Treatment of 100 patients with primary amyloidosis: A randomized trial of melphalan, prednisone, and...

Treatment of 100 Patients With Primary Amyloidosis: A Randomized Trial of Melphalan, Prednisone, and Colchicine Versus Colchicine Only Martha Skinner, MD, Jennifer J. Anderson, PhD, Robert Simms, MD, Rodney Falk, MD, Ming Wang, MS, Caryn A. Libbey, MD, Lee Anna Jones, MD, Alan S. Cohen, MD, Boston, Massachusetts

FURWSE: A clinical trial designed to test whether treatment with melphalan, prednisone, and colchicine (MPC) is superior to colchicine (C) alone was performed in patients with primary amyloidosis (AL), a nonmalignant plasma cell dyscrasia.

PATIENTS AND METIWS: Patients were randomized to MPC or C with stratification according to sex, time from diagnosis to study entry (ie, less than 3 months or 3 to 12 months), and dominant organ system involvement (ie, cardiac, renal, neurologic, or other). Data were gathered monthly from patients, quarterly from physicians, and annually in the Clinical Research Center. One hundred consecutive patients with AL amyloidosis admitted between 1987 and 1992 who met eligibility requirements were treated and followed for a minimum of 18 months. Fii patients (group A) received daily oral colchicine and 50 patients (group B) received cycles of oral melphalan and prednisone every 6 weeks for 1 year as well as colchicine.

RESULTS: The principal outcome measure was median survival, which was compared in the two treatment groups and in the subgroups. The overall survival of all patients from study entry was 8.4 months. Comparing group A (C) to group B (MPC), the survival was 6.7 months versus 12.2 months (P = 0.087). Both treatment groups had poor survival for patients in the cardiac sub- group, longest survival in the renal group, and significant differences favoring MPC treatment only in patients whose major system manifestations were neurologic (P = 0.037) or other (P = 0.007). Multivariate analysis showed a strongly significant treatment effect (P = 0.003) and improved survival associated with not having cardiac or gastrointestinal involvement.

From the Arthritis Center (MS, JJA, RS, MW, CAL, LAJ, ASC) and Cardiology Section (RF), Thorndike Memorial Laboratories, Boston City Hospital%d Department of Medicine, Boston University School of Medicine, Boston, Massachusetts.

Supported by grants from the National Institutes of Health (AR20613, AR40414, RR5331 and the Arthritis Foundation.

Requests for reprints should be addressed to Martha Skinner, MD, Arthritis Center, Boston University School of Medicine, 80 East Concord Street, Boston, Massachusetts 02118.

Manuscript submitted August 26, 1994 and accepted in revised form August 29, 1995.

CONCLUSKINS: MPC was advantageous for patients whose major manifestations of amyloid disease were other than cardiac or renal. Better survival regardless of treatment was noted in patients for whom a satisfactory supportive treatment such as transplant or dialysis exists for their organ failure.

rimary amyloidosis (AL) is a plasma cell dyscra- P sia characterized by an autonomous Iproliferation of plasma cells with an overproduction of a monoclonal immunoglobulin protein.‘12 The imununoglobu- lin light chains or fragments thereof deposit throughout the body as amyloid fibrils. Although AL amyloidosis is similar in pathogenesis to multiple myeloma, it lacks the “malignant” sheets of immature plasma cells in the bone marrow and usually does not have the clinical findings of anemia and osteolytic bone le- sions. Nevertheless, the survival of AL amyloidosis may be worse than that of uncomplicated multiple myeloma due to the deposition of amyloid fibrils in major organ systems.3,4

There is no specific therapy for AL amyloidosis. In 1972, colchicine was found to inhibit tlhe cyclic inflammatory illness of familial Meditenanean fever and thus prevent secondary (AA) amyloiidosis in this disease.5 Since then, it has been used to treat AA amyloidosis due to other causes as well as AL amyloidosis.W Our recent experience treating AL amyloidosis with colchicine suggested a modest increase in SW vival when compared to retrospective controls.’ Other therapies that have been tried include: vitamin E,g dimethyl sulfoxide,1° and interferorccx,ll but results have been disappointing.

Case reports of a few patients treated with immunosuppressive drugs, most commonly melphalan in combination with prednisone, suggest at marked improvement in measurable parameters such as decrease in urinary protein, decrease in hepatomegaly, decrease in monoclonal protein, and in one case, re- versal of an amyloid positive organ bialpsy to nega- tive. 12-18 None of these reports note whether other patients were treated but did not have measurable improvement. However, two large clinical trials of 55 and 101 patients reported by Kyle et a17Jg compare the surviva3 of AL amyloidosis after treatment with melphalan and prednisone versus placebo and versus colchicine. In the first study, treatment benefit

290 March 1996 The American Journal of Medicine* Volume 100

could not be determined, as more than half of those given placebo were treated with melphalan and prednisone due to disease progression. Similarly, in the second study, a crossover treatment design was used if disease progression occurred, which it did in nearly half of the patients. Nevertheless, survival was slightly longer in the melphalan and prednisone treated patients, although not significantly so, and limited measurable improvements occurred only in this group.

