InternatIonal MyeloMa FoundatIon
ASCO 2008
Highlights
for Physicians
Compiled by Lynne Lederman, PhD
This publication is sponsored by unrestricted educational grants from Celgene Corporation and Millennium: The Takeda Oncology Company.
ASCO 2008 Highlights for Physicians
News from the 2008 ASCO Meeting
collection. Dr. Rajkumar emphasized that high-dose dex as
a single agent and the combination therapy VAD (vincris-
New and updated data on novel and conventional therapies
tine/doxorubicin/dexamethasone) should not be used any
for patients with newly diagnosed or relapsed/refractory
more. He noted that thal/dex provides convenient oral dos-
multiple myeloma were presented at the 2008 meeting of
ing. Bortezomib is probably not an option for patients in
the American Society of Clinical Oncology (ASCO), May 30
remote areas who can't travel. The risk of DVT (deep vein
through June 3, in Chicago, Illinois. Presentations included
thrombosis) is higher for thal and lenalidomide (len) than
reports of data from clinical trials of novel agents in combi-
for other therapies. Neuropathy is a risk for all combina-
nation therapy, preliminary results of newer targeted agents,
tions other than len/dex. For patients with renal failure,
discussions of the role of stem cell transplantation (SCT),
lenalidomide, because it is excreted by the kidneys, may not
including the effects of novel agents on stem cell collection,
be the best option. Presentations on the use of lenalidomide
and continued research into how best to determine risk fac-
and bortezomib in patients with renal insufficiency are dis-
tors and to personalize therapy in an age of increasing use of
cussed below.
gene expression profiling (GEP) and other molecular tech-
niques. This year's educational session covered controver-
Dr. Rajkumar believes things are changing for patients who
sies surrounding the initial treatment of multiple myeloma,
are not candidates for transplantation. Until recently, MP
timing of high-dose therapy in myeloma, and risk-adapted
(melphalan plus prednisone) has been the standard of care
therapy of myeloma. There were also poster presentations
for this group of patients. MPT (MP plus thal) is better than
on these and other topics of interest to physicians treating
MP or MEL100, according to an IFM study. The recently
patients with multiple myeloma.
reported VISTA trial showed that btz plus MP (VMP) is better
than MP. To assist in the choice of treatment, Dr. Rajkumar
Newly Diagnosed Myeloma
suggested considering the route of administration and the
Controversies surrounding the initial treatment
patient's DVT risk. Both MPT and VMP have an associated
of multiple myeloma
risk of peripheral neuropathy (PN), and both regimens can
be used in patients with renal failure. VMP has no associated
In the educational session, S. Vincent Rajkumar, Mayo Clinic,
DVT risk but is administered intravenously (IV). The use of
Rochester, encouraged physicians to consider the patient as
MPT requires DVT prophylaxis.
a whole, and to take into account supportive care, quality
of life (QOL), and therapeutic toxicities, as well as the dis-
Although the results of trials showing efficacy of low-dose
ease itself. He discussed the choice of initial therapy, which
dex with lenalidomide should apply to its use in other
should be guided by whether or not a patient is a candidate
combinations, HD (high-dose) dex may be appropriate for
for transplant. For those who are transplant candidates, Dr.
patients with relapsed-refractory disease. Dr. Rajkumar also
Rajkumar believes that thalidomide plus dexamethasone
discussed the goal of therapy for myeloma. He believes that
(thal/dex) is not satisfactory because of toxicity. Newer
the goal of treatment should be complete response (CR),
options with equal or better response rates and lower toxic-
which may have some prognostic value. However, he cau-
ity may include lenalidomide plus lower-dose dexametha-
tions that patients should not be harmed by treating with
sone (Rd), bortezomib (btz) plus dexamethasone (dex),
more drugs to get to CR. Overall survival (OS), in addi-
other bortezomib-containing regimens, e.g., bortezomib
tion to the response rate (RR), is important. Some patients
with thalidomide (thal) plus dex (VTD). For patients who
with myeloma may do well if their disease is converted to
are transplant candidates, he noted that there are combina-
an MGUS-like state. CR is established by immunofixation,
tions such as bortezomib, cyclophosphamide, and dex that
which requires visually inspecting bands on a gel. This may
have high response rates yet do not compromise stem cell
result in determinations that are not precise or may not be
3
comparable between observers or trials. The critical role of
based on subsets of small numbers of patients. However, this
CR for patients with high-risk myeloma identified by GEP-70
analysis does raise an interesting question, namely, why is
is discussed below. Finally, Dr. Rajkumar discussed dura-
OS superior in spite of lower RR with Rd? This could be due
tion of therapy. This should be dictated by data obtained
to high toxicity with HD dex, leading to an early toxic death
from controlled clinical trials, tolerability of regimens, and
rate and toxicity-related reduction in duration of response
feasibility of using a given regimen on a long-term basis.
(DOR). Lower DOR results in less overall delivery of either
Some therapies can't be taken forever, although Rd lends
drug. Toxicity can decrease the ability of patients to tolerate
itself to long-term maintenance. He feels btz/dex and thal/
salvage therapies. This ultimately translates into a reduced
dex should be limited to pre-transplantation induction or
antitumor effect and a higher proportion of patients with
until maximum response is reached.
early deaths due to progressive disease (PD). Dr. Dhodapkar
agreed that the effect of Rd on survival with and without
Lenalidomide-based therapies
early SCT needs to be tested in a trial specifically designed
randomized trial of lenalidomide plus high-dose dex-
to address this issue, which will require a longer follow-up.
