Document 2jrbR40b0Jr7Dz69ov0BmqBb6

Thalidomide Alone or With Dexamethasone for Previously Untreated Multiple Myeloma By Donna Weber, Kim Rankin, Maria Gavino, Kay Delasalle, and Raymond Alexanian Purpose: To evaluate the activity of thalidomide in patients with asymptomatic multiple myeloma and of thalidomide-dexamethasone in patients with previously untreated symptomatic myeloma. Patients and Methods: Twenty-eight patients with previously untreated asymptomatic myeloma were treated with thalidomide 100 to 200 mg orally (PO) at bedtime (qhs) with serial increments of 50 to 100 mg at weekly intervals, as tolerated to a maximum of 600 mg PO qhs. Forty consecutive previously untreated patients with symptomatic myeloma were also treated as above (maximum dose 400 mg) and received dexamethasone 20 mg/m2 for 4 days beginning on days 1, 9, and 17; the second and third cycles of repeated dexamethasone were begun on day 30. Both groups of patients were treated for at least 3 months. Results: The response rate was 36% for patients treated with thalidomide alone and 72% for patients treated with thalidomide-dexamethasone, the latter including complete remission in 16% of patients. The median time to remission was 4.2 months with thalidomide alone and 0.7 months with thalidomide-dexamethasone. Grade 3 toxicity included infections (nine patients) and thrombotic/embolic events (seven patients). Five deaths have occurred as a result of multiple myeloma (two patients), infection (one patient), unknown cause (one patient), and a possible thromboembolic event (one patient). Conclusion: Thalidomide alone was effective in patients with newly diagnosed myeloma. The combination with dexamethasone induced a high frequency of response, rapid onset of remission, and low incidence of serious irreversible toxicity. These observations support further studies of this promising combination for patients with newly diagnosed multiple myeloma. J Clin Oncol 21:16-19. 2003 by American Society of Clinical Oncology. A LTHOUGH MANY treatments have been useful for patients with multiple myeloma, the disease eventually relapses and becomes resistant to treatment. Combinations of multiple alkylating agents with glucocorticoids or of vincristinedoxorubicin-dexamethasone (VAD) have not improved survival in comparison with melphalan-prednisone.1 Whereas treatments such as myeloablative therapy with autologous stem cell support have improved survival for many patients, other patients achieve limited or no gain because of early death, short remission, or persistent resistant disease. Recently, Singhal et al2 described a 26% response rate for patients with resistant myeloma treated with thalidomide, as determined by 50% reduction of myeloma protein, which has been reported and confirmed by others.3-6 After confirming a response rate of approximately 25% with thalidomide alone among patients with resistant disease, we then assessed a combination of thalidomide and dexamethasone in patients with disease who were resistant to a sequence of dexamethasonecontaining regimens and single-agent thalidomide (not necessarily consecutive treatments).3,4 The observed 46% response rate in patients with resistant disease despite multiple therapies was superior, perhaps by a synergistic effect of both drugs. This response rate was also confirmed by others.7 Although the mechanism of action of thalidomide remains unclear and may be related to antiangiogenic or direct antitumor effects against myeloma, the drug is effective despite resistance to multiple standard therapies including alkylating agents, anthracycline, glucocorticoid combinations, and intensive therapy with stem cell support. Consequently, thalidomide may also be effective at diagnosis, possibly leading to improved response rate, particularly complete response (CR), and eventually impacting survival. We therefore initiated trials for previously untreated disease with thalidomide alone for asymptomatic patients, and in combination with repeated dexamethasone for symptomatic disease. Based on our prior experience in patients with resistant disease, we postulated that the combination of thalidomide-dexamethasone would induce rapid reduction of tumor mass, allowing prompt collection of autologous blood stem cells with granulocyte-colony-stimulating factor alone. Therefore, in eligible patients, treatment with thalidomidedexamethasone was followed by intensive therapy with autologous stem cell support in an attempt to achieve maximum reduction of tumor burden. PATIENTS AND METHODS From the The University of Texas MD Anderson Cancer Center, Houston, TX. Submitted March 27, 2002; accepted September 6, 2002. Supported by the myeloma research funds established by Hill and Knowlton (in honor of William Max Watson) and Kay Laro. Thalidomide provided by Celgene Corporation, Warren, NJ. Address reprint requests to Donna Weber, MD, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 429, Houston, TX 77030, email: dmweber@mdanderson.org. 2003 by American Society of Clinical Oncology. 