For these reasons, we chose to examine whether or not melphalan and prednisone added to colchicine had any benefit over colchicine alone. The continu- ing availability of a large number of referred patients with AL amyloidosis allowed us to conduct a randomized clinical trial.

METHODS Patients

One hundred patients with AL amyloidosis seen between August 1987 and June 1992 were evaluated in the Boston City Hospital Clinical Research Center at Boston University School of Medicine, randomized to a treatment regimen, and followed for a minimum of 18 months. All patients with AL amyloidosis referred to the Clinical Research Center were included in the trial if they met the eligibility requirements of amyloid diagnosis by a Congo red stained biopsy within 1 year of referral, exclusion of multiple myeloma, and no prior treatment with melphakm and prednisone. The date of the positive biopsy was noted as the date of diagnosis. The AL type of amyloidosis was con- firmed by finding a monoclonal immunoglobulin protein in serum or urine, positive immunohistochemi- cal staining of bone mar&w demonstrating clonal plasma cells or tissue amyloid deposits with anti- kappa or anti-lambda antiserum, or by overwhelm- ingly convincing clinical features (eg, macroglossia associated with other systemic manifestations).

Protocol Design All patients were informed of the trial, and consent

was obtained for the evaluation and for treatment with colchicine. Patients subsequently randomized to melphalan and prednisone were asked to sign a second consent form. This treatment assignment method is termed randomized consent.20 The scheme was chosen to minimize asking some patients to give consent for a treatment (ie, melphalan and prednisone) they might not receive.

Each patient was placed in 1 of 16 strata defined according to sex, time from diagnosis to study entry (ie, c3 months or 3 months to 1 year), and dominant organ system involvement (ie, cardiac, renal, neurologic, or other). Randomization was carried out within each stratum by an investigator who was not

involved with patient care, assigning 50% of the patients to group A and 50% to group B. Patients randomized to group A received colchicine 0.6 mg t.wice daily. Patients randomized to group B received melphalan 0.15 to 0.25 mg/kg/d and prednisone 1.5 mg/kg/d for 4 days every 6 weeks for a total of 600 mg of melphalan, as well as colchicine 0.6 mg twice daily except on the days melphalan was I-aken. Pa- tients were started on the lower melphalan dose and increased until leukopenia or thrombocytopenia occurred at mid-cycle. The total dose of melp:halan was limited to 600 mg in an attempt to avoid the leuke- mogenic effect that is a greater risk with higher doses.“1,22 Colchicine was continued indefinitely in the group B patients.

The Clinical Research Center evaluation al; study entry examined the extent of organ system involvement in each patient. Subsequently, data were gathered by monthly medication records from patients, quarterly reports from primary physicians, and annual evalua- tions in the Clinical Research Center. Parameters of quantitative measurement were chosen to assess cardiac, renal, neurologic, and gastrointestinal function. Tests for cardiac function included: electrocardiogram, echocardiogram, and Holter monitor; tests for renal function included: serum creatinine, creatinine clearance, 24-hour urine protein, and serum albumin; tests for neuropathy included: blood pressure and pulse, both lying and standing, and sensory and motor neurologic exam; tests for gastrointestinal function included: weight, liver size by physical exam, stool fat, D-xylose, and gastric emptying scan; tests for overall amyloid burden assessment included: bone marrow plasma cells, serum M component, amyloid in abdominal fat, and number of organ systems involved.

In addition, during the course of treatment, subjec- tive assessments of patient improvement or worsening in each organ system were made by the primary physician and the Clinical Research Center evalua- tions. Patient worsening was also document.ed objec- tively by new events that indicated disease progression, such as, syncopal episodes, stroke, congestive heart failure, cardiac arrhythmia, or need for dialysis or other supportive therapies. Patient improvement was documented by decrease in organomegaly, improvement in renal or liver function, and disappear- ance of serum M component. New involvement was noted when an organ system not previously .known to be involved developed evidence of amyloid disease by above assessments or a new positive biopsy.

Statistical Analysis A sample size of 100 patients, each with at least 1

year of follow-up, was planned in designing the trial in order to have 80% power for the detection of a clin- ically important 25% increase in 1 year survival in the

TREATMENT FOR PRIMARY AMYLOIDOSIS/SKINNER ET AL

March 1996 The American Journal of Medicine@ Volume 100 291

TABLE I System involvement at Study Entry

!J:;;/ (n!El No. (%) Total

Hematologic 46 (92%) 44 (88%) 90 M component in serum or urine 34 39 Ig with lambda light chain 29 27 Bone marrow plasma cells 25% 30 34

Renal 41 (82%) 38 (76%) 79 Creatinine 2130 umol/L 9 11 Proteinuria >0.5 g/d 37 30 Creatinine clearance cl.00 mL/s 20 15 Serum albumin ~30.0 g/L

Gastrointestinal 37;;4%) 42;:4%) 79 Weight loss reported 24 26 Hepatomegaly 13 13 Alkaline phosphatase >200 U/L 6

Neurologic 36 (72%) 40 :800%) 76 Postural hypotension 13 15 Sensory neuropathy 17 18 Impotence (males) 10 10 Motor neuropathy 4 4

