Supplemental Material

Table of Contents

Supplemental Statistical Considerations Page 2

S1: CALGB 100104 CONSORT Flow Diagram Page 8

S2: Time to Progression and Overall Survival, crossover within 6 or 12 months Page 9

S3: Survival after Progression, crossover within 6 or 12 months Page 10

S4: Time to Progression and Overall Survival, stratified by prior lenalidomide exposure Page 11

S5: Time to Progression and Overall Survival, stratified by response at randomisation Page 12

S6: Time from Randomisation to SPM diagnosis Page 13

S7: Box plot of SPMs Page 14

S8: Cumulative incidence risk analyses of haematological and solid tumour SPMs Page 15

S9: Cumulative incidence risk analyses of lenalidomide vs placebo (non-crossover only) patients Page 16

S10: Cumulative incidence risk analyses of lenalidomide vs placebo (crossover only) patients Page 17

S11: Event-free Survival Page 18

Table S1: Follow-up times Page 19

Table S2: Characteristics of the patients Page 20

Table S3: Adverse events associated with treatment discontinuation Page 22

Table S4: TTP/PFS analyses with different censoring rules Page 24

Table S5: Summary of first salvage therapy Page 25

Table S6: Complete listing of initial salvage therapies Page 26

Table S7: > Grade 3 Adverse Events Page 27

Table S8: Haematologic Second Primary Malignancies Page 33

Table S9: Solid Tumour Second Primary Malignancies Page 34

Table S10: Second Primary Malignancies Diagnosed after Myeloma Progression Page 35

Table S11: Second Primary Malignancies Diagnosed in non-Randomised Patients Page 36

Table S12: Enrolling sites Page 37

1
STATISTICAL CONSIDERATIONS

Primary Objective
The primary endpoint of this study is the time to progression (TTP), starting at transplant, with progression defined in the
protocol document, as follows:
Progressive disease (for patients not in complete response) requires one or more of the following:
• >25% increase in the level of the serum monoclonal paraprotein, which must also be an absolute increase of at
least 0.5 g/dL and confirmed by at least one repeated investigation.
• >25% increase in the 24-hour urinary light chain excretion, which must also be an absolute increase of at least 200
mg/24-hour and confirmed by at least one repeated investigation.
• >25% increase in plasma cells in a bone marrow aspirate or trephine biopsy, which must also be an absolute
increase of at least 10%.
• Definite increase in the size of existing bone lesions or soft tissue plasmacytomas.
• Development of new bone lesions or soft tissue plasmacytomas (development of a compression fracture does not
exclude continued response).
• Development of hypercalcemia (corrected serum calcium > 11.5 mg/dL or 2.8 mmol/L) not attributable to any
other cause in Section 11.2.5 of the protocol document, in MM patients. Time to progression will start at Day 0 of
start of transplant. Any death will be considered as an event (i.e., will not be censored). We hypothesize that the
treatment arm will be superior with respect to this endpoint.

Throughout these discussions, we assumed that the laws of the time-to-event in both arms are exponential. The hypothesis of
interest may be canonically presented as testing H0: Δ=1 versus the general alternative H1: Δ >1, where Δ denotes the
treatment to control hazard ratio. In particular, we hypothesize that the median TTP is Mc=2 years (24 months) for the
control arm and Me=2.8 years (33.6 months) for the treatment arm. This corresponds to testing H0: Δ=1 versus the local
alternative H1:Δ=1.4.

Given that the observed accrual pattern has differed considerably from that assumed in the original design, an amendment to
the statistical considerations was necessitated so as to comply with the CTEP low accrual policy. Furthermore, the TTP
distribution assumptions in this amendment and the targeted number of events (309) are identical to those of the original
design. What is subject to amendment is only the assumed distribution of the administrative censoring mechanism. As such,
amendment should have no effect on the risk to benefit ratio of the original design.

Secondary Objectives
The rates for the various types of responses pre-randomization, at 3 months and at 12 months will be estimated. The
improvement in response (e.g., PR to CR) over the aforementioned time-points will be analyzed as well. The potential
discrepancies between the TTP profiles for the CR and non-CR group on the study drug arm will be described. We expect
little differences in patient responses by sex or race.

The marginal survival distributions for time to progression, overall survival and EFS are estimated using Kaplan-Meier
estimators.1 Time-to-event distributions between the two arms are compared using the log-rank test for univariable analysis
and the Cox score test2 for multivariable analysis adjusting for baseline patient characteristics. The corresponding effect size
is quantified on the basis of the hazard ratio under the framework of a proportional hazard model. To estimate and compare
cause-specific hazard (progression, death and second primary malignancy), marginal estimators of the corresponding
cumulative incidence profiles3 and the log-rank test proposed by Gray4 are used. To test whether there is an interaction
between a baseline co-variable and randomization arm, a two-way multiplicative Cox model is employed. A forest plot is
used to provide a graphical presentation of the absolute and relative effect sizes. For this plot, the effect size is presented as
the log hazard ratio (i.e., the regression coefficient from the Cox model). The radii of the circles are proportional the inverse
of the square of the standard error. It is noted that all hazard ratios discussed in the text and illustrated in the forest plot have
been quantified within the context of univariable Cox model without accounting for other additive or multiplicative effects.
For all of time to event analyses, standard asymptotics is employed to characterize the sampling distributions for the statistics
and estimators. As a descriptive analysis, the discrepancies between the response at time of randomisation profiles of the two
arms were assessed using Fisher’s exact test.5 For the progressive disease events, patients for whom no event had been
realized were censored at the last documented clinical assessment date at which they were found to be in remission. For the
overall survival endpoint, patients who are alive are censored at the last date of follow-up. Some patients withdrew consent to

2
further follow-up. For these patients, only follow-up information provided on or before the date of withdrawal of consent was
used.

While the progressive disease and second primary malignancy events are subject to interval censoring, in the analyses, they
have been right censored to the date of clinical assessment. The statistical analyses are carried out using the R statistical
environment (version 3.3.2) along with the survival (version 2.39-5) and cmprsk (version 2.2-7) extension packages.

Randomisation/blinding: The blinding table was generated by CALGB IS and was kept on a secure server. Only the study
statisticians had access to the unblinding application. The Executive Officer authorized unblinding in case of an emergency.
At that point either Vera Hars or a member of the IS department could unblind the treatment via an unblinding application.
Vera Hars had access to the randomisation table and received notice about each randomisation which included the
stratification factors of the newly randomised patient. As per CALGB guidelines regarding randomised double-blinded
studies, study statisticians confirmed that the patient was assigned to the correct treatment (the next treatment of the current
block of six treatments within his/her stratum) for the first 10-20 patients.

