Myeloma Dialysis-dependent renal failure in patients with ... - Nature

Bone Marrow Transplantation (2004) 33, 823–828 & 2004 Nature Publishing Group All rights reserved 0268-3369/04 $25.00

www.nature.com/bmt

Myeloma Dialysis-dependent renal failure in patients with myeloma can be reversed by high-dose myeloablative therapy and autotransplant C-K Lee, M Zangari, B Barlogie, A Fassas, F van Rhee, R Thertulien, G Talamo, F Muwalla, E Anaissie, K Hollmig and G Tricot The Myeloma Institute for Research and Therapy, The University of Arkansas for Medical Sciences, Little Rock, USA

Summary: To evaluate the role of high-dose melphalan and autologous transplant (AT) in reversing dialysis-dependent renal failure, 59 patients still on dialysis at the time of AT were analyzed. A total of 37 patients had been on dialysis p6 months. A 5-year event-free and overall survival rate of all patients after AT was 24 and 36%, respectively. Of 54 patients evaluable for renal function improvement, 13 (24%) became dialysis independent at a median of 4 months after AT (range: 1–16). Dialysis duration p6 months prior to first AT and pre-transplant creatinine clearance 410 ml/min were significant for renal function recovery: 12 of 36 (33%) p6 months vs one of 18 patients (6%) 46 months on dialysis recovered renal function; 10 of 26 (38%) with 410 ml/min vs three of 28 (11%) with p10 ml/min of creatinine clearance (both Po0.05). Quality of response after autotransplant was also significant: 12 of 31 (39%) being greater than partial remission after AT vs one of 21 patients (5%) attaining partial remission or less became independent of dialysis (Po0.05). Our data suggest that significant renal failure can be reversible and AT should be considered early in the disease course. Bone Marrow Transplantation (2004) 33, 823–828. doi:10.1038/sj.bmt.1704440 Published online 9 February 2004 Keywords: stem cell transplantation; multiple myeloma; renal failure

Renal insufficiency is not uncommon in patients with myeloma, with a 30% incidence at initial presentation increasing further as the disease progresses.1 Patients who require dialysis are known to have a poor prognosis,2 although myeloma has been known to be one of the reversible causes of the end-stage renal disease.3 Several studies have shown equivalent response rates to chemo-

Correspondence: Dr C-K Lee, The Myeloma Institute for Research and Therapy, The University of Arkansas for Medical Sciences, Slot 776, 4301 West Markham, Little Rock, AR 72205, USA; E-mail: [email protected] Received 8 July 2003; accepted 26 November 2003 Published online 9 February 2004

therapy in myeloma patients with mild to moderate renal insufficiency, compared with those who have normal renal function, and response is frequently associated with gradual improvement in creatinine clearance. Recovery from renal insufficiency appears to depend on the underlying cause of the kidney failure.2 Complete recovery was observed only in the absence of global tubular atrophy and interstitial damage, whereas cast-induced tubular obstruction by Tamm–Horsfall protein did not seem to influence the outcome.4 Partial improvement in patients with light-chain nephropathy, characterized as diffuse mesangial nodular lesion and tubulointerstitial changes, has been reported only occasionally, suggesting a possibility of response presumably related to the degree of involvement.5,6 Similar observations were made in patients with amyloid involvement of kidneys, especially after high-dose therapy and autotransplant.7 We have previously reported on the outstanding outcome of 81 patients with renal failure after high-dose melphalan and autotransplant, including 38 patients on dialysis at the time of first autotransplant.8 Of these, 21 and 31 patients achieved complete remission (CR) after the first and second autotransplant, respectively, with 48% eventfree and 55% overall survival probabilities at 3 years. On further follow-up of patients on dialysis, we have witnessed slow but persistent improvement in renal function in a limited number of patients to the extent that dialysis was no longer required. Therefore, in the present study, with an additional 21 patients on dialysis, the role of high-dose melphalan and autotransplant in reversing dialysis-dependent renal failure and potential prognostic factors for the improvement of renal function were investigated.