Cardiac 36 (72%) 38 (76%) 74 Low voltage on ECG 21 27 Cardiomyopathy on ECHO 28 30 Congestive heart failure 23 27 Syncope’ 13 3

Musculoskeletal 13 (26%) 13 (26%) 26 Arthropathy 1 2 Macroglossia 12 13

Pulmonary 10 (20%) 8 (16%) 18 Dermatologic 8 (16%) 5 (10%) 13 No. systems involved/person (mean f SD) 4.5 + 1.3 4.6 * 1.1

‘P = 0.012. C = colchicine; MPC = melphalan, prednisone, colchicine; Ig = immune globulin; ECG = electrocardiogram; ECHO = echocardiogram; SD = stan dard deviation.

chemotherapeutically treated group (one-sided test at the 0.05 level). All patients were analyzed with their assignment group, using the intent-to-treat principle.

The baseline status of the treatment groups was compared using &i-square tests for categorical variables and Student’s t-tests and Wilcoxon 2-sample tests for continuously measured variables. Kaplan- Meier survival curves, which take into account the censoring of observations because some patients are still living, were constructed in order to compare the survival distributions of the patients in the two treatment groups and also of patients with and without certain clinical features. The median was used to typ- ify the survival, and the shapes of the survival curves were compared statistically using the exponential likelihood test. This type of comparison was also performed within strata defined by time from diagnosis to referral and, separately, by major clinical manifestation. Survival was measured from the time of entry into this study. Survival until the earlier of death or organ death (ie, the point in time at which life-saving

interventions were required) was also examined in the same way, with only those living patients who had not required life-saving interventions providing censored data In multivariate analysis, survival in the two treatment groups was compared, adjusting simul- taneously for various types of system involvement, using a Weibull or accelerated failure time mode1.23

RESULTS Forty-one of the patients were female and 59 were

male. Two-thirds of the patients were referred within 3 months of diagnosis and the rest were seen 3 to 12 months after diagnosis. The major organ system involved was cardiac in 33 patients, renal in 29, neurologic in 11, and other in 27. The “other” category included 10 patients in whom the major involvement was hepatomegaly, 6 with macroglossia, 3 with pulmonary infiltrates, 3 with weight loss, and 1 each with splenomegaly, arthropathy, adenopathy, muscle thickening, and purpura Although the major prob- lems at study entry were varied in these :27 patients, many also had usual system involvement; 20 patients had cardiac, 21 patients had renal, 19 patients had neurologic, and 25 patients had gastrointestinal involvement with amyloidosis. The total number of systems involved with amyloid in patients in the “other” stratum was comparable to that of patients in the cardiac, renal, and neuropathy strata.

Within these strata of male, female, early referral, late referral, and four categories of major involvement, 50 patients were randomized to group A, treatment with colchicine (C); and 50 patients were randomized to group B, treatment with melphakm, prednisone, and colchicine (MPC). In group A (C), the median age was 62.7 years, and 20 of the 50 patients were female. In group B (MPC), the median age was 60.6 years, and 21 of the 50 patients were female.

System involvement and laboratory data at baseline are compared in Table I. A particular system was defined as involved if a patient had any one of the fol- lowing: for cardiac involvement, restrictive cardiomyopathy on echocardiogram or low voltage on electrocardiogram; for renal involvement, proteinuria 20.5 g/d or creatinine 2130 pmol/L, for neurologic involvement, peripheral sensory or motor loss, carpal tunnel syndrome, or orthostatic hypotension; for gastrointestinal involvement, enlarged liver, abnormal gastric emptying scan, diarrhea, or weight loss; for musculoskeletal involvement, amyloid arthropathy, macroglossia, or amyloid deposits in muscles; for hematologic involvement, bone marrow plasmacytosis 25Oh, serum monoclonal protein, or clotting factor deficiency; for dermatologic involvement, ecchy- moses; and for pulmonary involvement, abnormal pulmonary function tests. An amyloid positive biopsy also defined involvement within any syste:m. Although

292 March 1996 The American Journal of Medicine@ Volume 100


TABLE II Biopsy Results in 100 Patients With AL Amyloidosis

No. No. Biopsy Site Tested Positive Percent

Abdominal fat 97 86 88.7 Bone marrow 44 23 52.3 Kidney 21 21 100 Rectal 21 18 85.7 Heart 14 14 100 Liver 11 11 100 Small bowel 8 7 87.5 Gingiva 6 5 83.3 Skin 3 3 100 Nerve 2 2 100 All other 22 22 100 Total 249 212 85.1 AL = primary (light chainderived).

each patient had one identified major system involved for purposes of stratification, all patients were found to have at least two and some as many as eight (av- erage 4.4) systems involved at study entry.

The two groups were alike with respect to system involvement at baseline (Table I). Bone marrow plasmacytosis was present in 30 patients in the C group, with values for the whole group ranging from ~3% to 25% (median 5%), and in 34 patients in the MPC group, with values for the whole group ranging from ~3% to 20% (median 7%). The various abnormal parameters within each system were evenly distributed in each group, except for syncope, which was present in more patients in the C treatment group (13 versus 3, P = 0.012).