amethasone versus lenalidomide plus low-dose dex-
amethasone in newly diagnosed myeloma (e4a03), a
a randomized Southwest oncolog y Group study com-
trial coordinated by the eastern Cooperative oncolog y
paring dexamethasone (d) to lenalidomide + dexame-
Group: analysis of response, survival, and outcome
thasone (ld) as treatment of newly-diagnosed multiple
with primary therapy and with stem cell transplanta-
myeloma (ndMM): Impact of cytogenetic abnormali-
tion (abstract 8504)
ties on efficacy of LD, and updated overall study results
(abstract 8521)
Dr. Rajkumar presented a landmark analysis of data from
the ECOG E4A03 trial randomizing 445 patients with newly
Jeffrey A. Zonder, Karmanos Cancer Institute, Detroit, pre-
diagnosed myeloma to lenalidomide plus HD (standard)
sented results of SWOG S0232, which was originally planned
dex (RD, n=223) vs. Rd (n=222). After 4 months of treat-
for an enrollment of 500 untreated patients with an interim
ment, patients either went off study for SCT or continued
analysis after the first 300. This study was closed after an
on therapy. The purpose of this analysis was to determine
early interim analysis when 198 patients were randomized
if the high OS seen with Rd was due to the len/dex therapy
because of the release of the ECOG trial data. Follow-up
or to SCT. The landmark analysis showed that len/dex was
data are not yet complete. This study prospectively exam-
highly active in newly diagnosed myeloma regardless of the
ined the effect of abnormal karyotype (AK) and high-resolu-
patient age or transplant eligibility. Treatment with primary
tion chromosomal analysis (HRCA) (defined as deletion of
Rd continued beyond 4 cycles has high activity, with ORR
chromosomes 13 and/or 17 by FISH) on 1-year progression-
= 89%, including IF negative CR = 22%, VGPR/CR = 56%.
free survival (PFS) and OS. Dr. Zonder concluded that cyto-
The 2-year OS rate was 93% among patients who received
genetics still matter. Patients in either arm with cytogenetic
primary therapy with Rd. This is comparable to the original
abnormalities had inferior PFS and OS. Patients with AK
intent to treat (ITT) results in the Rd arm and to patients
treated with len/dex had lower PFS and OS compared with
who received RD or Rd for 4 cycles and then proceeded to
those without AK. Patients with AK treated with len/dex had
SCT. These results provide a rationale for future prospec-
higher PFS and OS compared to those on dex alone. FISH
tive trials comparing novel agents to SCT. Dr. Rajkumar sug-
HRCA as defined in this study did not seem to account for
gested that the most important question patients face today
this difference. The small sample size was also a concern.
in the era of new drugs is whether they still need a trans-
Bortezomib-based therapies
plant, a hypothesis that can be tested in the future.
Efficacy of induction with CyBorD in newly diagnosed
This presentation was discussed by Madhav V. Dhodapkar,
multiple myeloma (abstract 8517)
Yale University, who noted that it was important to remem-
ber that ECOG E4A03 was an induction trial with a primary
Craig B. Reeder, Mayo Clinic, Scottsdale
endpoint of RR and toxicity. OS was not a primary endpoint,
In this single-arm phase II study in patients with newly diag-
and 55% patients were off study within 6 months, so the
nosed myeloma, CyBorD (cyclophosphamide, bortezomib,
median duration of therapy was only 48 months. There-
dexamethasone) was used as induction therapy for 4 cycles
fore, one should be careful when looking at these outcomes
followed by ASCT. No survival data were presented. CyBorD
4
produced a rapid response as measured by percent reduc-
mum planned dose after phase I, with 10 patients at MTD.
tion of M-protein. Most responses occurred within the first
There were no unexpected toxicities, no grade 4 PN, a 5%
2 cycles of therapy. A response of at least very good partial
incidence of DVT/PE, and no treatment-related mortality.
response (VGPR) occurred in 71% of 28 evaluable patients
Response appears to be durable, but follow-up time is short
(of 33 enrolled) who completed 4 cycles of therapy. Near
(median 5 months). Time to progression (TTP), PFS, and OS
CR (nCR) plus CR occurred in 46% of these patients. All 21
have not yet been reached. Stem cells have been collected
patients in whom stem cell collection was attempted had at
from 15 patients, with data on 13, and engraftment data are
least adequate numbers of cells. Of 18 transplanted patients
pending. SCT in those 15 patients has had an unremarkable
who were evaluable, the CR/nCR rate was 72%. Ten patients
course to date. Responses by phase/dose level (subject to
were not transplanted. CyBorD-related grade 3 and 4 com-
confirmation) are summarized in Table 1.
mon toxicities include anemia, neutropenia, thrombocy-
topenia, hyperglycemia, diarrhea, hypokalemia, PN, and
Future plans include additional analyses for cytogenetics,
thrombosis. The trial has been re-opened to an additional
proteomics, and GEP. Len/btz/dex-based combination stud-
30 patients. The btz dose has been changed from twice to
ies are ongoing, including the phase I/II EVOLUTION trial
once weekly, and the dex dose is lowered after the first 2
of len/btz/dex and cyclophosphamide/btz/len/dex in newly
cycles. This regimen will be studied in a 3-arm (CRVD vs.
diagnosed patients. Phase III trials of len/btz/dex are in pro-
RVD vs. CyBorD), randomized trial of up-front therapy for
cess or planned.
myeloma.
Relapsed or Relapsed/Refractory
Regimens containing both lenalidomide and
Myeloma
bortezomib
Lenalidomide-based therapies
Safety and efficacy of lenalidomide (Len), bortezomib
Final analysis of MM-014: Single-agent lenalidomide in
(Bz), and dexamethasone (Dex) in patients (pts) with
patients with relapsed and refractory multiple myeloma
newly diagnosed multiple myeloma (MM): a phase I/II
(abstract 8524)
study (abstract 8520)
Mohamad. A. Hussein, Moffitt Cancer and Research Insti-
Paul G. Richardson, Dana-Farber Cancer Institute, Boston,
tute, Tampa, presented the final analysis of a single-arm,
discussed a phase I/II study defining the maximum tolerated
open-label, multi-center study to determine the efficacy
dose (MTD) and RR to len/btz/dex in previously untreated
and safety of len monotherapy in patients with relapsed/
patients who could proceed to autologous SCT (ASCT).
refractory myeloma. For the ITT population (n=222), the
Safety, response, and preliminary stem cell harvesting data
CR+PR rate was 26%, stable disease (SD) occurred in 48%,
were presented. Phase I and II enrolled 68 patients, 33 in
and the ORR was 44%. CR+PR for the efficacy-evaluable
phase I, 35 in phase II. Median age was 58 years, but the
population (n=183) was 32%, with 48% SD and ORR 51%.
patient population included individuals up to age 86 years.