Patients and methods Patients To evaluate whether autotransplantation after high-dose melphalan chemotherapy could reverse renal failure and to identify factors predicting such as reversal, clinical data were retrieved on all MM patients with renal failure who were still on maintenance dialysis at the time of autotransplant (n ¼ 59) and received a single or tandem autotransplants (UARK 96012 study) between December 1995 and December 2001. Of the 86 patients enrolled on the UARK 96012, 27 with renal insufficiency (median serum creatinine 282.9 mmol/l) who did not require maintenance

Stem cell transplantation in patients with myeloma C-K Lee et al

824

dialysis were not included in the analysis. The Institutional Review Board of the University of Arkansas for Medical Sciences and the Arkansas Cancer Research Center approved the treatment protocol and an informed consent was obtained from all patients. Eligible patients had symptomatic, active multiple myeloma that was measurable. Patients had been off all previous therapy except steroids, and local radiotherapy for X4 weeks prior to the autotransplant and had fully recovered from all previous chemotherapy and radiotoxicity effects. They had adequate spirometric values of pulmonary function test X50% of predicted, diffusion capacity (DLCO) X50% of predicted, left ventricular ejection fraction 440%, direct bilirubin p34.2 mmol/l and transaminases p2 times normal unless related to myeloma and Zubrod performance status p2 unless due to bone pain. Patients with the following conditions were ineligible: serious uncontrolled infection; positivity for hepatitis B or C virus, or human immunodeficiency virus; pregnant or nursing women; and those who had undergone a priori auto- or allotransplant. Prior to autotransplant, peripheral blood stem cells (42 106 CD34 þ cells/kg) were mobilized by growth factors following chemotherapy (n ¼ 17) or dexamethasone (n ¼ 12). Growth factor alone was used for stem cell mobilization for 30 patients who, before study enrollment, had received chemotherapy (n ¼ 16), dexamethasone (n ¼ 7), or involved field radiotherapy with dexamethasone (n ¼ 3). The median number of peripheral blood stem cells collected was 12 106 CD34 þ cells/kg (range, 2.5–50). For the first 27 patients, melphalan 200 mg/m2 was administered as a single agent 2 days prior to autologous transplantation. The dose of melphalan was subsequently reduced to 140 mg/m2 (n ¼ 32) due to its significant toxicity. Stem cell infusion was preceded by dialysis 24–36 h after melphalan. Standard supportive care included intensified dialysis because of the increased catabolic state, and antibacterial, antifungal and antiviral prophylaxis, with the dose adjusted to creatinine clearances. Hemoglobin and platelet levels were maintained at X6.2 mmol/l and X20 109/l, respectively.

Definitions of response CR required disappearance of monoclonal protein in serum and urine also on immunofixation analysis, and attainment of normal bone marrow aspirate and biopsy with o1% light-chain-restricted plasma cells on flow cytometry. Patients were considered to have achieved near CR (nCR) if bone marrow findings were normal and positive immunofixation analysis of serum or urine was the only evidence of disease. PR implied a normal marrow aspirate and biopsy and at least 75% reduction from baseline serum M protein and/or, in case of Bence–Jones proteinuria, reduction to o100 mg/day. Improvement required a 50% reduction in serum paraprotein production and/or a 75% reduction in Bence–Jones protein excretion and at least a 50% reduction in bone marrow plasmacytosis from the pretreatment values. Bone Marrow Transplantation

Statistical analysis Patients were studied for response, survival, transplantrelated mortality (TRM) and off-dialysis recovery from renal failure. Toxicity was assessed according to the Common Toxicity Criteria of the National Cancer Institute.9 TRM was defined as any death due to complications related to autotransplantation within 6-months post transplant; 1 week after the last dialysis was used as the time of event for off-dialysis recovery. Cumulative incidence curves were generated to calculate the probability of responses, TRM and off-dialysis recovery, using the NCSS statistical software, version 2001.10,11 For the cumulative incidence curve of response, death before response assessment was a competing risk for response. Similarly, early death within a month post transplant before assessment for off-dialysis recovery was a competing risk for recovery. Survival analyses were performed using the product limit estimate of Kaplan–Meier method, and comparisons of survival were carried out by the log-rank statistics.12,13 Analyzed variables related to off-dialysis recovery were isotype of myeloma (kappa light chain vs lambda light chain vs others), pre-transplant response (XPR vs oPR), dose of melphalan (140 vs 200 mg/m2), transplant response (CR þ nCR vs pPR), duration of dialysis before AT (p6 vs 46 months) and creatinine clearance at AT (410 vs p10 ml/min). Prognostic factors affecting off-dialysis recovery were evaluated by univariate analysis. Multivariate analysis was not performed due to the small sample size.