Each patient had at least one positive biopsy for amyloid deposits, and a review of 249 biopsies from the 100 patients found 212 (85.1%) were positive (Table II). Ninety patients had evidence of a monoclonal protein in serum or urine by protein elec- trophoresis or immunoelectrophoresis, or had mildly increased plasma cells in the bone marrow with a monoclonal light chain staining pattern. Of the 10 remaining patients, the diagnosis of AL amyloidosis was based on overwhelming clinical evidence in- cluding multisystem disease associated with macroglossia in 2, hypogammaglobulinemia in 7, and/or light chain staining of amyloid in a tissue biopsy in 2. None of these 10 patients had a history to suggest familial amyloidosis or an underlying inflammatory disease, such as, rheumatoid arthritis, to suggest re- active (AA) amyloidosis.

The overall survival of all patients from study entry was 8.4 months. Comparing group A (C) to group B (MPC), the survival from study entry was 6.7 versus 12.2 months (P = 0.087) (Figure 1). Survival comparison from diagnosis was 10.5 versus 14.8 months (P = 0.132), and from first symptom of disease was 29.7 versus 36.7 months (P = 0.006).

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0 0 12 24 36 48 60 72

Figure 1. Proportion of patients surviving by months from study entry are shown for 50 patients treated with colchicine (closed circles) and 50 patients treated with melphalan, prednisone, and colchicine (open circles). The curves are parallel for the first 5 months, after which the melphalan, prednisone, and colchicine group has a better survival.

According to the method of analysis used in this study, survival data were based on treatment assign- ments. Although every effort was made to encourage patients to take the medicine to which they had been assigned, for a variety of reasons some patients did not. There were 6 patients of the 50 assigned C alone (group A) who subsequently received MPC, and of these, 5 patients received 3 or more cycles. Of the 50 patients assigned MPC (group B), 3 refused the MP part of the treatment (received only C) and ;’ died before the MP portion of the treatment (undertaken by the primary physician when the patient returned home) could be started, leaving 40 patients of whom 29 lived long enough to receive 3 or more cycles of MP. The 6 patients in group A who received MPC and the 3 in group B who refused treatment were too few in number for statistical comparisons with their whole group, but they did not appear to fare better or worse than others within the same strata (Table III). The 6 patients assigned to C that recefved MPC survived 3, 4, 7, 12, 16, and 44 months from diagnosis. The 3 assigned to MPC but who only received C survived 12, 18, and 26 months from diagnosis.

There was a marked difference in survival between the two strata of time from diagnosis to referral. The overall survival from study entry of 67 paltients referred within 3 months was 7.0 months versus 16.0 months for the 33 patients referred between 3 and 12 months from diagnosis (Table IV). The tendency for longer survival in the patients referred later was present in both treatment groups. Survival from study entry of patients within each major organ system involvement stratum was 4.4 months for cardiac, 18.7 months for renal, 5.9 months for neurologic, and



TABLE Ill Treatment Versus Assienment

C MPC Number assigned 50 50

Refused 3 Died before MPC treatment 7

Number receiving any MPC 6 40 Number receiving 3 or 5 29 more cycles of MPC

Survival from entry (months) of 6 patients assigned to C, but received MPC

Renal (2) 44, 16 Neurologic (1) 4 Other (3) 12, 7, 3

Survival from entry (months) of 3 patients assigned to MPC, but refused and received only C

Renal (21 26, 18 Other (1) 12

C = colchicine; MPC = melphalan, prednisone, colchicine.

TABLE IV Median Survival of Patients From Study Entry

All Patients C MPC P Value’

All patients 8.4 6.7 12.2 0.087 Time from diagnosis to referral

c3 months (n = 67) 7.0 6.4 8.9 0.556 3-12 months (n = 33) 16.0 11.1 >16 0.053

Sex Male In = 59) 6.9 6.3 16.7 0.400 Female (n = 41) 13.5 13.5 12.2 0.095

Major clinical manifestation Cardiac (n = 33) 4.4 5.3 3.6 0.588 Renal (n = 29) 18.7 21.5 17.3 0.598 Neurologic (n = 11) 5.9 4.5 >6.1 0.037 Other (n = 27) 11.8 6.6 19.9 0.007

‘For C versus MPC. C = colchrclne; MPC = melphalan, prednisone, colchicine.

11.8 months for other. Both treatment groups had poor survival in the cardiac group, longest survival in the renal group, and significant differences favoring MPC treatment only in patients with major neurologic or “other” system manifestations. The patient’s sex did not influence survival overall or within the treatment groups.

Using a multivariate survival analysis, treatment with MPC versus C was compared adjusting for any involvement of cardiac, renal, neurologic, and gastrointestinal systems regardless of initial stratification. There was a strongly significant treatment effect (P = 0.003) and improved survival from study entry associated with not having cardiac (P lO.001) or not having gastrointestinal (P = 0.002) involvement, but no change in survival associated with having renal (P = 0.543) or neurologic (P = 0.895) involvement. In the C group, 6 patients had neither cardiac nor gastrointestinal involvement, and their median survival

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0 -1~-I 0 12 24 36 48 60 72

Figure 2. Proportion of patients with cardiac and gastrointestinal (GI) involvement surviving by months from study entry are shown for 29 patients treated with colchicine (closed circles) and for 33 patients treated with melphalan, prednisone, and colchicine (open circles). After 6 months of treatment, the melphalan, prednisone, and colchicine group has a better survival.

was 44 months. In the MPC group, 3 patients had neither cardiac nor gastrointestinal involvement, and their median survival was 33 months (2 patients still alive).