Interim analyses/stopping rules/DSMB monitoring: As per the protocol: A group sequential test design due to Emerson
and Fleming6 will be used to stop the trial early for superiority of the experimental drug. The study will also be monitored for
futility. In particular, at each of the seven interim analyses the hypothesis that the hazard ratio is at least 1.4 will be tested at a
fixed one-side significance level of 0.005. The upper boundary will be truncated by the 1-0.005 quantile of a standard normal
distribution. For calculating the futility bound, we will assume that the standardized log-rank statistic, after the d-th event has
occurred, is normally distributed with the unit variance and mean log [1.4] (d/4)0.5. Note that among other things, this
assumption not only depends on asymptotic results but also on approximating the mean of the asymptotic distribution. For
illustrative purposes, we provide the lower and the upper boundaries (in the so called normalized z-score scale) that would be
used under the specific assumptions in the table below. We reiterate that these bounds are for illustrative purposes only as
these calculations were based on various assumptions that were rarely realized in practice.

Analysis # 1 2 3 4 5 6 7 8

Projected Time of Analysis (in months

21 27 33 39 45 51 57 63
following start of accrual)
Expected proportion of events realized at time
0.21 0.33 0.47 0.61 0.73 0.83 0.92 1.00
of analysis
Upper Boundary (for superiority) 2.576 2.576 2.576 2.308 2.109 1.974 1.877 1.803
Lower Boundary (for futility) -1.225 -0.878 -0.546 -0.265 -0.047 0.126 0.266 1.803

This rather extensive monitoring plan has, as discussed in Freidlin et al.7, a negligible impact on the
planned Type 1 and II error rates of this trial.
Toxicity Stopping Rules: If > 20% of patients randomized to the CC-5013 arm permanently discontinue
the drug due to drug-related toxicity within one year following randomization, the treatment will be
considered overly toxic and consideration will be given to terminate the study.

The study was monitored by the CALGB Data and Safety Monitoring Board (DSMB) on a semiannual basis in June and
November of each year per CALGB policy. Per protocol, the first interim analysis was to be presented after at least 20% of
the expected events had been realized. A group sequential design was employed to monitor time to progression for
superiority. The bounds were truncated by the 1- 0.005 quantile of a standard normal distribution. Time to progression was
also to be monitored for futility at each interim analysis to test the specific hypothesis that the hazard ratio is at least 1.4 as a
fixed one-sided level of 0.005. A randomized permuted block procedure using three stratification factors (β-2M elevation,
prior thalidomide induction therapy and prior lenalidomide induction therapy) was employed.

The study was opened on 12/15/2004. Accrual to the study did not begin until 04/2005. Given that the observed accrual rate
had differed considerably from that assumed in the original design, an amendment to the statistical considerations was
necessitated so as to comply with an NCI CTEP policy regarding accrual. It is noted that the study team had not carried out
any interim analyses when this amendment was drafted and consequently approved. Furthermore, the time to progression

3
distribution assumptions in the amendment and the targeted number of events (309) were identical to those of the original
design. Only the assumed distribution of the administrative censoring mechanism had been revised.

The DSMB reports, including analysis results, summaries and recommendations, were drafted and presented by the statistical
team. The clinical team members, including the study chair and data coordinators, were blinded with respect to these analyses
and results. The PI and the treating team had no access to the DSMB. The DSMB met in closed sessions and the members
were required to destroy the DSBM reports after their meetings. The first time the PI had access to the data was at the time of
study unblinding.

1. June 2009 Interim analysis: According to the protocol, the first interim analysis for efficacy was scheduled to be carried
out for presentation at the first DSMB meeting after at least 20% of the 309 expected events (progressions or deaths) had
occurred. This corresponds to at least 61 events. Out of the 375 patients randomized to the study, 74 have realized an event.
This report provides a summary of the first interim analysis for this study. There have been 18 deaths among the randomized
patients all of whom have experienced a progression event on or before time of death. The median follow-up time for TTP is
360 days (95% CI=347,369). Among the 188 patients randomized to the CC-5013 arm, 24 experienced a progression event.
This compares to 50 events among the 187 patients randomized to the placebo arm. The observed value of the standardized
log-rank statistic for testing for the association between TTP and treatment is 3.94. This exceeds the first bound of the
superiority boundary which is 2.576 and corresponds to an asymptotic P-value of less than 0.0001. Under a log-linear
proportional hazards framework the estimated hazard ratio is 0.39 (95% CI=0.24,0.63). The data suggests that there is ample
statistical evidence that CC-5013 based maintenance therapy following ASCT reduces time to progression.

2. Sep 2009 Interim analysis: According to the results from the first interim analysis presented during the June 2009 DSMB
meeting, there was strong statistical evidence to support the hypothesis that adding CC-5013 based maintenance therapy
following ASCT prolongs time to progression for multiple myeloma patients. The results from this second interim analysis
reaffirm those findings. The results from this second interim analysis are based on more mature outcome data (28% rather
than 21% of the 309 events targeted in the statistical design) and a considerably more up to date database. The results are
rather striking and per the decision rule of the study design warrant consideration for release of the study data to the study
team. There is not enough evidence to support the hypothesis that CC-5013 maintenance therapy improves overall survival in
this patient population (two-sided P-value <0.2). Although there is an apparent, albeit weak, observed trend in favor of the
CC-5013 maintenance arm with respect to survival.

Based on the strong evidence in support of the primary objective hypothesis of this study, namely that addition of CC-5013
based maintenance therapy following ASCT prolongs time to progression in the protocol population, the primary hypothesis
of the study, we recommend that the data for CALGB 100104 are released to the study team. We emphasize that this
recommendation is solely based on the observed prolongation in time to progression in support of the primary hypothesis of
the protocol as it is too early to conclude anything about overall survival. If this recommendation is approved by the DSMB,
we will conduct one additional analysis after completing the official CALGB procedures for study database finalization. If
our recommendation is not approved, we will present a third interim analysis for efficacy at the November 2009 DSMB
meeting as mandated by the protocol.

Addendum to CALGB 100104 Interim Analysis Report: This report is submitted as an addendum to the second interim
analysis of efficacy report for CALGB 100104 produced on 09/14/2009. The latter was produced at the request of the DSMB
as an update to the June 2009 DSMB report. In this addendum, we report the results from a limited sensitivity analysis.

The analysis data set for the updated analysis is comprised of 418 patients randomized on or before 07/31/2009. Among these
there are 16 cases with missing ASCT dates. More specifically, twelve out of the 208 patients randomized to the placebo arm
and four out of the 210 patients randomized to the maintenance therapy arm are missing ASCT dates. In the updated analysis,
these cases are censored at time zero which effectively excludes them from the calculation of the test statistic.

It is reasonable assume that the TTP profile of the maintenance therapy will not be any worse than that of the placebo arm. It
is noted that these 16 patients were randomized in the period starting 05-27-09 and ending 07-30-09. As these are all recent

4
cases, it is also reasonable to assume that the missingness pattern is non-informative. In other words, it is the case of the
respective institutions not having submitted the paperwork.

Under this set of assumptions, we will simulate the 16 missing cases from the placebo arm. More specifically, for each
simulation replicate, we draw 16 pairs (observed time and censoring indicator) at random with replacement from the 196
(=208 - 12) pairs on the placebo arm with complete ASCT date information.