Results Patient characteristics Patient characteristics are described in Table 1. Renal failure requiring dialysis was present at diagnosis in 10 (17%), within 6 months of diagnosis in 27 (46%) and after 6 months (median 13 months, range 7–72 months) of diagnosis in 22 patients (37%). All patients were still on maintenance dialysis at the time of autotransplant, for a median duration of 4 months (range, 1–26 months). The median value of serum creatinine at AT was 804.4 mmol/l (range, 371.3–1184.6 mmol/l) and of creatinine clearance 10 ml/min (range, 0–35 ml/min). Combination chemotherapy with growth factors was used for stem cell mobilization in 17 patients (29%), including VAD (vincristine, doxorubicin, dexamethasone) (n ¼ 7), DTACE (dexamethasone, thalidomide, doxorubicin, cyclophosphamide, etoposide) (n ¼ 4),14 DEP (dexamethasone, cyclophosphamide, etoposide) (n ¼ 3),15 induction chemotherapy on total therapy II (n ¼ 2)16 and cyclophosphamide þ etoposide (n ¼ 1). In all, 12 patients received dexamethasone followed by growth factors for stem cell mobilization. A total of 30 patients mobilized stem cells after growth factors only: 26 of these patients were previously treated, prior to referral to our center, with combination chemotherapy (n ¼ 16), dexamethasone alone (n ¼ 7) and involved field radiation therapy with dexamethasone (n ¼ 3). At the time of


825 Table 1

Patient characteristics

Parameter

Median

Age (years) Male:female

51

Range

Cutoff

Number of patients

28–66

(455)

17 (29%) 35:24

Isotype Kappa light chain Lambda light chain Ig A Ig G Duration of dialysis (month) Serum creatinine (mg/dl) Creatinine clearance (ml/min) Albumin (gm/dl) CRP (mg/l) Newly diagnosed Previously treated

4 9.1 10 3.9 0.46

1–26 4.2–13.4 0–35 1.6–5.0 0.09–9.8

(46) (46.5) (p10) (o3.5) (42.0)

22 11 12 14

(37%) (19%) (20%) (24%)

22 40 30 11 20 23 36

(37%) (68%) (51%) (19%) (34%) (39%) (61%)

Pre-transplant response XPR

18 (33%)

Melphalan dose for transplant 140 mg/m2 200 mg/m2

32 (54%) 27 (46%)

Autotransplant First Second

59 (100%) 23 (39%)

Table 2

Response, toxicity and off-dialysis recovery after autotransplant 1st Autotransplant (n ¼ 59)

CR+nCR TRM Toxicities (X3 grades) Off dialysis

24 7 36 8

(46%) (12%) (61%) (15%)

autologous transplantation, 18 patients (33%) were in XPR, one CR, 11 nCR and six PR. All patients received at least one AT and 23 (39%) a second AT at a median of 5 months (range, 3–11 months) after the first AT. Of the 45 patients in whom data were available for evaluation of an infiltrative process of other organs including heart, liver and peripheral nerve, eight patients had either echocardiographic findings suggestive of myocardial involvement by deposition of light chain or amyloid (n ¼ 4), or biopsyconfirmed light-chain deposition in myocardium (n ¼ 2), amyloidosis of peripheral nerve (n ¼ 1) and amyloidosis of both the myocardium and liver (n ¼ 1). Renal biopsy was available in 28 patients prior to AT, of whom 15 showed findings compatible with cast nephropathy, 10 light-chain deposition disease, two both light- and heavy-chain deposition disease and one amyloid kidney.

Response, toxicity and TRM (Table 2) Seven patients died during the first 6 months after the first AT due to TRM. Of the 52 patients evaluable for response, 24 (46%) achieved a CR (n ¼ 12) or nCR (n ¼ 12), 14 (27%) had a PR, 12 (23%) had an improvement and two (4%) had

2nd Autotransplant (n ¼ 23) Seven additional CR+nCR 4 (17%) 7 (30%) 5 (22%)