When cardiac, but not gastrointestinal, involvement was present, the median survival was 15 months in the C group (n = 7) and 35 months in the MPC group (n = 5, with 2 still alive). This tendency towards longer survival in the MPC group is not significant with so few patients (P = 0.157). When gastrointestinal, but not cardiac, involvement was present, the median survival was 17 months in the C group (n = 8) and 27 months in the MPG group (n = 9). Likewise, this tendency towards longer survival in the MPC group was not significant with so few patients (P = 0.426).

When both cardiac and gastrointestinal involvement was present, survival was 4.4 months in the C group (n = 29) and 5.4 months in the MPC gr0u.p (n = 33). A significant difference is noted here (P = 0.006) even though survival is poor with either treatment. The rea- son for the significant difference is clear when the survival curves are examined (Figure 2). The MPC group starts to do better after 6 months, but by then more than half of each group have already died.

During the study period, 23 patients had one or more other major treatment intervention. Two patients had heart transplants, 8 received :pacemakem, 2 had kidney transplants, 18 were placed on dialysis, 7 were given jejunostomy tubes, and 8 were placed on total parenteral nutrition (Table v). The patients receiving these interventions were nearly equally di- vided between the treatment groups. The differential impact of the interventions could be realized when the survival until the earlier of death or organ death (ie, the time the intervention became necessary) was

294 March 1996 The American Journal of MedicineQ Volume 100


TABLE V Other Major Treatment Interventions

C MPC Total Heart transplant 1 1 2 Pacemaker 6 2 8 Kidney transplant 2 0 2 Dialysis 9 9 18 Jejunostomy tube 4 3 7 Total parenteral nutrition 3 5 8 Total 25 20 45 C = colchicine; MPC = melphalan, prednisone, colchicine.

examined (Table VI). The benefit in the C treated patients was minimal, as survival from study entry increased from 6.6 to only 6.7 months, but it increased from 10.6 to 12.2 months in the MPC treated patients. The overall difference between the C and MPG treatment in survival until organ death is significant (P = 0.017), suggesting a delaying effect of MPC with respect to organ failure.

The data gathered from patients’ subsequent visits to the Clinical Research Center and their physicians’ quarterly reports were examined for measurable parameters reflecting worsening, improvement, or new organ system involvement (Table VII). These parameters were tabulated for each system of each person in each of the treatment groups and defined by a notation of person/system. The worsenings far out- numbered the improvements in both groups, but a few more worsening parameters were noted for those treated with C (71 versus 63) and more improvements were noted for those treated with MPC (16 versus 4).

Cardiac improvement was noted by a decrease in symptoms of congestive heart failure in 6 patients. In 5 of these, 2 of 4 cardiac worsening events also occurred during the study. Four of the 6 patients were treated with melphalan, and improvement occurred after 320, 512, and 600 mg in 3 patients; 1 patient improved after a heart transplant. No patient had a decrease in cardiac wall thickness on echocardiogram Renal improvement was noted in 4 patients, all of whom were in the MPC group. A decrease in proteinuria occurred in 3 after 480,512, and 536 mg melphalan and after a kidney transplant in 1 patient. Gastrointestinal improvement was noted by a decrease in diarrhea or increase in weight in 4 patients (3 on melphalan after 450,556, and 588 mg), and in 1 patient, a decrease in liver size was noted after 512 mg melphalan. Four of the 5 patients also had gastrointestinal worsening events.

No patient noted an adverse effect of the colchicine treatment, although some patients decreased the dosage from 2 to 1 (0.6 mg) tablet daily because of diarrhea. Two patients had complica- tions of melphalan treatment with lowered blood

1 TABLEVI Median Survival Until Organ Failure or Death

All Patients C’ MPCt P Value* All patients 7.9 6.6 10.6 0.017 Time from diagnosis to referral

<3 months (n = 67) 6.4 3-12 months (n = 33) 15.4

Major clinical manifestations Cardiac (n = 33) 3.9 Renal (n = 29) 15.4 Neurologic (n = 11) 4.6 Other (n = 271 11.8

‘Four censored. IThirteen censored. ‘For C versus MPC.