The histogram of B=10000 bootstrap replicates is shown below. The median of these replicates is 3.89 while the first and
second quartiles are 3.77 and 4.02 respectively. The realized value of the test statistic (Z=3.85 is marked by a vertical line on
the histogram. According to the statistical design, the upper boundary for the second interim analysis is 2.57 (the -0.005
quantile of a standard normal law). It is noted that none of the bootstrap replicates are below this threshold.

Histogram of B=10000 bootstrap replicates of the standardized log-rank statistic obtained by imputing the missing cases from
the placebo arm

This has been an attempt to study the potential impact of the 16 missing cases under a set of assumptions. The purpose of this
analysis is not to approximate the null sampling distribution of the test statistic. Rather, this has been an attempt to study the
potential impact of the 16 missing cases by assuming that the missing cases are identically distributed according to the
distribution on the placebo arm. Based on the result of the actual analysis, we recommended that the study data be released to
the study team. The results from this sensitivity analysis does not suggest that we would reach a different conclusion if these
16 missing cases happen to be all distributed to the placebo arm. It is noted that this approach has several caveats. Most
notably, it neither takes into account the sequential nature of the statistical design nor does take into account stratification
factors.

3. October 2009 Interim analysis: The first interim analysis for efficacy was presented at the June 2009 CALGB DSMB
meeting as mandated by the protocol. What is presented in the current report are the results from the second interim analysis
for efficacy. This analysis is being presented at the request by the DSMB as an update to the first interim analysis. The results
presented are based on the data from all 418 patients randomized on or before 07/31/2009. The median follow-up time for

5
TTP is currently 372 days (95% CI=363,410). Among the 210 patients randomized to the CC-5013 arm, 29 have experienced
an event (disease progression or death). This compares to 58 events among the 208 patients randomized to the placebo arm.
The observed value of the standardized log-rank statistic for testing for the association between TTP and treatment is 3.85.
This exceeds the second bound of the superiority boundary which is 2.576 and corresponds to an asymptotic one-sided P-
value of less than 0.0001. Under a proportional hazards framework the estimated hazard ratio is 0.42 (95% CI=0.27,0.66).
The data suggests that there is ample statistical evidence that CC-5013 based maintenance therapy following ASCT prolongs
time to progression. There is not enough statistical evidence to conclude that the maintenance therapy improves overall
survival (two-sided P-value <0.2). It is noted that these estimates, test statistics and P-values have not been adjusted to take
into account that one interim analysis has been carried out. The upper boundary is adjusted for multiple testing. The adjusted
P-value is in the interval (0.0001,0.005).

Eligibility Criteria:

Active Multiple Myeloma: Patients must have active multiple myeloma requiring treatment (Durie-Salmon stage > 1) and
have stable disease or be responsive to at least two months of any induction therapy. Patients with smoldering myeloma were
not eligible unless the disease had progressed to > 1 stage 1.

Prior Therapy Patients must not have received more than 12 months of any prior therapy, including lenalidomide and
thalidomide. Patients must have been within 12 months of initiation of induction therapy. Patients must not have had prior
progression after initial therapy. No more than two regimens were allowed, excluding dexamethasone alone. Patients must
not have undergone prior peripheral blood, bone marrow, or solid organ transplant.

Peripheral Blood Stem Cell Collection: Patients must have peripheral blood stem cell collection of > 2 x 106 CD34+ cells/kg
(patient body weight) and preferable 5 x 106 cells/kg. Stem cells may be collected at any time prior to transplant and may
occur before or after registration.

Age Requirement: Patients must be > 18 and < 70 years of age.

ECOG Performance Status: Patients must have ECOG performance status of 0-1.

Additional Requirements: Patients must have DLCO >50% predicted with no symptomatic pulmonary disease, must have
LVEF > 40% by MUGA or echocardiogram, must not have uncontrolled diabetes mellitus and must not have an active
serious infection. Patients must not be HIV, HBSag or Hep C positive. Patients must be non-pregnant and non-nursing.

Initial Required Laboratory Values: ANC > 1000/µL, platelets > 100,000/µL, creatinine clearance > 40 cc/min, creatinine < 2
mg/dL, total bilirubin < 2 mg/dL, AST/alkaline phosphatase < 3 x upper limits of normal, and negative urine-HCG or serum
HCG if patient of childbearing potential. Creatinine clearance to be calculated by method of Cockcroft-Gault or after 24-hour
urine collection.

Myeloma Monitoring during Maintenance Therapy: Myeloma laboratory studies were to be done every three months for
four years then every six months thereafter. Skeletal surveys were to be done annually for five years. Bone marrow
aspirate/biopsy was to be done at 3 and 12 months after Day 0 of ASCT and then yearly until five years post-ASCT and at
time of progression.

Dose Modifications: Patients were scheduled to be re-staged between day 90-100 post-autologous hematopoietic cell
transplant (AHCT) and those with SD or better were scheduled to start therapy between day 100 to 120 post-AHCT. All
patients started on 2 pills (10 mg of lenalidomide) daily. Hematologic dose modifications were as follows:

Months 1-3: If on 2 capsules per day, the absolute neutrophil count (ANC) is <500/μL or the platelet count is <30,000/μL,
then the study drug will be held for 8 weeks. Study drug may be re-instituted at 1 capsule per day if ANC is ≥500/μL or the
platelet count is ≥30,000/μL. If, however, after an 8 week treatment delay, the ANC remains <500/μL or the platelet count
<30,000/μL, the patient will be removed from protocol therapy. If on 1 capsule per day, the ANC is <500/μL or the platelet

6
count is <30,000/μL, then the study drug will be held for 8 weeks. If ANC ≥500/μL or the platelet count is ≥30,000/μL, then
study drug may be re-instituted at 1 capsule per day for 21 of 28 days. If, however, after an 8 week treatment delay, the ANC
remains <500/μL or the platelet count <30,000/μL, the patient will be removed from protocol therapy. If on 1 capsule per day
for 21 of 28 days, the ANC is <500/μL or the platelet count is <30,000/μL, then the patient will be removed from protocol
therapy.

Beyond Month 3: If on 3 capsules per day, the ANC is <500/μL or the platelet count is <30,000/μL, then the study drug will
be held for 8 weeks. Study drug may be re-instituted at 2 capsules per day if ANC is ≥500/μL or platelet count is ≥30,000/μL.
If, however, after an 8 week treatment delay, the ANC remains <500/μL or platelet count <30,000/μL, the patient will be
removed from protocol therapy. If on 2 capsules per day, the ANC is <500/μL or the platelet count is <30,000/μL, then the
study drug will be held for 8 weeks. If ANC is ≥500/μL or the platelet count is ≥30,000/μL, then study drug may be re-
instituted at 1 capsule per day. If, however, after an 8 week treatment delay, the ANC remains <500/μL or the platelet count
<30,000/μL, the patient will be removed from protocol therapy. If on 1 capsule per day, the ANC is <500/μL or the platelet
count is <30,000/μL, then the study drug will be held for 8 weeks. If ANC ≥500/μL or the platelet count is ≥30,000/μL, then
study drug may be re-instituted at 1 capsule per day for 21 of 28 days. If, however, after an 8 week treatment delay, the ANC
remains <500/μL or the platelet count <30,000/μL, the patient will be removed from protocol therapy. If on 1 capsule per day
for 21 of 28 days, the ANC is <500/μL or the platelet count is <30,000/μL, then the patient will be removed from protocol
therapy.