Cumulative total 31 11 43 13

(60%) (19%) (52%) (24%)

no response. In total, 23 patients received a second AT, of whom four died of TRM during the 6 months post second AT. Of these, seven patients (37%) achieved a CR (n ¼ 5) or nCR (n ¼ 2) from PR and two (11%) a PR from improvement after the second AT. Thus, 31 (60%) achieved either CR (n ¼ 17) or nCR (n ¼ 14) and nine PR (17%) after AT. The median time to absolute neutrophil count 40.5 109/l was both 11 days after the first (range, 9–37 days) and second AT (range, 8–17 days), respectively. The median time to platelet 450 109/l was 39 days (range, 9– 142 days) and 26 days (range, 12–97 days) after the first and second AT. All autotransplants were performed as an inpatient procedure, with a median duration of hospital stay of 23 days (range, 1–143 days). Of the total of 82 autotransplants, Xgrades III toxicities encountered were 24 (29%) mucositis, 14 (17%) documented sepsis, 11 (13%) metabolic encephalopathy, seven (9%) neutropenic fever, six (7%) cardiac abnormalities (three arrhythmia, two congestive heart failure, one myocardial infarct), six (7%) idiopathic pneumonia syndrome and three (4%) hepatic abnormalities. Of these, 11 (19%) died of TRM, seven after the first and four after the second AT. The reasons for TRM were sepsis (n ¼ 6), multiorgan failure (n ¼ 2), adult Bone Marrow Transplantation


826

Recovery from dialysis-dependent renal failure (Figure 1) Of 54 evaluable patients (excluding five early deaths due to TRM in the first month post transplant), 13 (24%) became dialysis independent at a median of 4 months (range, 1–16 months) after AT, eight after the first and five after the second AT. The median creatinine clearance off dialysis was 32 ml/min (range, 17–55 ml/min), compared with 5 ml/ min (range, 0–19 ml/min) of the patients (n ¼ 41) who continued to require dialysis (Wilcoxon, P ¼ 0.0001). Factors affecting the recovery of renal function included time on dialysis (Table 3): 12 of 36 (33%) with p6 months of dialysis prior to the first AT became dialysis independent, compared with one of 18 patients (6%) who had been on dialysis 46 months (P ¼ 0.03); attainment of CR or nCR: 12 of 31 (39%) in CR or nCR after AT vs one of 22 patient (5%) attaining less than a CR or nCR (P ¼ 0.04). Of 28 patients with creatinine clearance at AT p10 ml/min, three (11%) became dialysis independent compared with 10 of 26 (38%) with 410 ml/min of creatinine clearance (P ¼ 0.04). None of the other factors were associated with recovery, including the isotype of myeloma (kappa light chain 29% vs lambda light chain 10% vs other myeloma 26%, P ¼ 0.78) or dose of melphalan (140 mg/m2–24% vs 200 mg/m2–24%, P ¼ 0.57). Of the 28 patients whose renal biopsy was available, five of 10 with light-chain deposition disease and six of 15 with cast nephropathy became dialysis independent after AT (P ¼ 0.93). One patient with amyloid kidney died of myocardial infarct and two with both light- and heavychain deposition disease did not improve in renal function.

Discussion The current study shows that high-dose melphalan supported by autotransplant can reverse dialysis-dependent renal failure due to myeloma, and long-term survival in these patients can be achieved, not too dissimilar from that obtained in patients with normal renal function. Although full recovery of renal function was not seen, sufficient Bone Marrow Transplantation

Cumulative Incidence of Off-dialysis Recovery

0.8 0.6 0.4 0.2 0 0

1

2 3 4 5 Years post transplant

6

7

0

1


6

7

0

1


6

7

0

1


6

7

b 1 Cumulative Incidence of Off-dialysis Recovery

The median survival of all 59 patients from AT was 41 months and a 5-year overall survival rate was 36%. The projected 7-year survival is 24% (CI: 13–35%). The median survival of patients in CR þ nCR (n ¼ 31) was not reached (5-year survival 58%), compared with 13 months in patients (n ¼ 21) in less than CR þ nCR (2-year survival 39%; 5-year survival 10%) (P ¼ 0.002). The median EFS was 22 months for all patients, with a 5-year EFS rate of 24%. A total of 22 patients are alive and event free from 6 to 76 months. Patients achieving CR or nCR post-AT had a median EFS of 54 months (5-year survival 41%), compared with 11 months for those attaining onCR (2-year EFS 20%; projected 5-year survival 7%) (P ¼ 0.0009).

1

0.8 0.6 0.4 0.2 0

c 1 Cumulative Incidence of Off-dialysis Recovery

Overall and event-free survival after AT

a

0.8 0.6 0.4 0.2 0

d 1 Cumulative Incidence of Off-dialysis Recovery

respiratory distress syndrome (n ¼ 1), veno-occlusive disorder of liver (n ¼ 1) and myocardial infarct (n ¼ 1).