6.9 5.9 0.242 6.5 18.7 0.024

5.3 3.0 0.533 15.4 16.7 0.804 4.0 >12.0 0.033 6.8 18.0 0.010 --

Worsening Improvement New Involvement c MPC c MPC CMPC

Cardiac 17 9 2 4 4 2 Renal 10 12 0 4 1 3 Neurologic 6 10 1 0 2 1 Musculoskeletal 2 4 0 0 0 3 Gastrointestinal 27 22 1 4 2 0 Hematologic 7 3 0 3 0 0 Pulmonary 2 2 0 1 0 0 Dermatologic 0 1 0 0 1 1 Person/System 71 63 4 16 10 10 Renal labs 24 25 3 2 Gl/weight loss 18 14 6 9 C = colchicine; MPC = melphalan, prednisone, colchicine; GI = gastrointestinal.

counts. In one, thrombocytopenia with. platelet counts as low as 15 X log/L occurred suddenly after 1 year of melphalan treatment totaling 454 mg. After treatment was stopped, the platelet count. sta- bilized at 135 X log/L with a hematocrit of 33?h and white blood cell count of 3.3 X log/L with a normal differential. In the other patient, a persistent leukopenia with a white blood cell count of 2.!3 X 10Q/L and a normal differential was noted after 1 year of treatment that totaled 576 mg of melphalan. Both patients are alive and maintain full- or part-time employment 40 and 39 months, respectively, after diagnosis. No patient treated with melphalan has developed leukemia during the study period. All patients continue to be followed.

A possibly related adverse effect occurred. in 2 other patients, both of whom developed infections. One, with pulmonary amyloidosis as the major involvement, developed KlebsieUu pneumonia after 3 cycles of melphalan. The other, with soft tissue amyloid in- cluding macroglossia and arthropathy as the major involvement, developed coccidiomycosis after 1 year of melphalan totaling 384 mg. These patients survived 14 and 36 months, respectively, after diagnosis.


TABLE VIII

Primary Cause of Death

C MPC

Heart failure 15 11 Sudden death 11 13 Arrhythmia 2 3 Renal failure 7 2 Hepatic failure 3 4 Pulmonary failure 2 1 Pneumonia 2 0 GI infection 1 1 GI bleeding 1 0 Non-G1 bleeding 0 1 Total* 44 36 ‘P ~0.05 for C versus MPC. C = colchicine; MPC = melphalan, prednlsone, colchicine; GI = gastroin- testlnal.

Eighteen months after the last person was admitted and 6 years after the trial began, 80 of the 100 patients have died (Table VIII). More deaths have occurred in the C treatment group (44 versus 36, P 50.05). Fifty- five of the deaths were due to a cardiac event (26 heart failure, 24 sudden death, 5 arrhythmia), 9 were from renal failure, 7 from hepatic failure, 3 from pulmonary failure, 4 from infection (2 pneumonia, and 2 GI infections), and 2 from bleeding.

DISCUSSION AL amyloidosis is known to be a fatal disease with

a short life expectancy. To date no cure is known. The disease process itself is a mild plasmacytosis with a monoclonal protein that is frequently difficult to de- tect because it is so small. However, this monoclonal protein forms amyloid fibrils within tissues that in- terfere with function and lead to death. Treatment for the disease targets the plasma cells and thus the pro- duction of monoclonal protein. There is no treatment to remove the deposited amyloid fibrils, and it is not known whether they can actually decrease in quan- tity within tissues.

There is difficulty in measuring therapeutic benefit when reversals in the disease are not present. In AL amyloidosis, it may be necessary to measure improvement by observing a slowing of disease progression. We have conducted a large clinical trial in patients with AL amyloidosis and suggested, by mea- sures of overall survival, by greater time until organ death, and by fewer worsening events, an advantage to treatment with MPG over C alone. The advantage was for patients presenting with neuropathy or “other” involvement, but not for those with cardiac or renal disease as a major feature. The survival advantage persists in the full set of patients after ad- justment for the presence of cardiac or gastrointestinal involvement, both of which were each strongly significant on multivariate analysis. No differences in

survival associated with renal or neurologic involvement were seen on multivariate analysis. Thus, patients with advantageous clinical features appeared to survive longer, and in general, there was benefit from the melphalan therapy.

Five patients survived 60 months or more since diagnosis, and 2 others survived 58 and 59 months. Although all patients have been followed for at least 18 months, only the first 43 entered have been followed for at least 58 months. The 7 patients who survived at least 58 months represent 16% of the patients in whom the diagnosis was made 58 or more months ago. It is worth noting that 2 of these patients had kidney transplants, 1 had a heart transplant, and the other 4 were assigned to and received MPC.

Two of the study patients received heart transplants, and 1 can be considered a long s-urvivor. She had cardiac involvement from the onset and received a heart transplant 9 months after diagnosis. She had been assigned MPC but did not receive :it initially in anticipation of transplant, then did not receive it because of the other immunosuppressive drugs being given to prevent rejection. She did receive MPC 40 months after transplant when a cardiac biopsy re- vealed amyloid deposits in her transplanted heart. She lived a busy and active life until just before her death 59 months after her initial diagnosis. The second patient who received a heart transplant presented with severe cardiomyopathy and was transplanted 4 months later. She died of pulmonary amyloidosis 14 months after the transplant. She had been assigned to C and received it for the duration of her disease. The appropriateness of heart transplantation has been questioned for patients with AL amyloidosis.“4 Our experience notes the benefit of such treatment, particularly in view of the bleak prognosis for patients presenting with cardiac involvemenLZ5

The remaining 4 long survivors all were assigned to and received full courses of MP totaling 588, 512, 480, and 556 mg of melphalan. Therapy with C alone continued after MP cycles ended. The first patient presented with macroglossia and also had renal and neuropathic involvement. He lived 75 months, and died from congestive heart and multisystem failure.