0732-183X/03/2101-16/$20.00 Patients and Drug Regimens Between May 1999 and September 2001 we defined the role of thalidomide alone (provided by Celgene Corporation, Warren, NJ) in 28 consecutive patients with asymptomatic multiple myeloma. Only those patients at high risk for progression were eligible based on previously described combinations of risk factors (lytic bone lesion, IgA type, monoclonal protein 3.0 g/dL, Bence Jones protein [BJP] 50 mg/d, or abnormal magnetic resonance imaging).8 High-risk disease was defined by lytic bone lesions or two risk factors, where the median time to progression was approximately 18 months. The initial dose of thalidomide was planned at 200 mg PO qhs, with serial increments of 50 to 200 mg at 1- to 2-week intervals 16 Downloaded from jco.ascopubs.org on September 8, 2010. FJooruprenarsloonfaCl luinsiecaolnOlyn.cNoologoyth,eVrouls2e1s, wNitoho1u(tJapneurmariys)s,io2n0.03: pp 16-19 Copyright 2003 American Society of Clinical Oncology. All rights reserved. DOI: 10.1200/JCO.2003.03.139 THALIDOMIDE FOR UNTREATED MULTIPLE MYELOMA 17 in absence of serious toxicity, to a maximum of 600 mg PO qhs: two patients Table 1. Patient Characteristics (median and range) were started at 100 mg, two patients at 150 mg, and 24 patients at 200 mg. With moderate side effects, the thalidomide dose was reduced by 50- to 100-mg decrements until an acceptable level was reached and maintained. For severe toxicity, thalidomide was held and restarted using the aforementioned decrements after the side effect had improved. The median maximum dose achieved was 400 mg (one patient received a maximum dose of 150 mg, nine patients 200 mg, three patients 300 mg, 10 patients 400 mg, two patients 500 mg, and three patients 600 mg). Between July 2000 and July 2001, we also treated 40 consecutive, previously untreated patients with symptomatic multiple myeloma with thalidomide in a similar schedule, but with a maximum dose of 400 mg and combined with dexamethasone 20 mg/m2 for 4 days beginning on days 1, 9, and 17. Whereas the initial thalidomide dose was planned to be 100 mg, escalated weekly by 100 mg increments to 400 mg, 28 patients started at 100 mg, seven patients at 150 mg, and five patients at 200 mg. Only patients with Thalidomide (n 28) Tumor mass (no. of patients) Low Intermediate Hemoglobin (g/dL) Monoclonal serum paraprotein (g/dL) Marrow plasmacytosis (%) 2microglobulin (mg/L) Bence Jones only (no.) 24 4 11.9 (9.2-14.4) 3.2 (0.9-6.1) 22 (5-52) 3.2 (1.4-12.9) 1 Thalidomide-Dexamethasone (n 40) 20 20 11.6 (8.6-14.8) 3.0 (0.2-9.6) 35 (0-85) 3.9 (1.6-15.7) 9 low or intermediate tumor mass were eligible. Second cycles of repeated dexamethasone began on day 30, as did third cycles in 10 patients whose myeloma had improved but was not in remission. The median maximum dose of thalidomide received was 200 mg (three patients received a maximum dose of 150 mg, 18 patients received 200 mg, one patient 250 mg, 10 patients 300 mg, and eight patients 400 mg). Treatment with the combination was continued for at least 3 months (unless there was disease progression) until the earliest occurance of maximum plateau of myeloma protein reduction, autologous transplant-supported intensification, other treatments (VAD, PS-341) for resistant disease, or unacceptable toxicities (thrombotic events in two patients occurred prior to more effective prophylaxis with therapeutic doses of coumadin). For responding patients not in CR, and for those not proceeding to myeloablative therapy, thalidomide alone response. This included 140 patients treated between June 1997 and May 2000 with dexamethasone alone (20 mg/m2 PO qd on days 1 to 4, 9 to 12, and 17 to 20 and repeated at 4 to 5 weeks with subsequent maintenance days 1 to 4, every 4 weeks), 170 patients treated between January 1992 and August 1997 with