Median time to first response (TTR) was 2.8 months and
In phase I there were two dose-limiting toxicities (DLT) at
median DOR was 13 months for both the ITT population and
dose level 4 related to the high-dose dex. Dose level 4M
efficacy-evaluable population. Median TTP was 5.2 months
was added based on the ECOG Rd study results. In phase
for the ITT population, and 10.4 months for responders
II, 35 patients were to be enrolled at the MTD or the maxi-
Table 1
renal impairment
none
mild
moderate
severe
CLCr
>80
50 CLCr 80
30 CLCr <50
<30
Patients
n=158
n=125
n=42
n=16
OR
64%
64%
62%
50%
% patients with
64%
75%
86%
improved renal function
TTP, months
11.3
12.1
11.4
7.9
PFS, months
11.1
10.6
11.1
7.8
OS, months
not reached
34.7
30.4
18.6
5
(CR+PR+MR, n=98). Progression-free survival (PFS) was
overall, with 67% CR+PR for len alone, and 78% CR+PR for
4.9 months vs. 10.4 months, and OS was 23.2 months vs.
len+dex. The median DOR was >36.4 months. The most
28.0 months. Median OS for the whole population was 23.3
common grade 3 or 4 AE was neutropenia (67%). No grade
months, with 37% of patients proceeding to HD therapy and
3 or 4 thrombocytopenia, anemia, PN, or DVT was reported.
SCT. The most common grade 3 and 4 AE were neutropenia
Dr. Richardson concluded that because a subset of patients
(60%), thrombocytopenia (39%). and anemia (20%), which
responding to len with or without dex could remain on
were managed by dose reduction or interruption of therapy,
treatment for at least 4.1 years, len can be considered as
and growth factor support. Only 4% of patients developed
an effective, durable, and well-tolerated therapy for heavily
febrile neutropenia and DVT, although anticoagulant ther-
pretreated patients with relapsed/refractory myeloma. This
apy was not mandated. No grade 3 or 4 treatment-related
long-term therapy requires close monitoring for AE and
PN was seen. Dr. Hussein concluded that len monotherapy
appropriate dose adjustment, which may optimize drug tol-
resulted in meaningful long-term benefit for patients with
erance and the potential to maintain a long-term response.
relapsed or refractory myeloma, and had a manageable
He also suggested that the role of len in maintenance ther-
safety profile.
apy may be important.
long-term responses observed with lenalidomide ther-
Effect of lenalidomide on renal function
apy for patients with relapsed or refractory multiple
The efficacy and safety of lenalidomide plus dexam-
myeloma (abstract 8525)
ethasone in relapsed or refractory multiple myeloma
Paul Richardson presented long-term follow-up of patients
patients with impaired renal function (abstract 8542)
remaining on therapy on the MM-007 study, which provided
Donna M. Weber, MD Anderson Cancer Center, Houston,
the platform for MM-009/010 phase III studies. ORR was
presented a poster on a subset analysis of pooled data from
25%, CR 4%, PR 13%, MR 9%. PFS was relatively short at
recent phase III trials, MM-009 and MM-010, in patients with
4.6 months, but OS at 27 months was described as impres-
relapsed or refractory myeloma who had varying degrees of
sive. Long-term analysis had not originally been planned,
renal insufficiency. Efficacy endpoints of TTP, PFS, and OS,
but has been performed in patients still on therapy after a
as well as safety were evaluated based on stratification by
median of 4.1 years. From May 2002 to July 2003, 102 were
creatinine clearance. Of 704 patients enrolled in study, 353
enrolled on the study, and 15 were still on therapy, which
patients randomized to receive len/dex were the subject of
is a very small number of patients. Most patients received
a subanalysis based on degree of renal impairment. Levels
once daily therapy, a few received twice daily dosing, and 9
of renal impairment were defined by creatinine clearance
patients also received dex. It had been determined early on
(CLCr) in mL per minute, and are defined in Table 2, which
that although the response was similar, the twice-daily dose
includes a summary of the results.
led to more myelosuppression and had an inferior safety
profile. No follow-up was done for patients who came off
There were more frequent dose reductions as renal impair-
the study regimen. OS was not included in the original fol-
ment increased. Thrombocytopenia was more marked in
low-up plan. All patients had len dose reductions or inter-
patients with renal impairment, so platelet counts should
ruptions, but all stayed on the study drug. CR+PR was 73%
be monitored carefully. Lenalidomide improved function in
Table 2
dose level
response-
Cr
nCr
VGPr
Pr
rr
Len/Bz/Dex
evaluable
Phase l
31
8 (26%)
14 (45%)
8 (26%)
30 (97%)
1
15/1.0/40
3
2
1
2
15/1.3/40
3
2
1
3
20/1.3/40
3
1
2
4
25/1.3/40
6
1
3
1
4M
25/1.3/20
16
2
9
5
Phase II
25/1.3/20
35
9 (26%)
7 (20%)
9 (26%)
10 (29%)
35 (100%)
TOTAL
66
17 (26%)
7 (11%)
23 (35%)
18 (27%)
65 (98%)
6
68% of patients with renal impairment by at least 1 level
bination Rev/Vel/Dex in patients with relapsed or relapsed/
within 4 months as assessed by peak CLCr response. Len
refractory myeloma (Dr. Kenneth Anderson was the senior
was associated with an improvement from any category
author). Secondary objectives were to evaluate objective
of renal impairment to the category of "no renal impair-
response rate (CR+nCR+VGPR+PR+MR), DOR, PFS, and
ment" in 47% of patients with renal impairment at baseline.
OS; to evaluate tolerability and toxicity; to identify possible
In patients with relapsed or refractory myeloma regard-
surrogate markers to better define the mechanism of action
less of renal impairment, lenalidomide is effective as mea-
(MOA) for the combination; to evaluate prospectively the
sured by response, TTP, PFS, and OS. Len improves OS to
markers of coagulation in patients with myeloma who are
a lesser degree in patients with severe renal impairment.
receiving therapy; and to correlate markers of hypercoagu-
Further trials are needed to define the efficacy and safety
lability with patient outcomes. Anticoagulation therapy with
of lenalidomide plus dexamethasone in patients with renal
daily aspirin, 81 mg or 325 mg, was required. In patients
impairment.
for whom there was any concern regarding thrombotic
risk, full-dose thromboprophylaxis was recommended with
Effect of bortezomib-based therapy on renal
either coumadin (target INR 2 to 3), or low molecular weight
function
heparin (LMWH) for the first 4 cycles followed by aspirin.
effect of peg ylated liposomal doxorubicin (Pld) plus
bortezomib in multiple myeloma patients with renal
Sixty-five patients were enrolled; of these, 64 patients were
insufficiency (abstract 8562)
treated and 55 patients were evaluable for response. The
best responses to date are 7% CR, 13% nCR, 38% PR includ-
Joan Blade, Hospital Clinic, Barcelona, presented a poster
ing 9% VGPR, and 20% MR. ORR is 78%. The median DOR in
on behalf of the DOXIL-MMY-3001 study investigators on
responding patients is 21 weeks. Median TTP, PFS, and OS
a retrospective subgroup analysis of a subset of patients
have not yet been reached. So far, 23% of patients discon-
with relapsed/refractory myeloma in the randomized phase
tinued due to PD, toxicity, or other reasons. Toxicities were
III trial comparing PLD plus bortezomib with bortezomib
mainly grade 1 and 2 myelosuppression, one grade 3 PN,
alone. Renal insufficiency was defined as CLCr <60 mL/min
two DVT, two grade 3 atrial fibrillation, one grade 3 rash.