The second patient presented with hepatomegaly and also had pulmonary and renal involvement. His liver decreased in size and pulmonary function improved after MPC treatment. He sustained a stroke 41 months after diagnosis but recovered nearly com- pletely and remains alive and active after 78 months.

The third patient presented with major cardiac involvement as well as renal disease. She sltopped C after 24 months due to gastrointestinal disturbance. Fifty-seven months after diagnosis, she received a cardiac pacemaker for heart block and a jejunostomy feeding tube for decreased gastric motility. The feed-

296 March 1996 The American Journal of Medicine@ Volume 100

ing tube was removed 1 year later. She is alive 77 months after diagnosis.

The fourth patient presented with major renal involvement and marked orthostasis and diarrhea. After MPC treatment, her renal function had worsened slightly (creatinine 90 to 150 pmol/L, proteinuria 8.7 to 9.4 g/d), but 24 months later had significantly improved (creatinine 120 pmol/L, proteinuria 1.3 g/d). During this time, her orthostasis and diarrhea also dramatically improved, and she was able to return to full-time employment. She remains well 61 months after diagnosis.

The long survivors in this study did not have less severe disease at onset. They had as extensive organ system involvement as other patients in this study and no one factor would have predicted a better survival when they were initially seen. Two of them had measurable improvements, with a decrease in liver size and lessening of proteinuria that may be attributable to MPC treatment. Three of them benefited from organ transplant (2 kidney, 1 heart) and another from major supportive treatment (pacemaker, jejunostomy feeding tube). The percentage of long survivors is comparable to our previous reported series of patients who were treated with colchicine alone. Of those, 18 of 127 (14%) survived 60 months or longer.26

The seriousness of the AL form of amyloidosis is out of proportion to the apparent degree of the underlying problem, that is, a benign plasma cell dyscrasia with a monoclonal gammopathy of mild degree. The inability of the monoclonal protein to undergo complete proteolysis, its lack of stability while normal processing occurs, or its inherent fibrillogenic propensity are possible factors that allow the protein to form amyloid fibrils within tissues throughout the body. This mass of amyloid librils is variable in size; it is sometimes small but located in a critical area such as the conducting system of the heart or renal glomeruli, and at other times, it is large, causing organomegaly of massive proportions, which ulti- mately leads to organ dysfunction or failure. The widespread deposition of amyloid in this disease was seen in this study with the extraordinarily large number of organ systems involved in every patient from the onset of their illness and the high percentage (85%) of all biopsies that were positive.

A notable finding in this trial was the poorer survival of patients referred early. Sixty-seven patients referred within 3 months of diagnosis survived a median of 7.0 months from study entry as compared with a median of 16.0 months survival for the 33 patients referred between 3 and 12 months from diagnosis (P = 0.053). For each of these time periods, there was no significant difference in survival within treatment groups, but survival from study entry to organ death was significantly longer for MPC patients in the group

referred more than 3 months from diagnosis (18.7 versus 6.5 months for the C patients, P = 0.024). The number of organ systems involved at the time of study en-

try was essentially the same for patients in both time periods, totaling 4.4 for patients referred within 3 months of diagnosis and 4.6 for those referred within 3 to 12 months of diagnosis (P = 0.601). Two-thirds of the patients in this series were referred early and the speed of the referral may have reflected the primary physicians sense of urgency regarding the seriousness of the illness. Another possibility is that patients who survived 3 months before referral represent a biased group in which early deaths were already excluded. It is possible that in some reported series with longer median survivals the early referral patients have not been included.17,27

Although we recommend melphalan, prednisone, and colchicine for all patients diagnosed with AL amyloidosis, there is a need for a better treatment. The opportunity for treatment between diagnosis and death is brief, and methods that deliver <adequate chemotherapy early may be most beneficial. Recent experience with intravenous melphalan and autolo- gous stem cell rescue has promising early results.28~29 In this study, we did not examine the role of colchi- tine alone, which we previously reported by comparison with retrospective controls and complex statistical analysis. Likewise, we cannot conclude that melphalan, prednisone, and colchicine are superior to melphalan and prednisone alone, or that greater benefit would not have been achieved with higher total doses of melphalan. Perhaps future detailed comparisons with data from other trials will help to de- termine this. We did find that better surviv<al occurs in individuals for whom a satisfactory supportive treatment such as organ transplant or dialysis exists for their organ failure. Where feasible, these treat- ments should accompany treatment with MPC.