Dose Escalation Beyond Month 3: If a dose reduction has occurred and ANC ≥1000/μL and platelet count is ≥75,000/μL, the
study drug dose may be re-escalated by one level (i.e., one capsule every day to two capsules every day, etc.). Hematologic
parameters must remain at these threshold values for one month before another dose escalation may occur. Maximum study
drug dose will be 3 capsules per day. If for any reason, a patient is not able to be dose escalated, dose escalation should be
attempted by the time of the next re-staging. If at next restaging, the patient has not recurred or progressed, and the patient is
not able to be dose escalated, patient may continue on treatment at current dose level. If for any reason the drug is held for a
non-grade 3 hematologic toxicity, the drug will be held until the toxicity resolves and the drug started at one dose level lower.
The drug should be re-escalated to the original dose within 4 weeks. The drug should be escalated as per the criteria listed
above.

References

1. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457-81.
2. Cox DR, Oakes D. Analysis of Survival Data; 1984.
3. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time data, 2nd edition; 2002.
4. Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988;
16(3): 1141-54.
5. Fisher RA. The logic of inductive inference. J R Stat Soc 1935; 98: 39.
6. Emerson SS, Fleming TR. Symmetric group sequential test designs. Biometrics 1989; 45(3): 905-23.
7. Freidlin B, Korn EL, George SL. Data monitoring committees and interim monitoring guidelines. Control Clin
Trials 1999; 20(5): 395-407.

7
Figure S1. CONSORT flow diagram of patient disposition since time of enrollment to time of intent-to treat analysis.

8
Figure S2: Kaplan-Meier curves for TTP/overall survival for lenalidomide and placebo groups where lenalidomide
contains placebo patients who crossed over within 6 months (n=19) or within 12 months (n=46). A) TTP, crossover with
6 months; B) overall survival, crossover within 6 months; C) TTP, crossover within 12 months; D) overall survival, crossover
within 12 months.

9
Figure S3: Kaplan-Meier curves for survival after progression where the lenalidomide group includes placebo
patients who crossed over within 6 (A) or 12 (B) months.

10
Figure S4: Kaplan-Meier curves for (A) TTP or (B) overall survival for lenalidomide and placebo patients
who received lenalidomide (Prior Len) induction versus no lenalidomide (No Prior Len) induction.

11
Figure S5: Kaplan-Meier curves for (A) TTP and (B) overall survival for lenalidomide and placebo patients in
complete response (CR) or not in CR at randomisation

12
 Solid Len
Solid PBO
Heme Len
Heme PBO

0 1 2 3 4 5 6 7 8 9 10

Time from Randomization to SPM (Years)

Figure S6: Swimmers plot showing time from randomisation to invasive (hematological and solid tumor) SPM
diagnosis. Abbreviations: Len, lenalidomide; PBO, placebo.

13
Figure S7: Box plot representing the length of time from autologous stem cell transplant (ASCT) to SPM of each
invasive SPM type (haematological and solid tumour) that occurred prior to disease progression among the patients
treated with lenalidomide or placebo. The median line indicates the median time from ASCT to SPM. The time to SPM
was not different between the lenalidomide and placebo groups.

14
Figure S8: Cumulative incidence risk (CIR) of haematological (heme) vs solid tumour (solid) second primary
malignancies (SPMs) comparing patients in the lenalidomide arm to those patients in the placebo arm. A) Cumulative
incidence of haematological vs solid tumour SPMs. B) Cumulative incidence of death from haematological vs solid tumour
SPMs.

15
Figure S9: Cumulative incidence risk (CIR) of PD, death, and second primary malignancies (SPMs) comparing
patients in the lenalidomide arm to those patients in the placebo arm who did not cross over to receive lenalidomide.
A) The CIR of PD or death from any cause is higher with placebo compared to lenalidomide (p<0·0001). The CIR of
developing a SPM is higher with lenalidomide compared with placebo (p=0·0002). B) The CIR of death from any cause is
higher with placebo compared with lenalidomide (p<0·0001). C) The CIR of death from myeloma is higher with placebo
than with lenalidomide (p<0·0001) while the CIR of death from SPM is higher with lenalidomide than placebo (p=0·033).

16
Figure S10: Cumulative incidence risk (CIR) of progressive disease, death, and second primary malignancies (SPMs)
comparing patients in the lenalidomide arm to those patients in the placebo arm who crossed over to receive
lenalidomide. A) There is no difference in the CIR of progressive disease or death from any cause between the lenalidomide
and crossover groups (p=0·85). There is no difference in the CIR of developing a SPM between the lenalidomide and
crossover groups (p=0·83). B) There is no difference in the CIR of death from any cause between the lenalidomide and
crossover groups (p=0·54). C) There are no differences in the CIR of death from myeloma (p=0·56) or death from SPM
(p=0·30) between the lenalidomide and crossover groups.

