousands of such patients
are alive today because of this procedure; moreover, refinements of
HDC/AHSCT as well as other strategies (eg, the use of
reduced-intensity allografts)may be expected to further improve
results. Nonetheless, we must be vigilant to the development of
t-MDS/AML using our newer techniques and develop methods to identify,
prevent, and/or treat such patients early in their course.
Gordon L. Phillips, II
University of Rochester Medical Center
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
|
|