(n=193); adequate renal function was defined as CLCr 60
One on-study death was possibly due to dexamethasone-
mL/min (n=453). PLD plus bortezomib resulted in signifi-
related fungal pneumonia. This combination is active in this
cantly longer median TTP compared with bortezomib alone
patient population, which includes patients with prior len,
(331 days vs. 190 days for inadequate renal function and
btz, thal, and SCT. Dex dose reduction was required and the
270 days vs. 190 days for adequate renal function). PLD
protocol was amended to a lower dose of dex. Surrogate
plus bortezomib was associated with higher response rates
studies and correlative studies are in progress.
in both groups than bortezomib alone. Steady, statistically
significant, and clinically meaningful improvement from
Discussant James R. Berenson, Institute for Myeloma and
baseline in mean CL
Bone Cancer Research, Los Angeles, felt that the somewhat
Cr was seen in patients in the renal
insufficiency subgroup treated with either PLD plus bort-
disappointing DOR of 21 weeks might reflect the length
ezomib or bortezomib alone. The occurrence of grade 3 and
of follow-up, and although the CR/nCR rates are low, they
4 anemia and diarrhea were higher in patients with renal
might improve with longer follow-up. He also wondered
insufficiency, but the incidence of hand and foot syndrome
what the cost of treatment relative to the benefit was, asking
was lower.
if it will it be better than less expensive alternatives includ-
ing VM, VMP, len/mel, BiRD, VTD, or others. Dr. Berenson
Regimens containing both lenalidomide and
asked if the drugs can be used sequentially with the same
bortezomib
benefit over the long run rather than all at once, and if this
Phase II study of lenalidomide (len), bortezomib
might provide a platform from which to add other agents,
(Bz), and dexamethasone (Dex) in patients (pts) with
e.g., cytotoxic agents, including cyclophosphamide, mel,
relapsed or relapsed and refractory multiple myeloma
PLD, or new targeted agents.
(MM) (abstract 8545)
Dr. Irene Ghobrial, Dana-Farber Cancer Institute, Boston,
presented a poster on an evaluation of TTP following com-
7
Stem Cell Transplantation
in the era of effective salvage therapies. She didn't think
that btz/dex should be considered the standard induction
Bortezomib/dexamethasone versus VAD as induction
treatment prior to ASCT to which other regimens including
prior to autologous stem cell transplantation (aSCt) in
novel agents should be compared at this time. To become
previously untreated multiple myeloma (MM): updated
the standard of care, a clinically relevant endpoint should be
data from IFM 2005/01 trial (abstract 8505)
shown, e.g., QOL, difference in cost-effectiveness, observing
Jean-Luc Harousseau, CHU Hôtel Dieu, Nantes, presented
that btz is expensive, and/or decrease in number of patients
the results of a phase III study of btz/dex compared with
going to second transplant. She noted however, that second
VAD as induction prior to autologous SCT (ASCT) in 482
transplant is not the standard of care in the US and many
patients aged <65 years. Consolidation therapy with two
other countries. Finally, an improvement in PFS and OS will
cycles of dex, cyclophosphamide, etoposide, and platinum
be required to see an effect of this regimen on patient care.
(DCEP) was also evaluated. All patients underwent MEL200
Bortezomib, peg ylated-liposomal-doxorubicin and
and ASCT. A second transplant, either ASCT or reduced-
dexamethasone (Pad) as induction therapy prior to
intensity conditioning (RIC) alloSCT, was performed for
reduced-intensity autologous stem cell transplant
patients not in at least VGPR after the first ASCT. Treatment
(aSCt) followed by lenalidomide and prednisone (lP)
and results are summarized in Table 3.
as consolidation and lenalidomide alone as mainte-
Btz/dex was superior to VAD for all ISS stages and for high-
nance (abstract 8518)
risk disease as measured by high beta 2-microglobulin lev-
Antonio P. Palumbo, University of Torino, presented the
els and chromosome 13 deletion. At this early follow-up,
results of PAD (btz, PLD, dex) as induction therapy prior to
differences between treatment arms are not significant for
ASCT compared with btz/dex, followed by len plus predni-
OS or PFS. There were no striking differences in AE, SAE,
sone (LP) as consolidation therapy, then single-agent len as
or discontinuation between arms. Stem cell collection was
maintenance therapy in patients aged 65 to 75 years. Younger
performed after G-CSF alone, and more patients in the btz/
patients who couldn't tolerate MEL200 were also enrolled.
dex arm needed a second collection. A study of btz/dex vs.
Cyclophosphamide plus G-CSF were used to harvest stem
btz/dex plus thal has just started.
cells. The response rate of 86 of 105 enrolled patients after
Discussant Suzanne Trudel, University of Toronto, agreed
PAD was 21% CR/nCR and 59% VGPR; after PAD MEL100
that failure to demonstrate improvement in outcome may
(n= 51) there were 59% CR/nCR and 88% VGPR. Response
reflect the need for longer follow-up, although CR may be
rates after PAD-MEL100-LP (n=34) were 70% CR/nCR
inadequate as a surrogate for OS for non-high risk patients.
and 88% VGPR, so LP was observed to increase the
Also, it may be difficult to demonstrate improvement in OS
response rate.