17
Figure S11. Kaplan-Meier curve for event-free survival (time to progression, death, or SPM).

18
Table S1. Follow-up times from time of ASCT

Estimate (mos) 95% Confidence Interval

Median 91·0 90·0 - 93·7

Q1 83·7 82·1 - 84·7

Q3 103·3 99·9 - 106·3

19
Table S2. Characteristics of the Patients*

Characteristic Lenalidomide Placebo Total

(n=231) (n=229) (n=460)
Age – yr
Median 59 58 59
Range 29-71 40-71 29-71
Male sex – no. 121 129 250
β2-microglobulin at registration – no. (%)
>2·5 mg/liter 50 (22) 55 (24) 105 (23)
<2·5 mg/liter 170 (74) 163 (71) 333 (72)
Data missing 11 (5) 11 (5) 22 (5)
Durie-Salmon stage at registration – no. (%)
I 35 (15) 28 (12) 63 (14)
II 71 (31) 59 (26) 130 (28)
III 112 (48) 129 (56) 241 (52)
Data missing 13 (6) 13 (6) 26 (6)
M component – no. (%)
Serum
IgG kappa 70 (30) 76 (33) 146 (32)
IgG lambda 43 (19) 31 (14) 74 (16)
IgA kappa 21 (9) 20 (9) 41 (9)
IgA lambda 13 (6) 13 (6) 26 (6)
IgM kappa 2 (1) 1 (<1) 3 (1)
IgM lambda 0 1 (<1) 1(<1)
Urine
Kappa light chains only 13 (6) 12 (5) 24 (5)
Lambda light chains only 4 (2) 10 (4) 14 (3)
Data missing 35 (15) 41 (18) 76 (17)
Nonsecretory myeloma – no. (%) 30 (13) 24 (10) 54 (12)
Serum calcium at registration – mg/dL
Median 9·1 9·1 9·1
Range 7·2-12·8 7·3-10·8 7·2-12·8
Serum albumin at registration – g/dL
Median 4·0 3·9 4·0
Range 1·4-4.9 2.9-5.0 1·4-5·0
Serum creatinine at registration –mg/dL
Median 0·9 0·9 0·9
Range 0·4-1·9 0·5-2·2 0·4-2·2
ISS stage at registration – no. (%)
I 177 (77) 170 (74) 347 (75)
II 11 (5) 16 (7) 26 (6)
III 4 (2) 3 (1) 7 (2)
Data missing 39 (17) 40 (17) 79 (17)
Induction regimen – no.
Any use of bortezomib 98 91 189 (41)
Any use of lenalidomide 79 81 160 (35)
Any use of thalidomide 102 103 205 (45)
Bortezomib-lenalidomide 20 21 41 (9)
Bortezomib-thalidomide 33 27 60 (13)
Bortezomib without lenalidomide or thalidomide 43 40 83 (18)
Bortezomib with glucocorticoids, without lenalidomide or thalidomide 40 32 72 (16)
Bortezomib with lenalidomide and thalidomide 2 3 5 (1)
Lenalidomide without bortezomib 57 57 114 (25)
Thalidomide without bortezomib 67 72 139 (30)
Lenalidomide-glucocorticoids without bortezomib 56 56 112 (24)
Thalidomide-glucocorticoids without bortezomib 65 72 137 (30)
Other induction regimen without bortezomib, lenalidomide, or thalidomide 15 13 28 (6)
Other induction regimen not determined
0 1 1 (<1)
Response to autologous HSCT at day 100—no. (%)
Complete response 67 (29) 79 (34) 146 (32)
Partial response 115 (50) 109 (48) 224 (49)
Marginal response 11 (5) 5 (2) 16 (3)
Stable disease 38 (16) 32 (14) 70 (15)
Progressive disease 0 3 (1) 3 (1)
Data missing 0 1 (<1) 1 (<1)
Mean time from autologous HSCT to randomization - mo 3.3 3.3

20
*From New England Journal of Medicine, Lenalidomide after stem-cell transplantation for multiple myeloma, PL McCarthy,
K Owzar, CC Hofmeister, DD Hurd, H Hassoun, PG Richardson, S Giralt, EA Stadtmauer, DJ Weisdorf, R Vij, JS Moreb,
NS Callander, K Van Vesien, T Gentile, L Isola, RT Maziarz, DA Gabriel, A Bashey, H Landau, T Martin, MH Qazilbash, D
Levitan, B McClune, R Schlossman, V Hars, J Postiglione, C Jiang, E Bennett, S Barry, L Bressler, M Kelly, M Seiler, C
Rosenbaum, P Hari, MC Pasquini, MM Horowitz, TC Shea, SM Devine, KC Anderson, C Linker, 366(19): 1770-81.
Copyright © (2012) Massachusetts Medical Society. Reprinted with permission.

21
Table S3. Adverse events associated with treatment discontinuation as determined by chart review1

AE Lenalidomide (n) Placebo (n) Crossover (n)

Cytopenias 9 1
Neutropenia 2 1
Thrombocytopenia 4 1
Anemia 1
Hemolytic anemia 1
Infection 7
Venous thrombosis 1 1
Pericarditis 1
Supraventricular and nodal 1
arrhythmia
Coronary artery disease 1
Diarrhea 4
Renal insufficiency 1 1
Creatinine phosphokinase 1
Rash/desquamation 6 2
Pain 2
Fatigue 1
Hypothyroid 1
Stroke 1 1 1
Neuropathy 1 1 1
Seizures 1
Dizziness 1
Syncope 1
Encephalopathy 1
Ataxia 1* 1
Tremor 1
Total AEs** 45 5 14
Total number of patients 42 5 13
1
This study’s off-treatment form did not require the sites to specify which adverse event led to treatment discontinuation,
although in some cases this information was provided as a comment on the form. For those patients for whom such
comments were not provided, chart review was performed by S.A.H. to adjudicate the adverse event/s associated with
treatment discontinuation. This review involved the utilization of study adverse event reporting forms from the time period of
treatment discontinuation, AdEERs reports, or primary source documents (i.e., clinic notes).
*This patient was subsequently suspected to have amyotrophic lateral sclerosis
**A few patients had two adverse events which contributed to study discontinuation.

22
Table S4. TTP/PFS analyses with different censoring rules

Censoring Number of Number of Number of Number of Median TTP Median TTP Hazard ratio
system evaluable evaluable events (n), events (n), (mos), (mos), placebo (95% CI)
patients (n), patients (n), lenalidomide placebo lenalidomide
lenalidomide placebo
Alliance (EMA)1 231 229 146 176 57·3 28·9 0·57 (0·46 to
0·71)
FDA2 231 228 115 114 62·7 19·2 0·38 (0·29 to
0·50)
1
Using Alliance (EMA) censoring rules: The transplant date is used as the reference date for determining the number of
evaluable patients. Events include disease progression (as defined by the IMWG criteria) or death. The following events are
censored: 1) no baseline assessments and alive after two scheduled assessments and 2) no progression. For new anti-myeloma
therapy or non-protocol therapy started prior to progression (including those patients on the placebo arm who crossed over to
receive lenalidomide on study), the event occurs at date of PD or death. For death or progression immediately after an
extended lost-to-follow-up time (two or more missed scheduled assessments defined by at least 252 days), the event occurs at
date of PD or death.
Reference: (2013). Appendix 1 to the Guideline on the Evaluation of Anticancer Medicinal Products in Man: Methodological
consideration for using progression-free survival (PFS) or disease-free survival (DFS) in confirmatory trials. EMA/CHMP.
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2013/01/WC500137126.pdf
2
Using FDA censoring rules: The randomisation date is used as the reference date for determining the number of evaluable
patients. Events include disease progression (as defined by the IMWG criteria) or death. The following events are censored:
1) no baseline assessments and alive after two scheduled assessments; 2) no progression; 3) new anti-myeloma therapy or
non-protocol therapy started prior to progression, including those patients on the placebo arm who crossed over to receive
lenalidomide on study; 4) death or progression immediately after an extended lost-to-follow-up time (two or more missed
scheduled assessments defined by at least 252 days).
Reference: (May 2007). Guidance for Industry Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics. US
FDA/CDER/CBER. https://www.fda.gov/downloads/Drugs/.../Guidances/ucm071590.pdf

23
Table S5. Summary of first salvage therapy

Len (n) Plac (n) Crossover (n)