melphalan-dexamethasone (melphalan 8 mg/m2/d PO for 4 days and same regimen of dexamethasone every 5 weeks), and 128 patients treated between December 1984 and October 1993 with VAD (vincristine 0.4 mg/d and doxorubicin 9 mg/m2/d by continuous infusion IV over 24 hours for 4 days with dexamethasone as above repeated every 4 to 5 weeks).12,13,14 RESULTS was continued in a maximum dose with tolerable side effects alone or with dexamethasone for 4 days each month. Patients in CR received at least 4 Frequency of Remission months of thalidomide-dexamethasone and then were observed without maintenance therapy. For both trials, more than 80% of patients received thalidomide in an average daily dose of 100 to 200 mg each evening during the initial 2 months of therapy. Disease remission was noted in 10 patients (36%) treated with thalidomide alone and in 29 patients (72%) treated with thalidomide-dexamethasone. Five patients treated with the combina- The first 24 patients who received the combination were given coumadin tion achieved CR (16%) based on disappearance of serum 1 mg PO daily (qd) for prophylaxis of deep venous thrombosis (DVT). After it was clear that this was not preventive, therapeutic doses of prophylactic anticoagulation with coumadin or low molecular weight heparin were given to the remaining 16 patients. These treatments were conducted after approval by our institutional review board and in accordance with an assurance filed myeloma protein and, if present, urine paraprotein by immunofixation; two of nine additional patients with only Bence Jones protein had complete disappearance by urine immunofixation. Two patients achieved CR by immunofixation, but follow-up with and approved by the Department of Health and Human Services. All marrow was not available at the time of this report. Five patients patients provided written informed consent and were observed on the Celgene system for thalidomide education and prescribing safety (STEPS) program. Table 1 lists patient characteristics before therapy. included as PRs had reductions of serum paraprotein by 75% (median pretreatment serum level of 3.6 g/dL, median posttreatment value of 0.8 g/dL) but did not achieve 75% reduction of Staging and Clinical Response The diagnosis of multiple myeloma was based on standard criteria: all showed bone marrow plasmacytosis of more than 15% except for 16 patients who had biopsy-proven plasmacytomas and multiple lytic bone lesions, all patients showed a monoclonal globulin on serum or urine electrophoresis, minimal BJP ( 200 mg/d; median pretreatment BJP 0.021 g/d, median posttreatment BJP 0.012 g/d). All responding patients showed relief of bone pain, correction of anemia when present, and in 25 patients with posttreatment marrow available, reduction of marrow plasmacytosis to less than 10%. Two responding and 28% of asymptomatic patients and 80% of symptomatic patients had patients required vertebroplasty for relief of persistent back pain. lytic bone lesions. Plasma cell tumor mass was defined in each patient as high, intermediate, or low.9,10 High tumor mass was defined by either corrected serum calcium more than 11.5 mg/dL or hemoglobin less than 8.5 g/dL, and patients with such advanced disease were excluded. Low tumor mass required normal serum calcium, hemoglobin more than 10.5 g/dL, and The median time to remission for patients responsive to thalidomide alone was 4.2 months, in contrast to 0.7 months for those who received thalidomide-dexamethasone. For patients responsive to combination therapy, 86% of patients were in serum myeloma protein less than 4.5 g/dL. All other patients were classified remission within 2 months; the median time to complete remis- as intermediate tumor mass. Thus, as summarized in Table 1, asymptomatic patients treated with thalidomide alone had less advanced disease than symptomatic patients given thalidomide-dexamethasone. Partial response (PR) was defined as reductions by 75% of serum and/or urine myeloma protein synthesis on two measurements taken at least 1 month sion was 2.3 months (range, 1.6 to 2.9 months). No clinical or laboratory features, including 2-microglobulin and paraprotein level, correlated with response. At a median follow-up of 25 and 9 months, respectively, relapse had occurred in three patients on apart; CR required disappearance of serum and urine myeloma protein by thalidomide alone after 4, 12, and 21 months, and in one patient immunofixation