Table 3
arm
a1: Vad
a2: Vad+dCeP
B1: btz/dex
B2: btz/dex +DCEP
(n=110)
(n=96)
(n=112)
(n=96)
CR+nCR
9.1%
12.5%
23.2%
27.1%
VGPR
19.1%
28.1%
47.3%
55.2%
after first ASCT
n=185
n=197
(actually performed)
CR
11.4%
21.3 (p=.0051)
CR+nCR
27.6%
39.6 (p=.01)
VGPR
51.9%
71.6 (p<.0001)
PR
93.5%
95.4 (NS)
response to
n=87
n=55
second ASCT
CR+nCR
9%
7%
VGPR
24%
36%
8
Dr. Palumbo suggested that age 70 years might be the upper
monitis, and hypotension. In phase II, the response rate
limit for this regimen. Most AE were seen in the first two
must be at least 50% for further testing. There will be an
courses, when the most cytoreduction occurred. Grade 3
interim analysis for response. Accrual for phase II is planned
and 4 AE included 12% neutropenia, 15% thrombocytope-
for 53 to 75 patients. Of 23 patients treated in phase I, there
nia, 15% PN, 15% infections, 6% DVT, and 4% early deaths.
were 2 CR, 9 VGPR, and 8 PR after high-dose melphalan,
The projected event-free survival (EFS) at 2 years is 83% and
and 10 CR, 7 VGPR, and 6 PR after tandem ASCT at a median
the projected OS at 2 years is 92%. PAD-MEL100 is feasible in
follow-up of 12 months. Dr. Somio concluded that image-
patients age 65 to 75 years, but longer follow-up is needed
guided intensity-modulated TMI using helical tomotherapy
to determine upper limit of patient age. LP consolidation
delivers targeted radiation while decreasing organ doses to
further improves the response after ASCT. In this regimen,
unintended targets. Discussant Dr. Berenson found it dis-
len is a new drug to which patients have not been exposed.
concerting that there were two deaths among four patients
Decreasing the frequency of administration of btz to weekly
who had a relapse, occurring at 5 and 13 months after trans-
may reduce the incidence of PN. This sequential regimen
plant, which is a short follow-up period. In addition, the
may become a new strategy, especially for patients who can-
response was over 80% even prior to HD mel, although CR
not tolerate full-dose ASCT.
rate increased after ASCT.
application to high-dose therapy (Hdt) and autolo-
Effects of Novel Agents on Stem Cell Collection
gous transplantation: timing of high-dose therapy in
Some prior studies showed a decrease in CD34+ cells,
multiple myeloma
a decrease in average daily collection, an increase in the
In the educational session, Jean-Paul Fermand, Hôpital
number of aphereses, and/or failure to collect (FTC) suf-
Saint-Louis, Paris, described HD therapy and ASCT (HDT)
ficient stem cells for even one transplant after treatment
as the standard of care, noting that there are trial data that
with lenalidomide. Two poster presentations addressed this
both support and fail to support this modality. Younger
issue, and one presentation examined the effect of borte-
patients and those with better renal function may benefit
zomib on stem cell collection.
more, and more recently diagnosed patients appear to have
Impact of lenalidomide therapy on stem cell mobiliza-
a higher 10-year survival rate because of the introduction of
tion in myeloma (abstract 8543)
novel therapies. He does not think there are sufficient data
to support the risk and cost of tandem transplants, which
Harshita Paripati, Mayo Clinic, Scottsdale, presented a ret-
need further study. Other studies that Dr. Fermand sug-
rospective chart review of 61 patients undergoing stem cell
gested that are still needed include addition of btz to HDT
collection between January 2005 and October 2007. Records
with MP, confirmation of molecular markers associated with
were analyzed for type and duration of induction therapy,
poor prognosis in the era of novel agents, and the role of
collection efficacy, and engraftment in patients who went
thal maintenance therapy.
to transplant. G-CSF was used for stem cell mobilization.
Failure to collect (FTC) occurred in 45% of patients receiv-
a new tandem autologous peripheral blood progenitor
ing len-based induction therapy vs. 7.3% of patients receiv-
cell transplant (at) approach incorporating total mar-
ing other induction therapies (P=.0011). Re-collection in
row irradiation (tMI) for patients (pts) with multiple
the majority of len-treated patients when combined with
myeloma (MM) (abstract 8544)
the first collection provided cells sufficient for at least one
George Somlo, City of Hope Cancer Center, Duarte, pre-
transplant. Dr. Paripati suggested that other strategies for
sented a poster on results of a phase I/II study to test the
mobilization of stem cells following lenalidomide induc-
feasibility of using sequential tandem ASCT with melphalan
tion, such as cyclophosphamide or AMD-3100 (plerixafor)
and TMI for patients with responding or stable state I, II,
along with G-CSF, may be considered in patients treated
or III myeloma, followed by thal/dex maintenance for 12
with lenalidomide. A lenalidomide "holiday" prior to stem
months or for 6 months post CR. Stem cells were collected
cell collection may also be considered. Limiting the num-
with G-CSF and cyclophosphamide. The MTD of TMI was
ber of monthly cycles of lenalidomide prior to mobiliza-
1600 cGy, and is being used for the phase II dose. DLT at
tion may be optimal. Discussant Dr. Berenson wondered if
1800 cGy were reversible, radiation-induced enteritis, pneu-
lower doses, less frequent dosing, fewer cycles, and/or drug
9
holidays prior to collection could overcome the problem
patients treated with either len or with non-len-based regi-
without affecting post-transplant outcome. Given that G-CSF
mens. These results also might be different in a larger study
was used alone, he asked if other mobilization agents, e.g.,
population of patients.
cyclophosphamide or plerixafor, could improve collection.
He also wondered if these results should raise concerns
Bortezomib, cyclophosphamide, and filgrastim leads
regarding marrow health related to subsequent myelosup-
to high-yield Cd34+ stem cell collections in multiple
pressive therapies.
myeloma (abstract 8587)
lenalidomide and stem cell collection in patients with
Ruben Niesvizky, Weill-Cornell Medical College and New
multiple myeloma (abstract 8547)
York Presbyterian Hospital, New York, presented an open-
label study to maximize pre-transplant response and success-
Rachel I. Cook, University of Pennsylvnia Abramson Cancer
ful mobilization and harvest of stem cells. Patients received
Center, Philadelphia, also discussed concerns regarding the
btz/dex with or without PLD. Of 24 patients accrued, 16
ability to mobilize stem cells due to the effect of lenalido-
have been mobilized and 10 completed transplant. The goal
mide on bone marrow. This single-institution retrospective
of at least 10 x 106 CD34+ cells/kg in fewer than 7 leuka-
analysis of 290 patients compared collections using G-CSF
phereses was exceeded. Median collection was 21.3 x 106
alone with collections using G-CSF/cyclophosphamide (cy).