IMiD-baseda 32 54 13
PI-basedb 40 16 16
IMiD+PI-basedc 8 15 6
Investigational agentsd 8 15 3
Othere 13 4 4
Unknownf 19 14 0
Died before salvage 5 1 6
Total number of patients with 125 119 48
disease progression
a
Includes Len, Len/Dex, Thal, Thal/Dex, Len/Thal/Dex, Len/Cy/Dex, Pom
b
Includes Bor, Bor/Dex, Bor/Dox/Dex, Vel/Cy/Dex, Car, Car/Dex
c
Includes Len/Bor/Dex, Bor/Thal/Dex, Car/Len/Dex, Pom/Bor/Dex
d
Includes BB10901, CNTO238, Bor/tanespimycin, anti-Kir antibody, curcumin, NPI-0052, EMND2076, marizomib,
siltuximab, Len/temsirolimus, Vel/anti-CS-1 antibody, Len/Dex/bevacizumab, unspecified carfilzomib trial, Bor/azacitidine,
Len/everolimus, Vel/vorinostat, Len/anti-Kir antibody, AR-42, Bor/Dex/afuresertib, daratumumab trial
e
Includes Cy/Dex, hyperCVAD, VD-PACE, Dox, steroids, radiation, ASCT, allogeneic transplant,
daratumumab/dexamethasone
f
Includes patients for whom salvage regimens were not reported, or based on timing of reported salvage regimens, they could
not be confirmed as first salvage.
Abbreviations: Bor, bortezomib; Car, carfilzomib; Cy, cyclophosphamide; Dex, dexamethasone; Dox; liposomal
doxorubicin; hyperCVAD, hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone; IMiD,
immunomodulatory; Len, lenalidomide; Pom, pomalidomide, Thal, thalidomide; VD-PACE, bortezomib, dexamethasone,
cisplatin, doxorubicin, cyclophosphamide, etoposide.

24
Table S6. Complete listing of initial salvage therapies
Lenalidomide (n) Placebo (n) Crossover (n)
Len 14 26 3
Len/Dex 12 25 9
Len/Bor (+/- Dex) 3 12 5
Len/Car/Dex 1 2 0
Len/Cy/Dex 1 1 1
Len/Thal/Dex 1 0 0
Thal 1 0 0
Thal/Dex 3 2 0
Thal/Bor/Dex 2 1 0
Pom 0 1 0
Pom/Bor/Dex 2 0 1
Bor 9 7 5
Bor/Dex 16 3 9
Bor/Cy (+/- Dex) 8 1 2
Bor/Dox (+/- Dex) 4 2 0
Bor/Mel/Pred 0 1 0
Bor/Cy/Dox 0 1 0
Car 1 0 0
Car/Dex 2 0 0
Len/Dex/perifosine 0 1 0
Len/temsirolimus 0 3 0
Len/everolimus 0 1 0
Len/Dex/bevacizumab 0 1 0
Len/anti-Kir antibody 0 1 0
Bor/tanespimycin 1 0 0
Bor/azacitidine 0 1 0
Bor/Dex/afuresertib 0 0 1
Bor/vorinostat 0 1 0
Bor/CS-1 0 1 0
Bor/Dex/perifosine 1 0 0
Anti-Kir antibody 1 3 0
Dara (trial) 1 0 0
AR-42 0 0 2
BB10901 1 0 0
Siltuximab 1 0 0
Marizomib 0 1 0
Carfilzomib-based trial 0 1 0
EMND2076 1 0 0
Curcumin 1 0 0
Radiation 2 4 1
Radiation + steroid 1 0 0
Steroid 2 0 1
VD-PACE 2 0 0
HyperCVAD 0 0 1
Cy/Dex 1 0 0
Doxil regimen 1 0 0
Dara/Dex 1 0 0
ASCT 3 0 0
AlloSCT 0 0 1
Abbreviations: Bor, bortezomib; Car, carfilzomib; Cy, cyclophosphamide; Dara, daratumumab; Dex, dexamethasone; Dox;
liposomal doxorubicin; hyperCVAD, hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone; IMiD,
immunomodulatory; Len, lenalidomide; Pom, pomalidomide, Thal, thalidomide

25
Table S7: Grade > 3 adverse events at least possibly related to treatment1

Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

Hematologic Adverse Events
Blood/Bone Marrow
Blood/Bone Marrow - Other (Specify LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
CD4 count LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Hemoglobin LEN 9 ( 4%) 2 ( 1%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Hemolysis (e.g. immune hemolytic LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Leukocytes (total WBC) LEN 28 ( 12%) 3 ( 1%) 0 ( 0%) 231
PBO 1 ( 1%) 1 ( 1%) 0 ( 0%) 143
CRS 8 ( 9%) 1 ( 1%) 0 ( 0%) 86
Lymphopenia LEN 20 ( 9%) 1 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 1 ( 1%) 0 ( 0%) 143
CRS 5 ( 6%) 0 ( 0%) 0 ( 0%) 86
Neutrophils/granulocytes (ANC/AGC) LEN 82 ( 35%) 34 ( 15%) 0 ( 0%) 231
PBO 7 ( 5%) 4 ( 3%) 0 ( 0%) 143
CRS 22 ( 26%) 4 ( 5%) 0 ( 0%) 86
Platelets LEN 23 ( 10%) 11 ( 5%) 0 ( 0%) 231
PBO 0 ( 0%) 7 ( 5%) 0 ( 0%) 143
CRS 2 ( 2%) 2 ( 2%) 0 ( 0%) 86

SUMMARY
Maximum Hematologic AE LEN 80 ( 35%) 40 ( 17%) 0 ( 0%) 231
PBO 6 ( 4%) 8 ( 6%) 0 ( 0%) 143
CRS 20 ( 23%) 6 ( 7%) 0 ( 0%) 86
Non-Hematologic Adverse Events
Allergy/Immunology
Allergy/Immunology - Other (Specify LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Cardiac Arrhythmia
Conduction abnormality/atrioventricular heart LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 1 ( 1%) 143

26
 Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Cardiac General
Hypotension LEN 0 ( 0%) 1 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Restrictive cardiomyopathy LEN 0 ( 0%) 1 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Coagulation
INR (International Normalized Ratio) LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Constitutional Symptoms
Fatigue (asthenia, lethargy, malaise) LEN 14 ( 6%) 0 ( 0%) 0 ( 0%) 231
PBO 4 ( 3%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Fever (in the absence of neutropenia LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Weight loss LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Dermatology/Skin
Pruritus/itching LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Rash/desquamation LEN 9 ( 4%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Gastrointestinal
Anorexia LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Dental: periodontal disease LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Diarrhea LEN 12 ( 5%) 0 ( 0%) 0 ( 0%) 231
PBO 2 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 2 ( 2%) 0 ( 0%) 0 ( 0%) 86
Distension/bloating, abdominal LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231