and less than 5% plasma cells in bone marrow.11 Patients with complete disappearance of BJP in the absence of a serum myeloma protein were not included in the determination of CR. Relapse was defined on the combination, so it is premature to assess remission duration and survival. by earliest of a 20% increase of myeloma protein from nadir value, new lytic Autologous Blood Stem Cells bone lesions, or hypercalcemia. Using the saDmoewcnrilotearidae, dwefroevmaljucaote.adsocuorppurbevsi.oourgs torinalSs efoprtesimmbilear g8r,o2u0p1s 0. ForEpleigrsiobnilaitlyusfeoronhlyig. hN-odoostheemr ueslepshawliathnousut ppepromrtiessdiobny. autologous of previously untreated patients fCoropfyrerqiguhetncy20an0d3 Aspmeeedricoafn mSoyeclioemtyaof CslitneimcalcOelnlscorleoqguyi.rAedll rbigehintsgrelessesrvtehda.n 70 years of age, having good 18 WEBER ET AL health and performance, having adequate insurance support, and Table 2. Number and Percentage of Patients With Side Effects having the capacity to live near our center for 2 months. Twenty-one patients treated with thalidomide-dexamethasone proceeded to stem cell collection after 1.5 to 9.8 months Side Effect Thalidomide (%) No. of Patients % ThalidomideDexamethasone No. of Patients % (median, 5.5 months). Collection was rapid (median, 2 days) and efficient (median, 7.7 106 CD 34 cells/kg) with neupogen alone in all but two patients who received chemo- therapy priming. Total no. of patients Constipation Neurologic Numbness/tingling Unsteadiness 28 40 19 68 22 55 19 68 20 50 12 43 5 13 Toxicity Tremors Fatigue 10 36 12 30 11 39 22 55 Although side effects with thalidomide were frequent, they Rash/dry skin were nearly always mild, short term, and reversible. One patient Edema Infections 17 61 22 55 7 25 14 35 4 14 5 13 on the combination died unexpectedly of possible pulmonary Thrombotic/embolic events 1 4 6 15 embolus less than 1 week after thalidomide-dexamethasone was discontinued. Side effects noted most frequently are summarized in Table 2. Virtually all side effects were of grade 1 or 2 degrees; grade 3 toxicities included infections (nine patients), thrombotic/ further characterization of coagulation factors at baseline and embolic events (six patients) despite low-dose coumadin pro- during treatment and the preventive role of therapeutic antico- phylaxis and in one asymptomatic patient, constipation (one agulation in high-risk patients should be addressed. Whereas patient), and fatigue (one patient). No patient experienced grade most other side effects were mild and reversible with dose 2 neutropenia or thrombocytopenia. Most patients treated with reduction or temporary cessation of thalidomide, neuropathy was dexamethasone developed mild, reversible cushingoid features. rarely irreversible. These uncommon side effects should not Hospitalization for infection was required for three patients on detract from the higher response rate of thalidomide-dexametha- thalidomide alone and for five patients on thalidomide-dexam- sone with little or no serious toxicity in the majority of patients. ethasone because of pneumonia in seven patients and cellu- The newer immunomodulatory derivatives (IMiDs) of thalido- litis in one patient. Five deaths have occurred, two patients mide, which seem more potent in vitro, may avoid or reduce who received thalidomide alone (both of progressive my- serious side effects such as neuropathy while retaining signifi- eloma) and three patients after combination therapy (one of cant antitumor effects.17,18 infection, one of unknown cause after discontinuation of The potential value of thalidomide alone as prophylactic thalidomide, and one of postoperative complications after therapy for asymptomatic myeloma remains unclear. While the spinal surgery who showed a preoperative cardiac thrombus 1 time to disease progression may be prolonged for those with week after stopping thalidomide). responsive disease, premature and long-term therapy in asymp- DISCUSSION tomatic patients may contribute to later drug resistance when the more effective thalidomide-dexamethasone combination may be Thalidomide produced remissions in 36% of patients with necessary for disease progression. In some asymptomatic pa- previously untreated asymptomatic multiple myeloma, similar to tients who experience neuropathy, prolonged therapy with tha- the frequency observed in patients with resistant disease.2,3 This lidomide may also prevent future use of