To find a new cure for a disease, or at least to significantly prolong patient survival, requires a bit of insight and a lot of luck. Recent history in the hematologic malignancies has demonstrated the important role played in this process by an understanding of the biology of the disease, which can either serve as the foundation for the development of a new drug (eg, imatininb (Gleevec) in chronic myelogenous leukemia) or help subsequently explain an agent's efficacy (eg, all-trans-retinoic acid [ATRA] in acute promyelocytic leukemia and the resulting identification of the PML-RARα fusion gene from the balanced translocation of t[15;17]). A notable exception to this rule of thumb is rituximab(Drug information on rituximab) (Rituxan), which binds to CD20 and is effective in CD20-positive malignancies; our knowledge of how the antibody actually works—beyond simply antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC)—and why it sometimes doesn't, is inadequate.
One consequence of the availability of highly successful therapies is that the number of existing disease entities increases. Book chapters on hematologic malignanices once consisted simply of “lymphoma” and “leukemia;” these subsequently multiplied to include acute and chronic leukemias, Hodgkin and non-Hodgkin lymphomas. Coverage of chronic myelogenous leukemia might well now be divided into a chapter on the T315I-negative disease that responds nicely to tyrosine kinases, and another on the T315I-positive variety that does not. There is the acute leukemia that responds to ATRA, which we call now call acute promyelocytic leukemia, and the others that do not. There are activated B-cell diffuse large-cell lymphomas (ABC) as well as the germinal center sub-type (GCB), which have different outcomes and different responses to a variety of therapeutic agents. It is likely that all of the hematologic malignancies, and solid tumors as well, will eventually have new names determined by molecular and genetic characteristics and/or responsiveness to treatment. Thus, when a patient doesn't respond to a standard therapy for a disease, this does not mean that the treatment is ineffective; it simply means that the patient has another biologic variant of that histologic entity.
In this paper, Mougalian and O'Brien review the published data on prognostic factors in chronic lymphocytic leukemia (CLL)—eg, ZAP-70, thymidine kinase, FISH results, unmutated immunoglobulin heavy chain variable (IgVH) gene status, and others—and how those features predict an adverse outcome.
Nevertheless, a number of important questions remain unanswered regarding the use of these factors by the clinicians who actually care for these patients. Why is there such heterogeneity in outcome even within risk groups? Some patients with a 17p deletion may do just fine for years, while others with even more favorable cytogenetics succumb rapidly to their disease. Which combination of factors predicts clinical course more strongly than the others?
The question most often asked by physicians is how do we utilize this information for treatment decisions?[1] It is tempting to assume that pre-emptive intervention in patients who have early-stage disease but who also have potentially adverse prognostic factors should improve patient outcomes; however, studies to support that conclusion have been difficult to complete. Thus, patients with CLL should not be treated according to standard definitions until a benefit for doing so has been demonstrated in clinical trials.[2]
Clinical researchers have expended considerable financial and patient capital on shuffling drugs into various combinations, with limited success.[3] After the dramatic benefit from rituximab (Rituxan) combined with CHOP (cyclophosphamide, doxorubicin(Drug information on doxorubicin), vincristine, and prednisone(Drug information on prednisone)) (R-CHOP) in diffuse large B-cell lymphoma (DLBCL), numerous studies have examined whether further intensification of the chemotherapy might improve outcomes in the 40% of patients who fail to benefit from R-CHOP (R-CHOP-resistant DLBCL). These attempts have summarily failed. While fludarabine-rituximab, with or without cyclophosphamide(Drug information on cyclophosphamide), or bendamustine with rituximab, all appear to be associated with improved outcomes in CLL compared with single-agent therapy, the disease remains incurable. Adding an alkylating agent in patients with an 11q deletion may modestly enhance therapeutic efficacy, but there is considerable room for improvement. We are faced with that persistent holy grail—a drug or therapy that overcomes the stigmata of the 17p deletion; alemtuzumab(Drug information on alemtuzumab) (Campath) may be selectively effective for patients with a 17p deletion, but the results are still dismal.[4]
Fortunately, a new collection of novel agents has recently entered clinical trials in CLL, including some that target the PI-3 kinase pathway (eg, CAL-101) or Bruton's tyrosine kinase (eg, PCI-32765), and others that induce apoptosis. Both CAL-101 and PCI-32765 have demonstrated promising activity in patients with CLL and are suitable agents to combine with other effective drugs because of their excellent safety profile. In addition, early data suggest activity in patients with 17p deletions, 11q deletions, and IgVH unmutated status.[5,6]
We are not yet at a point where different cytogenetic, molecular, or immunologic abnormalities should select patients out of our clinical trials for some alternative therapy, even though such an approach is not far away. Thus, while it is tempting to evaluate every CLL patient for CD38, ZAP-70, IgVH status, and other prognostic factors, it is difficult at present to justify these extensive and expensive assessments until such time as the results will dictate a treatment approach. Resources would be better directed at improving our understanding of this disparate collection of diseases we have been calling CLL.
Financial Disclosure: The author has served as a consultant for Gilead and for Pharmacyclics