27
 Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Fistula GI LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Mucositis/stomatitis (functional/symptomatic) LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Nausea LEN 4 ( 2%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Obstruction GI LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Vomiting LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Hepatobiliary/Pancreas
Cholecystitis LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Infection
Febrile neutropenia (fever of unknown origin LEN 14 ( 6%) 1 ( 0%) 0 ( 0%) 231
PBO 2 ( 1%) 1 ( 1%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Infection (documented clinically or LEN 13 ( 6%) 2 ( 1%) 0 ( 0%) 231
microbiologically) with grade 3 or 4 neutropenia
(ANC<1.0 x 10e9/L)
PBO 3 ( 2%) 0 ( 0%) 0 ( 0%) 143
CRS 3 ( 3%) 0 ( 0%) 0 ( 0%) 86
Infection - Other (Specify __) LEN 3 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Infection with normal ANC or Grade 1 or 2 LEN 13 ( 6%) 0 ( 0%) 1 ( 0%) 231
neutropenia
PBO 3 ( 2%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Infection with unknown ANC LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Infection without neutropenia LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231

28
 Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Opportunistic infection associated with >=Grade LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
2 lymphopenia
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Metabolic/Laboratory
Alanine aminotransferase LEN 4 ( 2%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Aspartate transaminase LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Alkaline phosphatase LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Bilirubin (hyperbilirubinemia) LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
CPK (creatine phosphokinase) LEN 0 ( 0%) 1 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Creatinine LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 1 ( 1%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
GGT (gamma-Glutamyl transpeptidase) LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Glucose serum-high (hyperglycemia) LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 2 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Phosphate serum-low (hypophosphatemia) LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Potassium serum-high (hyperkalemia) LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Potassium serum-low (hypokalemia) LEN 6 ( 3%) 0 ( 0%) 0 ( 0%) 231
PBO 2 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Sodium serum-low (hyponatremia) LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231

29
 Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

PBO 0 ( 0%) 1 ( 1%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Uric acid serum-high (hyperuricemia) LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Neurology
CNS cerebrovascular ischemia LEN 0 ( 0%) 1 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Mood alteration LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Neurology - Other LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Neuropathy: motor LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Neuropathy: sensory LEN 4 ( 2%) 0 ( 0%) 0 ( 0%) 231
PBO 2 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Syncope (fainting) LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Pain
Pain LEN 6 ( 3%) 0 ( 0%) 0 ( 0%) 231
PBO 6 ( 4%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Pulmonary/Upper Respiratory
Cough LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Dyspnea (shortness of breath) LEN 1 ( 0%) 2 ( 1%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Hypoxia LEN 3 ( 1%) 1 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Pneumonitis/pulmonary infiltrates LEN 6 ( 3%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 1 ( 1%) 0 ( 0%) 143

30
 Grade of Adverse Event

Arm 3-Severe 4-LifeThr 5-Lethal Total

n (%) n (%) n (%)

CRS 1 ( 1%) 0 ( 0%) 0 ( 0%) 86
Renal/Genitourinary
Renal failure LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Renal/Genitourinary - Other (Specify LEN 1 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Vascular
Thrombosis/embolism (vascular access-related) LEN 0 ( 0%) 0 ( 0%) 0 ( 0%) 231
PBO 1 ( 1%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Thrombosis/thrombus/embolism LEN 2 ( 1%) 0 ( 0%) 0 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
Vascular - Other LEN 0 ( 0%) 0 ( 0%) 1 ( 0%) 231
PBO 0 ( 0%) 0 ( 0%) 0 ( 0%) 143
CRS 0 ( 0%) 0 ( 0%) 0 ( 0%) 86
SUMMARY
Maximum Non-Hematologic AE LEN 80 ( 35%) 7 ( 3%) 2 ( 1%) 231
PBO 22 ( 15%) 3 ( 2%) 1 ( 1%) 143
CRS 12 ( 14%) 0 ( 0%) 0 ( 0%) 86
SUMMARY
Maximum Overall AE LEN 98 ( 42%) 46 ( 20%) 2 ( 1%) 231
PBO 26 ( 18%) 8 ( 6%) 1 ( 1%) 143
CRS 23 ( 27%) 6 ( 7%) 0 ( 0%) 86

1
S.A.H. and P.L.M. reviewed all grade 5 AEs and reviewed the AEs reported by each center to the Alliance Data Center with
the data management personnel.

31
Table S8. Hematological SPMs details

SPM Arm Myeloma Sex Age at Cause of Time on Time from Time from SPM Cyto Other
induction SPM death Len (days) Rand to heme SPM SPMs
therapy (yrs) SPM (days) to death
(days)
AML** Len PAD F 55.9 Alive 957 987 Complex; 5-, 7-
AML Len VAD M 67.7 SPM 210 774 131 45X(Y-)
AML Len VAD M 43.2 SPM 846 946 85 Complex
AML Len RD M 74.2 SPM 1203 2475 411 Complex, 5q-,
7q-
AML** Len TD M 67.2 Alive 1387 1393 MLL rearrange-
ment
AML Len TD M 70.8 SPM 1679 1706 424 Complex, 5q-, SCC
7q-
MDS/AML Len TD M 58.4 SPM 695 743 762 Del(7)
MDS/AML Len RD M 74.3 SPM 2592 2666 38 NA SCC
MDS Len VD M 62.2 Alive 1440 1422 Complex, 5- colon
MDS Len RD F 60.1 Alive 2103 2093 Ring(7)
MDS Len RD M 52.9 SPM 315 376 344 Complex colon
MDS** CO VTD M 63.6 SPM* 824 999 NA melanoma
B ALL** CO TD M 64.5 SPM 1622 1931 321 20q-
B ALL CO RD M 66.2 Alive 1641 1756 NA
B ALL** Len TD M 50.4 SPM 776 780 215 Normal
B ALL Len VD M 62.6 Alive 2915 2955 +8, +10, +21,
del20, gain
RUNX1
B ALL** Len TD M 61.4 Alive 1322 1470 NA
B ALL** Len TD/Dox F 68.3 Alive 1537 1873 NA
B ALL Len VAD/Thal F 61.4 Alive 3399 3427 Hyperdiploid
with 20q-
HL1 Len VAD F 60.5 SPM 1257 1289 24
WM Len RD M 71.2 Alive 23 2547 NA
Abbreviations: CO (crossover); Cyto (cytogenetics); Dox (liposomal doxorubicin); HL (Hodgkin lymphoma); Len
(lenalidomide); NA (not available); PAD (bortezomib, liposomal doxorubicin, dexamethasone); Rand (randomisation); RD
(lenalidomide, dexamethasone); SCC (squamous cell carcinoma); TD (thalidomide, dexamethasone); VAD (vincristine,
liposomal doxorubicin, dexamethasone); VD (bortezomib, dexamethasone); VTD (bortezomib, thalidomide,
dexamethasone); WM (Waldenstrom macroglobulinemia).
1
With EBV positivity and hemophagocytic lymphohistiocytosis
*Cause of death was metastatic melanoma
**Underwent allogeneic stem cell transplant for the SPM