this and other potentially effect was similar to that of other single agents, such as neuropathic drugs (thalidomide, vincristine, PS-341) when the melphalan and high-dose dexamethasone, supporting the disease becomes symptomatic. Randomized controlled studies potential for more frequent control with combinations of these that include data on later sensitivity to induction therapy for drugs that act by different mechanisms. Furthermore, our symptomatic disease, remission time, and late side effects are results with thalidomide as a single agent were nearly required to clarify these questions. identical to those reported by Rajkumar et al,15 who observed The mechanism of myeloma control with thalidomide remains six responses in 16 similar patients (38%) using doses ranging unclear. Thalidomide inhibits angiogenesis in vitro, induces from 200 to 800 mg/d. apoptosis, and reduces the high levels of certain angiogenesis Among previously untreated patients with more advanced and factors, such as VEGF and bFGF, that are present in patients symptomatic disease, the combination of thalidomide-dexam- with multiple myeloma and other hematologic malignancies.19-21 ethasone doubled the response rate and induced remissions more However, no correlation between marrow vascularization and rapidly, indicating that the combination of both drugs was clinical response of myeloma has been noted.2 Recently, a direct superior to either alone. This experience conforms well with the effect inducing G1 growth arrest and subsequent apoptosis, superiority of the same combination among patients with resis- perhaps through interleukin-6 downregulation, has been demon- tant disease, even those who had been resistant to sequential strated in vitro.21 trials of dexamethasone and thalidomide separately.4,5 Our Regardless of the mechanism, thalidomide-dexamethasone results with this combination were also similar to the response provided a simple, relatively safe, and remarkably effective rate of 64% observed by Rajkumar et al16 in 50 comparable, primary treatment for nearly three-fourths of patients with newly diagnosed patients. previously untreated symptomatic multiple myeloma. Results Given the 25% incidence of thrombotic/embolic events in the seemed superior to previous experiences in comparable patients first 24 patients treated with thalidomide-dexamethasone and with dexamethasone alone, or in combination with melphalan, warfarin 1 mDogwPnOloaqdde,dafrnodmthjceo.aabssceonpcuebso.ofrgthoronmSbepotteicmebveren8t,s20in10. Faonrdpesrismoinlaarl utsoe tohnolyse. NwoioththeVrAusDe,s bwuitthowuitthpefremwisesriocno. mplications the subsequent 16 patients givCeonptyhriegrhatpeu2t0ic03doAsmeesriocfanwSaorfcaieritny,of C(lTinaicballeO3n)c.oAlogcyu.rArellnrtigrhatnsdreosmeirzveedd.comparison of dexamethasone THALIDOMIDE FOR UNTREATED MULTIPLE MYELOMA Table 3. Response Rates in Previously Untreated Patients With Low or Intermediate Tumor Mass Melphalan-dexamethasone13 Dexamethasone14 Vincristine- doxorubicin-dexamethasone12 Thalidomide Thalidomide-dexamethasone No. of Patients 170 140 128 28 40 Early Death 2 months 2 1 3 2 3 Response Rate (%) 51 47 66 36 72 CR (%)* 8 7 13 0 16 Myeloma Halving (median months) 0.3 0.4 0.5 1.5 0.3 *Complete remission (CR) defined by disappearance of serum myeloma protein by immunofixation. For patients with only Bence Jones protein, urine disappearance for historical data was assessed primarily by electrophoresis and did not usually include immunofixation studies. Therefore, patients with only Bence Jones protein were excluded for assessment of CR in both historical and current data. Among responding patients with serum myeloma protein. 19 versus thalidomide-dexamethasone by the Eastern Cooperative Oncology Group should clarify this question. We also showed a major effect against myeloma with low and acceptably tolerated doses of thalidomide, especially when combined with dexamethasone, so that the frequent and severe side effects observed with higher doses were usually avoided.2,3,15,16 With therapeutic doses of coumadin to maintain international normalized ratio between 2.0 to 3.0, we have observed only rare occurrences of deep venous thrombosis in recently treated patients. Because the combined effect with dexamethasone poses a risk of gastrointestinal bleeding (or bleeding from other causes), future study of risk factors for hypercoagulability may identify more clearly those patients who