Issues in the Management of Cancer-Related Thrombocytopenia

Cancer-related thrombocytopenia is a clinical problem.Unfortunately, the qualitative nature and quantitative extent of the problem hasbeen poorly defined to date. Without knowing these two parameters, therisk/benefit ratio of any management option for cancer-related thrombocytopeniais impossible to calculate accurately. Drs. Goodnough and DiPersio have done anexcellent job of delineating many of the potential risks of managing theproblems associated with platelet transfusions.

Although transfusion-related viral infections are high on the list ofpotential toxicities, data suggest that, with the exception of cytomegalovirus (CMV),these occur infrequently. Transfusion-related bacterial infections are not oftenconsidered to be a serious clinical problem. However, if one considers that therisk of sepsis with apheresis platelets is 1 in 12,000 and the mortality rateassociated with this sepsis is 26% (as the authors suggest), then patientsreceiving 10 platelet transfusions over the course of their cancer therapy havea 1 in 4,800 risk of death from transfusion. That is not a negligible risk,especially in light of the number of patients who undergo myelosuppressivetreatments for cancer.

Platelet Refractorinessand Other Limitations

Refractoriness to platelet transfusion due to alloimmunization is listed as apotential problem with the use of transfusion therapy, but its clinicalsignificance may be overestimated given the lack of clinical improvement seen inpatients in whom alloimmunization was prevented with the use of leukoreducedplatelets.[1]

The financial expense of platelet transfusion therapy was briefly mentioned,but deserves more detailed consideration. In calculating its expense, the costof all factors associated with the collection of platelets for transfusion(including, for example, time off from work for the platelet donors) should beincluded. These factors are often not accounted for in such analyses.

Another potential downside to platelet transfusions that is not oftenmentioned in this type of review is their potential contribution to thecomplications that occur after administration of chemotherapy. In blood andmarrow transplant patients, increased use of prophylactic platelet transfusionis associated with subsequent development of hepatic veno-occlusive disease,pulmonary dysfunction, and central nervous system (CNS) dysfunction.[2] Althoughthis association does not confirm a causative role for platelet transfusions inthe generation of CNS, pulmonary, or hepatic complications, data from animalmodels and other clinical disorders suggest this possibility.[3-5]

Level of Supportive Evidence

The authors point out that prophylactic platelet transfusion is a commonpractice, but they do not discuss the basis for this practice. Given the actualand potential risks associated with platelet transfusion (CMV infection,bacterial sepsis, cost, alloimmunization, and possibly transplant-related organdysfunction), one would suppose that its prophylactic use in cancer patientswould be based on appropriately designed clinical studies demonstrating improvedclinical outcomes (with prophylactic transfusions vs transfusions to treatbleeding complications). Unfortunately, that is not the case.

As pointed out in another authoritative review, this practice is based onobservational studies (level 4 evidence) of patients treated prior to 1976.[6]Recommendations for the use of prophylactic platelet transfusion in general havebeen based on this level 4 evidence and "expert consensus" rather thanon clinical outcomes of studies in patients treated in contemporary settings.[6]In the studies (cited by the authors) of patient outcomes after prophylacticplatelet transfusions, the only hemorrhagic deaths were in patients whoseplatelet counts exceeded 30,000/µL at the time of death, raising the possibilitythat strategies of prophylactic platelet transfusion based solely on a plateletcount threshold may not eliminate bleeding deaths.

Current recommendations from authoritative sources suggest that the use ofprophylactic platelet transfusion be individualized for specific clinicalsettings and that factors beyond the platelet count be considered in thedecision.[6] The paucity of data supporting the basic concept of prophylacticplatelet transfusions should be kept in mind when discussing the threshold forprophylactic transfusion and the dose of platelets to be administered. Evenlevel 1 evidence and grade A recommendations that "fine tune" the useof an unproven premise are of little clinical value.

Limited Scope

The limited basis for the use of prophylactic platelet transfusion shouldalso be kept in mind when discussing new "pharmacologic options to managecancer-related thrombocytopenia and prevent its attendant morbidity andmortality." As stated by other reviewers, "Fatal hemorrhage is now anunusual event, even in patients with bone marrow failure or in those receivingintensive antineoplastic therapy."[6]

The contribution of thrombocytopenia to the morbidity and mortality of cancerpatients in the contemporary practice of clinical oncology is currently unknown.We cannot assume that improvement of the platelet count, either by transfusionor medication, will improve patient outcome. Indeed, in the only clinical trialof one of the new agents discussed by the authors, patient outcome was no betteramong those who received the active drug than among those who received placebo,save for the platelet count.[7] Hopefully, the history of prophylactic platelettransfusions will not be repeated with these new agents. Before they achievewidespread popularity, they should be shown to produce better clinical outcomesthan placebo, not just a cosmetic improvement in the platelet count.

Conclusions

While it is probably true, as the authors suggest, that "thrombocytopeniaremains a significant clinical problem for patients with cancer" it is alsotrue that the nature and degree of this problem has been poorly defined. Thecurrent therapy for thrombocytopenia—prophylactic platelet transfusion—hasreceived widespread acceptance based on marginal proof of its safety andefficacy. The emergence of new pharmaceuticals designed to replace thetime-honored practice of prophylactic transfusion will provide anotheropportunity to understand the role of platelets in the overall outcome ofpatients with cancer. Hopefully, clinical oncology as a specialty is ready totake advantage of this opportunity.

References:

1. Trial to Reduce Alloimmunization to Platelets Study Group: Leukocytereduction and ultraviolet B irradiation of platelets to prevent alloimmunizationand refractoriness to platelet transfusions. N Engl J Med 337:1861-1869, 1997.

2. Gordon B, Tarantolo S, Ruby E, et al: Increased platelet transfusionrequirement is associated with multiple organ dysfunction in patients undergoinghematopoietic stem cell transplantation. Bone Marrow Transplant 22:999-1003,1998.

3. Heffner J, Sahn S, Repine JE: The role of platelets in adult respiratorydistress syndrome: Culprits or bystanders? Am Rev Respir Dis 135:482-492, 1987.

4. Mannel D, Grau GE: Role of platelet adhesion in homeostasis andimmunopathology. Mol Pathol 50:175-185, 1997.

5. Lou J, Lucas R, Grau GE: Pathogenesis of cerebral malaria: Recentexperimental data and possible applications for humans. Clin Microbiol Rev14:810-820, 2001.

6. Schiffer CA, Anderson KC, Bennet CL, et al: Platelet transfusion forpatients with cancer: Clinical practice gudelines of the American Society ofClinical Oncology. J Clin Oncol 19:1519-1538, 2001.

7. Tepler I, Elias L, Smith JW 2nd, et al: A randomized placebo-controlledtrial of recombinant human interleukin-11 in cancer patients with severethrombocytopenia due to chemotherapy. Blood 87:3607-3614, 1996.