0.8 0.6 0.4 0.2 0

Figure 1 Cumulative incidence of off-dialysis recovery: (a) whole group; (b) prior dialysis duration o6 months (solid line) vs X6 months (dashed line); (c) CR þ nCR (solid) vs onCR (dashed) post transplant; (d) creatinine clearance 410 ml/min (solid) vs p10 ml/min (dashed).


827 Table 3

Analyses of factors associated with off-dialysis recovery Incidence of recovery (%)

Isotype: KLC vs LLC vs heavy CL Pre-transplant response: XPR vs oPR Dialysis duration: p6 months vs 46 months CrCl at transplant: 410 vs p10 (ml/min) Melphalan dose (mg/m2): 140 vs 200 Transplant response: CR+nCR vs pPR

29 vs 10 vs 26 33 vs 23 33 vs 6 38 vs 11 24 vs 24 39 vs 5

recovery without the need for dialysis occurred more frequently when AT was applied early in the disease course, and after attaining a CR or nCR post transplant. The toxicity rate and TRM were clearly higher than in patients with normal renal function, and precluded the routine application of AT in this population outside a specialized center. In myeloma patients with renal failure treated conservatively including standard chemotherapy, renal function has been shown to improve more frequently when associated with hypercalcemia, mild to moderate renal insufficiency or relatively low Bence–Jones proteinuria.2,17 Survival of patients recovered from renal insufficiency is better than patients not recovered (31 vs 25 months), but still shorter than those without renal failure (43 months).2 In a study of 94 patients treated with standard chemotherapy, 26% showed improvement in renal function with a median survival of 28 months, compared with 4 months in patients with nonreversible renal failure.17 Similar observations were made in a study of 775 patients, with a median survival of only 3.5 months in patients who required dialysis, suggesting that dialysis-dependent renal failure is a poor prognostic factor for survival.2 The risk of TRM or death due to progressive disease in these patients was the highest during the first few months of initiation of standard chemotherapy, but the response rate and median survival of those who survived the first 2 months in one study were 40% and 2 years, respectively, suggesting that dialysis dependency per se should not be taken as a reason not to provide appropriate care for these patients.18 We have previously shown that patients with moderate to severe (serum creatinine 42 mg/dl) renal failure could achieve a response rate and survival after high-dose melphalan combined with autologous stem cell transplantation almost equivalent to those with newly diagnosed myeloma.8,19 In our study on the high-dose melphalan (100 mg/m2) pharmacokinetics in patients with renal failure, the median half-life (t1/2), area under the concentration curve and clearance of melphalan were 1.1 h, 5.5 mg h/land 27.5 l/h, respectively, in patients with a creatinine clearance of o40 ml/min compared to 1.9, 7.9 and 23.6 for the others, indicating no adverse effect of renal insufficiency.19 Further clinical study has revealed that the quality of peripheral blood stem cell collections and post transplant engraftment were not jeopardized by renal failure, with a median of 10 106 CD34 þ cells/kg and a median time to absolute neutrophil count 40.5 109/l of 11 days.8 In our previous study, CR was achieved in 21 patients (26%) after the first

Univariate values b

P-values

0.18 0.25 1.57 1.38 0.33 1.46

0.78 0.62 0.03 0.04 0.57 0.04

autotransplant and 31 patients (38%) after the second autotransplant, with a 48% probability of EFS and 55% of OS at 3 years, respectively. The present study shows that the quality of life after autotransplant could also be significantly improved in these patients, with 24% of the patients requiring no further dialysis. In particular, those who had achieved a CR or nCR after AT showed a 5-year EFS and OS of 41 and 58%, respectively, with one-third recovering from dialysis dependency. Although the results may be partly related to the short duration (p6 months) of dialysis prior to AT in 2/3 of the patients (n ¼ 37) and inclusion of a relatively large number of newly diagnosed patients (n ¼ 23), the current study demonstrates a favorable outcome of AT following high-dose melphalan even in patients with compromised renal function. The median time on dialysis in our patients was 4 months. Recovery after a short duration of dialysis prior to AT and attainment of CR or nCR post transplant suggests that reversibility of the damage may be related to abrupt termination of ongoing kidney damage in patients with relatively new onset, but reversible complete renal failure. In the current study, we have seen myeloma patients with cast nephropathy (n ¼ 6) and light-chain deposition disease (n ¼ 5) become dialysis independent, although renal biopsies were available in only 28 of 59 patients. Further investigation is needed to determine the role of a kidney biopsy with respect to the type and extent of renal damage, in order to predict more accurately as to which patient group is most likely to reverse dialysis dependency after high-dose melphalan and autotransplant. The present study suggests that myeloma patients with renal failure be evaluated promptly at diagnosis for possible high-dose melphalan followed by AT unless significant comorbid conditions exist. Dialysis dependency should not be viewed as a contraindication for transplant. Successful implementation of AT requires well-coordinated supportive care preferably in the setting of a tertiary care center.