32
Table S9. Solid tumour SPMs details

Malignancy Arm Myeloma Sex Age at Cause of Time on Reason for Time from R to Time from
induction SPM (yrs) death Len Len d/c SPM (days) ST SPM to
therapy (days) death (days)
Breast, metastatic Len RD F 64.1 SPM 895 SPM 925 2171
Breast Len VTD M 61.9 MM 95 Cytopenias 94
Breast + Len VTD F 59 Alive 1280 SPM 1265 (breast),
endometrial 1306 (endometrial)
Breast PBO TD F 72.5 Alive 3097
Colon Len TD M 61.7 Alive 1298 SPM chemo 997
Colon* Len VD M 64.6 Alive 1440 MDS diagnosis 2324
Colon* Len RD M 53 septic 315 Cytopenias 418
shock
Endometrial Len TD F 61.4 Alive 2112 SPM recurrence 750
Endometrial CO VTD F 71 Alive 110 Patient choice 974
Endometrial + PBO TD F 71 MM 458
ovarian
Glioblastoma Len TD F 68.8 SPM 104 SPM 112 374
multiforme
Lung carcinoid PBO TD F 69.3 stroke 195
Melanoma Len PAD M 60.9 Alive 1247 PD 507
Melanoma** CO VTD M 64.2 SPM 824 MDS diagnosis 1196 1271
Melanoma PBO RD F 69 Alive 725
Melanoma CO VTD M 63.3 Alive 92 Fatigue 643
Papillary thyroid Len VD M 57.4 Alive 1923 PD 405
Prostate Len RD M 67.8 Alive 280 Neutropenia 564
Prostate Len RD M 66 Alive 69 Rash 240
Renal cell CO TD F 46.4 Alive 819 PD 1256
Salivary gland Len TD M 52.7 Alive >2000 1238
SCC, invasive CO VD M 63.2 Alive 1259 Patient choice 1179
Abbreviations: CO (crossover); d/c (discontinuation); Len (lenalidomide); MDS (myelodysplastic syndrome); NA (not
available); PAD (bortezomib, liposomal doxorubicin, dexamethasone); PD (myeloma disease progression); PBO (placebo); R
(randomisation); RD (lenalidomide, dexamethasone); SCC (squamous cell carcinoma); ST (solid tumor); VD (bortezomib,
dexamethasone), VTD (bortezomib, thalidomide, dexamethasone).
*This patient also had MDS which was diagnosed prior to the colon cancer.
**This patient also had MDS which was diagnosed prior the melanoma. Cause of death was metastatic melanoma.

33
Table S10. Second primary malignancies that occurred after myeloma progression
Malignancy Arm Cause of death Time from Time from
randomization to PD randomization to SPM
(days) (days)
Hodgkin disease Len Alive 2011 2066
Renal cell carcinoma* Len Alive 1607 2020
Lung adenocarcinoma Len Alive 1016 2739
Common bile duct Len SPM 1224 1563
adenocarcinoma
Prostate cancer Len Alive 1363 2412
BCC Len Alive 256 1509
SCC Len Alive 1060 2600
Pre-B cell ALL CO SPM 962 1805
MDS CO MM 413 1006
SCC CO Alive 1363 2259
MDS PBO MM 939 1905
MDS PBO SPM 878 2149
Histiocytic sarcoma PBO SPM 264 908
Leiomyosarcoma PBO SPM 448 1005
Breast PBO Alive 624 1244
BCC PBO MM 936 1853
Abbreviations: BCC (basal cell carcinoma); CO (placebo crossover to lenalidomide); MM (multiple myeloma); MDS
(myelodysplastic syndrome); Len, lenalidomide; PBO (placebo); SCC (squamous cell carcinoma).
*This patient also had BCC and SCC

34
Table S11. Second primary malignancies in patients who were enrolled but never randomized.

Hematological (n) Solid tumor (n) Noninvasive skin cancer (n)

MDS (1) Melanoma (2) BCC (1)
Adenocarcinoma of unknown primary (1) SCC (1)
Bladder (1)
Breast (1)
Cholangiocarcinoma (1)
Intraductal papillary mucinous cystic neoplasm of the pancreas (1)

35
Table S12. Study Investigators and Investigational Sites

Investigational Site Investigator(s) Number of patients recruited

Ohio State University Comprehensive Cancer Center LAPS Craig Hofmeister 50
Memorial Sloan Kettering Cancer Center LAPS Hani Hassoun 41
UCSF Medical Center-Mount Zion Thomas Martin 39
University of Texas MD Anderson Cancer Center LAPS Muzaffar Qazilbash 36
Wake Forest University Health Sciences David Hurd 35
University of Minnesota/Masonic Cancer Center Daniel Weisdorf 33
Dana Farber/Harvard Cancer Center LAPS Paul Richardson 32
Washington University School of Medicine LAPS Ravi Vij 25
University of Pennsylvania/Abramsom Cancer Center Edward Stadtmauer 24
University of Florida, Gainesville Jan Moreb 22
University of Wisconsin Carbone Cancer Center LAPS Natalie Callander 19
State University of New York Upstate Medical University Teresa Gentile 15
University of Chicago Comprehensive Cancer Center LAPS Koen van Besien 15
Mount Sinai Hospital Luis Isola 14
Roswell Park Cancer Institute LAPS Philip McCarthy 14
Oregon Health and Science University Richard Maziarz 13
UNC Lineberger Comprehensive Cancer Center LAPS Thomas Shea 12
Blood and Marrow Transplant Group of Georgia Asad Bashey 11
Columbia-Presby-St Luke's Hospital Jeffrey Matous 10
Colorado Cancer Research Program NCORP
Rutgers Cancer Institute of New Jersey Mecide Gharibo 9
Northwell Health NCORP Ruthee-Lu Bayer 8
University of Pittsburgh Cancer Institute LAPS Markus Mapara 8
City of Hope Comprehensive Cancer Center F Sahebi 8
Western Pennsylvania Hospital John Lister 8
Vanderbilt Adetola Kassim 6
University Hospitals of Cleveland Tamila Kindwall-Keller 5
Medical College of Georgia Arnand Jillella 5
University of Nebraska Medical Center Julie Vose 5
Marshfield Clinic-- Wisconsin NCI Community Oncology Research Program Ali Bseiso 5
University of Vermont College of Medicine Barbara Grant 4
University of Illinois at Chicago Damiano Rondelli 4
Froedtert and the Medical College of Wisconsin Parameswaran Hari 4
Jewish Hospital Edward Broun 4
Walter Reed Army Medical Center David Van Echo 4
Fox Chase Cancer Center Robert Emmons 4
Mayo Clinic-Scottsdale LAPS Craig Reeder 2
Indiana University – Melvin and Bren Simon Cancer Center LAPS Rafat Abonour 2
Illinois Oncology Research Association CCOP Patrick Gomez 2
Weill Medical College of Cornell University Tsiporah Shore 2
University of California San Diego Edward Ball 2
Geisinger Cancer Institute NCI Community Oncology Research Program SS Khan 1
University of Mississippi Medical Center Tondre Buck 1
University of Miami Miller School of Medicine-Sylvester Cancer Center Mark Goodman 1
Northwest NCI Community Oncology Research Program John Keech 1
Virginia Commonwealth University/Massey Cancer Center John McCarty 1
LDS Hospital Clyde Ford 1
Case Western Metro Health Medical Center Howard Lazarus 1

36