CD34+ cells/kg in a median of 1.5 collection days. Of 23
Initial treatment did not include a len/dex-based regimen in
response-evaluable patients, 70% had at least PR, and 48%
247 patients. Of these, FTC occurred in 3 of 184 patients fol-
had CR/nCR. Post-transplant, the median time to ANC recov-
lowing one regimen, and FTC occurred in 7 patients follow-
ery was 11 days. The median time to platelet recovery was
ing multiple regimens. The other 4 FTC were in 43 patients
18 days.
following len/dex only (1 patient) or len/dex as one of mul-
tiple regimens (3 patients). Limitations of this study were
Risk-Adapted and Individualized
that toxicity data for patients who are mobilized with cy are
Approaches to Therapy
not available for this sample of patients; interpretability of
risk-adapted therapy of multiple myeloma
P-values is limited by the small patient sample size in some
comparisons; and it is a retrospective, non-randomized
Keith A. Stewart, Mayo Clinic, Scottsdale, discussed the
study. There was no correlation found between the num-
need to consider unique features of individual patients to
ber of cycles of len/dex and the number of collections or
tailor their therapy in practice, rather than just using these
the number of stem cells collected in patients. Len induc-
features to estimate their prognosis. He reviewed the high
tion therapy in this study does not prevent collection of
genetic variability of myeloma, describing at least 9 genetic
adequate numbers of stem cells for transplant when G-CSF
events associated with the development of myeloma, each
and cy are used together for mobilization. The median stem
of which confers a different prognosis. He emphasized that
cell yields are lower for patients who have taken lenalido-
it is no longer acceptable to think of myeloma as a single
mide compared with those patients on non-lenalidomide
disease entity, nor as a stable disease, because multiple
containing regimens, but the difference is not clinically sig-
mutations occur over time. Although patients can be classi-
nificant. G-CSF/cy mobilization increases stem cell yield and
fied as having high- or standard-risk disease, some therapies
decreases the number of collections compared with G-CSF
may abrogate high-risk profiles. An example of this is bort-
alone. An increased failure to collect rate is seen in patients
ezomib, which, as part of VMP, overcomes the high risk asso-
mobilized with G-CSF compared with G-CSF/cy, regardless
ciated with t(4;14), t(14;16), or 17del. Dr. Stewart believes
of the induction regimen. The number of cycles of len does
that patients with high-risk disease merit being subjects in
not correlate with stem cell yield.
their own clinical trials.
Discussant Donna M. Weber thinks the lack of correlation
Gene expression profiles defining molecular sub-
with len and numbers of aphereses or yield of stem cells /
types, coupled with signatures of tumor biology and
failure to collect may change if larger numbers of patients
chemotherapy sensitivity, provide a novel therapeutic
are analyzed. Although cy does significantly reduce num-
approach to multiple myeloma (abstract 8501)
ber of aphereses and increase the stem cell yield, it does
Ariel Anguiano, Duke University, discussed good and
not decrease the number of failed stem cell collections in
poor prognostic profiles based on patterns of oncogenic
10
deregulation. GEP reflecting hyperploidy (HY ) or specific
in TT3 demonstrated the reproducibility of the patterns.
genetic lesions leading to translocations, including C-MAF,
This analysis has shown distinct patterns of altered tumor
MAFB, CCND1, CCND3 and MMSET (MS), have been used
biology and chemotherapy sensitivity among the clinically
to define 7 distinct phenotypes of multiple myeloma.
relevant subtypes of multiple myeloma.
They are:
Dr. Anguiano concluded that patterns of oncogenic path-
1. HY, with low CCND1, high Wnt inhibitors FRZB and
way deregulation and altered tumor environment may fur-
DKK1; OS=84% at 3 years
ther differentiate molecular subtypes of multiple myeloma.
2. Low bone disease (LB) HY determined by aCGH, with low
Genomic analysis based on molecular subtype rather than
DKK1, CCND2 activation by an unknown mechanism;
clinico-pathologic stage may provide more clinical insight
OS=87% at 3 years
into the biology of disease progression and treatment. Evi-
3. CCND1/CCND3 (CD-1), with t(11;14) or t(6;14) activat-
dence of altered tumor biology may provide an opportunity
ing CCND1 or CCND3, respectively; OS=81% at 3 years
for selection of molecular targets. Genomic signatures of
chemotherapy sensitivity may help guide therapy in mul-
4. CCND1/CCND3 (CD-2), with t(11;14) or t(6;14) activat-
tiple myeloma in the future.
ing CD20 expression; OS=88% at 3 years
This presentation was discussed by Wee J. Chng, National
5. MMSET/FGFR2 (MS), with t(4;14) and CCND2 activation;
University Hospital, Singapore, who said that it will be nec-
OS=71% at 3 years
essary to have robust means to dissect patient and tumor
6. MAF/MAFB (MF), with t(14;16) or t(14;20), and CCND2
heterogeneity to select the right treatment for the right
activation; OS=71% at 3 years
patient to achieve maximal efficacy with minimal toxicity.
7. proliferation (PR) with a proliferation signature; OS=55%
The current study is not at a level where this technology can
(may include any of above subtypes)
be embraced. To move forward, data should be validated
for oncogenic pathways using independent technologies,
Several of these subtypes (5, 6, 7) proffer higher risk of
e.g., flow cytometry, to see what level of activity corresponds
progression and death, emphasizing the need for novel
with true biologic activity. For determining drug sensitivity,
therapeutic strategies. Molecular subtyping employed gene
the process must be validated in patient cell sets and should
expression analysis in a cohort of 560 newly diagnosed
include pathways relevant to agents used to treat myeloma,
patients with multiple myeloma (training n=352, valida-
e.g., involving NF kappa B and IL-6. It is important to create
tion n=208). Expression signatures reflecting deregulation
more relevant databases and/or signatures for analysis and
of oncogenic signaling pathways (Myc, Ras, Src, B-catenin,
to validate these strategies for different tumors and treat-
E2F1 and PI3K), profiles representing altered tumor
ments. For high confidence in therapeutic candidates, pro-
microenvironment (chromosomal instability, angiogen-
spective studies in a trial setting will be required to confirm
esis, hypoxia and TNF-a), and signatures of chemotherapy
the clinical utility of these strategies.