should receive therapeutic anticoagulation. Since the complete remission rate remained low, this program of induction therapy should be followed by intensive consolidation therapy in most eligible patients to achieve complete remission that translates into longer survival.22 Because serious irreversible toxicity was rare and myelosuppression virtually nonexistent, patients with responsive disease proceeded early to stem cell collection with granulocyte colony-stimulating factor alone, followed by high-dose melphalan therapy with stem cell support. Considering the ease of oral administration, the infrequency of serious irreversible side effects, and the rapidity of response permitting early stem cell collection, the combination of thalidomide-dexamethasone may well represent the treatment of choice for myeloma patients with disease of low or intermediate tumor mass. ACKNOWLEDGMENT We thank Rose Guevara for secretarial support. REFERENCES 1. Myeloma Trialists Collaborative Group: Combination chemotherapy versus melphalan plus prednisone as treatment for multiple myeloma: An overview of 6,633 patients from 27 randomized trials. J Clin Oncol 16:3832-3842, 1998 2. Singhal S, Mehta J, Desikan R, et al: Antitumor activity of thalidomide in refractory multiple myeloma. N Engl J Med 341:1565-1571, 1999 3. Alexanian R, Weber D: Thalidomide for resistant and relapsing myeloma. Semin Hematol 37:22-25, 2000 4. Weber DM, Rankin K, Delasalle K, et al: Thalidomide alone and in combination for previously untreated multiple myeloma. VIII International Myeloma Workshop, Banff, Canada 109:2001 (abstract S66) 5. Rajkumar SV: Angiogenesis and anti-angiogenic therapy with thalidomide for myeloma, Banff, Canada. VIII International Myeloma Workshop, p112, 2001 (abstract S68) 6. Durie BGM, Stepan DE: Efficacy of low dose thalidomide in multiple myeloma. Blood 94:316a, 1999 (abstract) 7. Dimopoulos MA, Zervas K, Kouvatseas G, et al: Thalidomide and dexamethasone combination for refractory myeloma. Ann Oncol 12:991995, 2001 8. Weber DM, Dimopoulos M, Moulopoulos LA, et al: Prognostic features of asymptomatic multiple myeloma. Br J Haematol 97:810-814, 1997 9. Durie B, Salmon S: A clinical staging system for multiple myeloma. Cancer 36:842-847, 1975 10. Alexanian R, Balcerzak S, Bonnet J, et al: Prognostic factors in multiple myeloma. Cancer 36:1192-1201, 1975 11. McLaughlin P, Alexanian R: Myeloma protein kinetics following chemotherapy. Blood 60:851-855, 1982 12. Alexanian R, Barlogie B, Tucker S: VAD-based regimens as primary treatment for multiple myeloma. Am J Hematol 33:86, 1990 13. Alexanian R, Weber D, Dimopoulos M, et al: Randomized trial of -interferon or dexamethasone as maintenance treatment for multiple myeloma. Am J Hematol 65:204-209, 2000 14. Alexanian R, Dimopoulos M, Delasalle K, et al: Primary dexamethasone treatment of multiple myeloma. Blood 80:887-890, 1992 15. Rajkumar S, Dispenzieri A, Fonseca R, et al: Thalidomide for previously untreated indolent or smoldering multiple myeloma. Leukemia 15:1274-1276, 2001 16. Rajkumar V, Hayman S, Gertz M, et al: Combination therapy with thalidomide plus dexamethasone for newly diagnosed multiple myeloma. Blood 98:849a, 2001 (abstract) 17. Richardson P, Schlossman R, Hideshima T, et al: A phase I study of oral CC5013, an immunomodulatory thalidomide derivative in patients with relapsed and refractory multiple myeloma. Blood 98:775a, 2001 (abstract) 18. Zangari M, Tricot G, Zeldis J, et al: Results of phase I study of CC5013 for the treatment of multiple myeloma patients who relapse after high dose chemotherapy. Blood 98:775a, 2001 (abstract) 19. D'Amato RJ, Loughnan MS, Flynn E, et al: Thalidomide is an inhibitor of angiogenesis. Proc Natl Acad Sci 91:4082-4085, 1994 20. Kenyon BM, Brown F, D'Amato RJ: Effects of thalidomide and metabolites in a mouse corneal model of neovascularization. Exp Eye Res 64:971-978, 1997 21. Hideshima T, Chauhan D, Shima Y, et al: Thalidomide and its analogs overcome drug resistance to human multiple myeloma cells to conventional therapy. Blood 96:2943-2950, 2000 22. Alexanian R, Weber D, Giralt S, et al: Impact of complete remission with intensive therapy in patients with responsive multiple myeloma. Bone Marrow Transplant 27:1037-1043, 2001. Downloaded from jco.ascopubs.org on September 8, 2010. For personal use only. No other uses without permission. Copyright 2003 American Society of Clinical Oncology. All rights reserved.