REFERENCES 1. Kyle RA, Bayrd ED. Amyloidosis: review of 236 cases. Medicine. 1975; 54:271-299. 2. lsobe T, Osserman EF. Patterns of amyloidosis and their association with plasma cell dyscrasia, monoclonal immunoglobulins and Bence-Jones proteins. NEJM. 1974;290:473-477. 3. Kyle RA, Greipp PR, O’Fallon WM. Primary systemic amyloidosis: multiiariate analysis for prognostic factors in 168 cases. Blood. 1986;68:220-224. 4. Gertz MA, Kyle RA. Response rates and survival in primary systemic amyloidosis. Blood. 1991;77:257-262. 5. Goldfinger SE. Colchicine for familial Mediterranean fever. NEJM. 1972;287: 1302-1306. 6. Meyers S, Janowitz HD, Gumaste W, et al. Colchicine therapy of the renal amyfoidosis of ulcerative colitis. Gastroenterology. 1988;94:150.?-1507. 7. Kyle RA, Greipp PR, Garton JP, Gertz MA. Primary systemic amytoidosis: comparison of melphalan/prednisone versus colchicine. Am J Med. 1985;79:708-716. 8. Cohen AS, Rubinow A, Anderson JJ, et al. Survival of patients with primary (AL) amyloidosis and the effect of colchicine treatment: case’s treated with colchicine from 1976-1983 compared with cases seen in previous years. Am J Med. 1987;82:1182-1190.


March 1996 The American Journal of Medicinee Volume 100 297


9. Gertz MA, Kyle RA. Phase II trial of alpha-tocopherol (Vitamin E) in the treatment of primary systemic amyloidosis. Am J Hematol. 1990;34:55-58. 10. Osserman EF, Sherman WH, Kyle RA. Further studies of therapy of amyloidosis with dimethylsulfoxide (DMSO). In: Glenner GG, Costa PP, Freitas AF, eds. Amyloid and Amyloidosis. Amsterdam: Exerpta Medica; 1980:563-569. 11. Gertz MA, Kyle RA. Phase II trial of recombinant interferon alfa-2 in the treatment of primary systemic amyloidosis. Am J Hematol. 1993;44:125-128. 12. Cohen HJ, Lessin LS, Hallal J, Burkhold P. Resolution of primary amyloldosis during chemotherapy. Ann Int Med. 1975;82:466-473. 13. Buxbaum JN, Hurley ME, Chuba J, Spiro T. Amyloidosis of the AL type. Am J Med. 1979;67:867-877. 14. Gertz MA, Kyle RA. Response of primary hepatic amyloidosis to melphalan and prednisone: a case report and review of the literature. Mayo Clinic Proc. 1986;61:218-223. 15. Schwartz RS, Cohen JR, Schrier SL. Therapy of primary amyloidosis with melphalan and prednisone. Arch Int Med. 1979;139:1144-1147. 16. Bradstock K, Clancy R, Uther J, et al. The successful treatment of primary amyloidosis with intermittant chemotherapy. Aust NZ J Med. 1978;8: 176-179. 17. Benson MD. Treatment of AL amyloidosis with melphalan, prednisone and colchicine. Arthritis Rheum. 1986;29:683-687. 18. Brown MP, Walls RS. Amyloidosis of immunoglobulin origin: useful treatment? Med J Australia. 1990;152:95-97. 19. Kyle RA, Greipp PR. Primary systemic amyloidosis: comparison of melphalan and prednisone versus placebo. Blood. 1978;52:818-827. 20. Zelen M. A new design for randomized clinical trials. NEJM. 1979;300:1242-1245.

21. Green MH, Harris EL, Gershenson KM, et al. Melphalan may be a more potent leukemogen than cyclophosphamide. Ann Int Med. 1986;105:360-367. 22. Gertz MA, Kyle RA. Acute leukemia and cytogenetic abnormalities complicating melphalan treatment of primary systemic amyloidosis. Arch fnt Med. 1990;150:629-633. 23. Cox DR, Oakes D. Analysis of Survival Data. New York: Chapman and Hall; 1984. 24. Hosenpud JD, DeMarco T, Frazier OH, et al. ProgressIon of systemic disease and reduced long-term survival in patients with celrdiac amyloidosis undergoing heart transplantation. Circulation. 1991;84(suppl lll):338-343. 25. Dubrey S, Falk RH. Heart transplantation in AL amyioidosis. Int J Exp Clin hvest In press. 26. Ferrante N, Cohen AS, Anderson JJ, Skinner M. Long survival in patients with AL amyloidosls. In: Natvig JB, Forre 0, Husby G, Husebekk A, Skogen 8, Sletten K, Westermark P, eds. Amyloid and Amyloidosis 1990. Dordrecht, Kluwer Acad Pub; Dordrecht, the Netherlands: Kluwer Academic Publishers. 1991:813-816. 27. Marinone G, Quaglini S, Bellotti V, et al. AL amyloiclosis: clinical and therapeutic aspects of an Italian study protocol. in: Kisilevsky R, Benson MD, Frangione B, Gauldie J, Muckle T, Young ID, eds. Amyloidosis. New York: Parthenon Press; 1994:206-208. 28. Comenzo RL, Simms RW, Vosburgh E, Skinner M. Advanced therapy for AL amyloidosis: IV melphalan with blood stem-cell rescue. Arthritis Rheum. 1995;38:S402. 29. van Buren M, Hene RJ, Verdonck LF, et al. Clinical remission after syngeneic bone marrow transplantation in a patient with AL alnyloidosis. Ann Int Med. 1995;122:508-510.

298 March 1996 The American Journal of Medicine” Volume 100

Treatment of 100 patients with primary amyloidosis: A randomized trial of melphalan, prednisone, and...

Documents

Transcript of Treatment of 100 patients with primary amyloidosis: A randomized trial of melphalan, prednisone, and...