sensitivity (melphalan, vincristine, Adriamycin, cyclophos-
phamide, cisplatin, and etoposide) were determined in
total therapy (tt) for myeloma (MM)--10% cure rate
a training cohort of patients previously enrolled in Total
with tt1 suggested by >10yr continuous complete
Therapy 2 (TT2). Expression patterns and observed dis-
remission (CCR): Bortezomib in TT3 overcomes poor
tinct patterns of altered tumor biology and chemosensitivity
risk associated with t(4;14) and deltP53 in tt2
between the molecular subtypes (CD1, CD2, LB, HY, MF,
(abstract 8516)
MS and PR) were compared. The low-risk HY group had
Bart Barlogie, University of Arkansas for Medical Sciences,
increased activation of Myc (P<0.0001) and PI3K (P =0.03)
Little Rock, reviewed the history of the TT1, TT2 ± thal,
pathways when compared with the high-risk MS group. The
and TT3 btz-thal regimens, all developed with the strategy
MS group had increased evidence of chromosomal instabil-
of using all agents and all modalities with efficacy against
ity (P =0.0002). The HY group had increased sensitivity to
myeloma to reduce the tumor burden maximally up front,
vincristine (P =0.04) and Adriamycin (P =0.003), whereas
rather than waiting for relapse to use second- and third-line
the MS group had increased sensitivity to cyclophosphamide
agents. Since 1989, 1200 patients have been enrolled in suc-
(P <0.0001) and to cisplatin (P =0.02). A similar analysis
cessive TT protocols, with steady improvement in outcomes
in an independent validation set of 208 patients enrolled
11
accompanying transitions from TT1 to TT2 to TT3. GEP can
use of high-density SnP-array analysis to identify novel
be used to define low-risk myeloma in 85% of patients. For
chromosomal abnormalities that predict survival in
these patients, TT3 results in better outcomes than TT2.
multiple myeloma (abstract 8522)
However, in the 15% of patients defined as having high-risk
disease, despite high CR rates, TT3 and TT2 have similar
Herve Avet-Loiseau, Institut de Biologie, Nantes, on behalf
poor outcomes. Dr. Barlogie was surprised to see that the
of the Intergroupe Francophone du Myelome (IFM), dis-
best survival is in patients treated with TT2, resulting in CR
cussed a study of DNA extracted from plasma cells of a
sustained for 3 years or longer. If CR is lost within 3 years,
cohort of newly diagnosed patients (n=192) younger than
outcome is poor, and is worse than for those who never
age 65 years, homogeneously treated with HD mel in the
had CR in the first 3 years. For patients treated with TT3,
IFM 99 trials, with a median follow-up of 60 months. The
the hazard of relapse declines more rapidly in high-risk vs.
study used Affymetrix 500K SNP arrays. After labeling, DNA
low-risk myeloma. If patients with high-risk disease survive
was hybridized on the two Nsp and Sty chips, containing
3 years, then their chance of survival will be more like that
260,000 SNPs each. As expected, copy number variations
of patients with low-risk disease.
were observed in all the cases. The most frequently involved
chromosomes were chromosomes 13, 16, 1, 8 and 6. In a
Dr. Barlogie is looking toward development and use of TT4
prognostic analysis to correlate survival and chromosomal
and TT5. For patients with low-risk myeloma, it is difficult
abnormalities, many previously unreported chromosomal
to improve upon TT3 results within the time span of an NCI
regions were identified. The most frequently lost regions
grant, he observes. For patients with low-risk disease, TT4
associated with overall survival were 1p, 6q, 8p, 12p, 14q,
is designed to reduce toxicities. In a trial, patients will be
16q (in 20% of patients), 17q, 20p, and 22q. Global analy-
separated by GEP risk and will receive TT3 vs. TT3-lite (one
sis revealed a marked heterogeneity in the gains observed
induction, one consolidation, pulsed MEL). For patients
in hyperdiploid karyotypes, explaining the inconsistency in
with high-risk disease, the goal is to sustain CR with a dose-
the literature regarding the prognostic value of hyperdip-
dense but less dose-intense therapy (TT5). TT5 is based
loidy. In a unichromosomal prognostic analysis, good prog-
on MEL80-VTD-PACE, that is 8 drugs in lower doses, then
nosis was associated with hyperdiploidy of chromosomes
alternating R-VD or M-VD as maintenance to avoid bone
5, 9, 15, 19. Poor prognosis was associated with gain of 1q,
marrow toxicity. Dr. Barlogie believes that CR duration is
loss of 8p, 11q, del 12p, del 14q, del 16q, abnormality of
the best surrogate for survival, although the most difficult
22. A multivariate prognostic analysis to see if abnormalities
one to measure in terms of the time needed for observa-
were related identified three abnormalities that were inde-
tion, compared with observing the CR rate. He concludes
pendent, i.e., gain of 1q, gain of chr 5, and del 12p. These
that cure is a realistic outcome in myeloma with a >10 year
were confirmed by resampling and permutation analysis. By
sustained CR for TT1, which may be shorter for TT2. Late
multivariate analysis, hazard ratios for these abnormalities
relapse in some patients shouldn't be an argument against
were 1.9, 0.37, and 2.32 respectively, all of which were sig-
curability, he says, because this happens in all cancers,
nificant. For patients with no risk factors, OS at 5 years was
including childhood leukemia. Bortezomib has changed
about 80%; for those with 2 or 3 risk factors, median sur-
the outlook of myeloma subgroups. The presence of t(4;14)
vival was 2 years. Beta-2M and t(4;14) were then included
does not confer a poor prognosis regardless of GEP risk
as variables in a multivariate analysis. Hazard ratios were
group, and delTP53 no longer confers poor prognosis in
as follows: beta-2M= 2.78, gain of chr 5= 0.37, del 12p
low-risk disease.
=3.17. All of these were significant; however, t(4;14) and
1q gain were no longer significant in this analysis. There
Kenneth C. Anderson discussed this presentation, calling it
were only two recurrent regions with double deletions, chr
a landmark presentation, and agreeing that the concept of
14 (TRAF3, in 4.7% of patients) and chr 16 (CYLD, in 2.6%
talking about the cure of myeloma is here. The new stud-
of patients). These genes previously have been reported as
ies described will ask whether high-dose therapy is needed
negative regulators of the NF kappa B pathway. Dr. Avet-
for patients with low-risk disease when incorporating bort-
Loiseau concluded that expression arrays should enhance
ezomib into the regimen.
understanding of myeloma oncogenesis and ultimately lead
to an individual therapeutic approach.