References 1 Knudsen LM, Hippe E, Hjorth M et al. Renal function in newly diagnosed multiple myeloma – a demographic study of 1353 patients. The Nordic Myeloma Study Group. Eur J Haematol 1994; 53: 207–212. 2 Knudsen LM, Hjorth M, Hippe E. Renal failure in multiple myeloma: reversibility and impact on the prognosis. Nordic Myeloma Study Group. Eur J Haematol 2000; 65: 175–181. Bone Marrow Transplantation


828 3 Pichette V, Querin S, Desmeules M et al. Renal function recovery in end-stage renal disease. Am J Kidney Dis 1993; 22: 398–402. 4 Rota S, Mougenot B, Baudouin B et al. Multiple myeloma and severe renal failure: a clinicopathologic study of outcome and prognosis in 34 patients. Medicine 1987; 66: 126–137. 5 Komatsuda A, Wakui H, Ohtani H et al. Disappearance of nodular mesangial lesions in a patient with light chain nephropathy after long-term chemotherapy. Am J Kidney Dis 2000; 35: E9. 6 Pozzi C, Fogazzi GB, Banfi G et al. Renal disease and patient survival in light chain deposition disease. Clin Nephrol 1995; 43: 281–287. 7 Comenzo RL, Vosburgh E, Falk RH et al. Dose-intensive melphalan with blood stem-cell support for the treatment of AL (amyloid light-chain) amyloidosis: survival and responses in 25 patients. Blood 1998; 91: 3662–3670. 8 Badros A, Barlogie B, Siegel E et al. Results of autologous stem cell transplant in multiple myeloma patients with renal failure. Br J Haematol 2001; 114: 822–829. 9 National Cancer Institute: Cancer Therapy Evaluation Program: Common Toxicity Criteria, version 2, Bethesda, MD, National Cancer Institute, April 30, 1999. 10 Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18: 695–706. 11 Hintze JL. Number Cruncher Statistical System 2001. NCSS Kaysville, UT.

Bone Marrow Transplantation

12 Kaplan EL, Meier P. Nonparametric estimation from incomplete observation. Journal of the American Statistical Association 1958; 53: 457–481. 13 Collett D. Modeling Survival Data in Medical Research. Chapman & Hall: London, 1994. 14 Lee CK, Barlogie B, Zangari ME et al. D.T.PACE, an effective, novel combination chemotherapy with thalidomide for previously treated patients with myeloma. J Clin Oncol 2003; 21: 2732–2739. 15 Fassas AB, Spencer T, Desikan R et al. Cytotoxic chemotherapy following tandem autotransplants in multiple myeloma patients. Br J Haematol 2002; 119: 164–168. 16 Barlogie B, McCoy J, Shaughnessy J et al. Total therapy II (TT II) for newly diagnosed multiple myeloma (MM): preliminary data on feasibility and efficacy in the first 231 enrolled patients; Comparison with Predecessor Trial Total Therapy I (TT I) (N ¼ 231). Blood 2001; 98: 683 (abstract 2857). 17 Blade J, Fernandez-Llama P, Bosch F et al. Renal failure in multiple myeloma: presenting features and predictors of outcome in 94 patients from a single institution. Arch Intern Med 1998; 158: 1889–1893. 18 Torra R, Blade J, Cases A et al. Patients with multiple myeloma requiring long-term dialysis: presenting features, response to therapy, and outcome in a series of 20 cases. Br J Haematol 1995; 91: 854–859. 19 Tricot G, Alberts DS, Johnson C et al. Safety of autotransplants with high-dose melphalan in renal failure: a pharmacokinetic and toxicity study. Clin Cancer Res 1996; 2: 947–952.