12
Complete remission and survival in multiple myeloma
analyses at 1, 3, and 7 years of survival by duration of CR
(abstract 8523)
show that CR is generally associated with longer survival
time, and that the longer patients remain in CR, the longer
Michael Wang, MD Anderson Cancer Center, Houston, pre-
they survive. Significant variables independently associated
sented a retrospective review of outcomes among newly
with survival include CR vs. NR, PR vs. NR, stage I vs. stages
diagnosed patients who received the sequence of primary
II or III, and HDT vs. no HDT. CR is the strongest indicator
chemotherapy followed by intensive therapy within the first
for longest time of survival and represents the major surro-
year. He applied landmark and multivariate analyses to inves-
gate marker for long-term survival. CR, however achieved,
tigate whether CR can be used as a major surrogate marker
was associated with longer survival (median 12 years) than
for long-term survival. Clinical outcomes were assessed in
other response outcomes. ASCT in CR did not further pro-
758 consecutive patients treated at MD Anderson from Janu-
long survival.
ary 1987 to April 2007, who were age 65 years or younger,
and received dexamethasone-based induction therapies.
In his discussion of this presentation, Dr. Rajkumar described
Of these, 395 patients (52%) also received HD-melphalan-
it as having the perfect type of statistics to show CR is an
based intensive therapy supported by ASCT within the first
excellent prognostic factor. He cautioned that there are
year. This new statistical approach included landmark analy-
caveats, because even the best regimens available now give
sis at 12 months of OS, response status, HD chemotherapy,
CR rates of only 30% to 40%, yet a lot of patients who don't
and a Cox regression (multivariate) analysis. Patients in the
have CR can have a long survival. Therefore, it is important
group receiving transplants (n=395) had a significantly
to see how these patients can be identified. For any regi-
younger median age (age 52 years) than those who did not
men, even if it is ineffective, there will be responders who
receive transplants (n=363, age 56 years). The response
do better than non-responders. Also, although retrospec-
rate among all patients was 66% (including CR in 8% and PR
tive studies aren't as good as prospective ones, a landmark
in 58%) after induction chemotherapy alone, which means
analysis helps overcome this statistical problem. Although
that 34% of patients had primary resistant disease. Dr. Wang
CR is desirable, is very prognostic, and is a prerequisite of
noted that the population of patients with primary resistant
cure, the way it is currently defined, CR is not indicative of
myeloma is disappearing with the advent of novel therapies.
cure. This is because, as Dr. Rajkumar previously pointed
Of the 59 patients in CR, 34 received ASCT and remained
out, immunofixation is not an adequate measurement. CR
in CR. Of 440 patients in PR, 256 received ASCT; of these
may be a good predictor of OS in many but not all trials. Dr.
24% converted to CR and 76% remained in PR. ASCT was
Barlogie has made the point that myeloma is often preceded
used to treat 106 of the 259 patients with primary resistant
by MGUS, which is present in 3% of the US population. This
disease. In this patient group, 11% converted to CR, 53%
represents 10 million people who are doing well. That sug-
converted to PR, and 36% remained non-responsive (NR).
gests that a subset of patients with myeloma don't need to
CR+PR in primary resistant disease after ASCT was 66%, so
be in CR, they could be in VGPR and still do quite well. They
ASCT changed the response status. Dr. Wang concludes that
won't benefit from getting a little M-spike reduced to zero,
patients with primary resistant myeloma have a more urgent
and may actually be harmed by it.
need for ASCT.
New Therapies and Targets in
For primary therapy alone with no ASCT, landmark analy-
Development
sis at 12 months predicted significantly longer survival for
patients whose disease was in CR (median 9.6 years), than
Targeting Cdk4/6 in combination therapy of chemore-
for those with PR or NR (median 4.0 and 2.2 years, respec-
sistant multiple myeloma (abstract 8503)
tively). For patients who received ASCT, survival with land-
Selina Chen-Kiang, Weill-Cornell Medical College, New
mark analysis is also longer than for patients with PR or NR.
York, discussed preclinical studies targeting the cell cycle,
For patients with PR after primary therapy, those who con-
which she believes is absolutely necessary to control aggres-
vert to CR after ASCT have significantly longer survival than
sive tumor growth and relapse and stem cell reactivation.
those who remained in PR. There are four pathways to CR:
She described the rationale for targeting Cdk4/6, which is
primary therapy alone CR, CR after primary therapy plus
deregulated in nearly all cancers and may be important at
CR after ASCT, PR to primary therapy then CR after ASCT,
times when myeloma is active and even more so at relapse
and NR after primary therapy then CR after ASCT. Landmark
13
and refractory relapse. PD 0332991, now in clinical trials,
deacetylase inhibitor, in combination with btz for up to 8
is a potent specific reagent and induces exclusive G1 arrest
cycles in relapsed myeloma (n=23). The MTD was 400 mg
in the absence of apoptosis, distinguishing it from other
vorinostat daily on days 4 to 11 of 21-day cycles (rather than
Cdk4/6 inhibitors. It can be combined with other agents.
8 days of therapy as in other trials) plus btz at the standard
Strategies for use involve adding an agent like bortezomib
dose. Of 21 patients evaluable for response, there were
to kill cells after cell cycle synchronization or after sustained
two with VGPR. DLT at 500 mg vorinostat were fatigue and
G1 arrest induced by PD 0332991. PD 0332991 plus btz plus
prolonged QTc. Most of the AE were hematologic, although
dex is in phase I/II clinical trials.
hyponatremia also occurred, which requires monitoring.
The pharmacokinetics of vorinostat in combination with
Phase I trial of vorinostat plus bortezomib (btz) in
bortezomib were similar to those of single-agent vorinostat.
relapsed/refractory multiple myeloma (mm) patients
The combination appears promising in myeloma previously
(pts) (abstract 8548)
refractory to bortezomib alone.
Ashraf Z. Badros, University of Maryland Greenebaum
Cancer Center, Baltimore, presented results of a phase I
dose-escalation trial to identify the MTD and toxicity of vor-
Lynne Lederman, PhD, is a medical writer based in
inostat (SAHA, suberoylanilide hydroxamic acid) a histone
Mamaroneck, New York
14
International Myeloma Foundation
12650 Riverside Drive, Suite 206
North Hol ywood, CA 91607 USA
Telephone:
800-452-CURE (2873)
(USA & Canada)
818-487-7455
Fax: 818-487-7454
TheIMF@myeloma.org
www.myeloma.org