Supplemental Appendix

1

SUPPLEMENTAL APPENDIX

Table of Contents 1

DATA SOURCES 2

STATISTICAL ANALYSIS 2

SUPPLEMENTARY TABLES

Table S1 3

Table S2 4

Table S3 6

IMPACT OF PATTERN OF EARLY THERAPY FAILURE ON OUTCOMES 7

Table S6 8

MULTIVARIATE ANALYSIS 10

Data Sources

The CIBMTR is a working group of more than 500 transplantation centers worldwide contributing detailed data on HCT to a statistical center at the Medical College of Wisconsin (MCW). Participating centers are required to report all transplantations consecutively and patients are followed longitudinally. Computerized checks for discrepancies, physicians' review of submitted data, and on-site audits of participating centers ensure data quality.

Observational studies conducted by the CIBMTR are performed in compliance with all applicable federal regulations pertaining to the protection of human research participants. Protected Health Information used in the performance of such research is collected and maintained in CIBMTR’s capacity as a Public Health Authority under the HIPAA Privacy Rule. The Institutional Review Boards (IRB) of MCW and the National Marrow Donor Program approved this study.

The NLCS was a joint venture between Genentech Inc. (San Francisco, CA) and Biogen Idec (Cambridge, MA). All patients provided informed consent before participation, and the protocol was approved by IRB. Consecutive patients with newly diagnosed FL (from March 2004 to March 2007 at participating sites throughout the United States) were recruited. There was no central pathology review. All treatment strategies (including observation) for patients were recorded. Treatment and outcomes (including response, time to progression, and survival) were collected quarterly. Academic investigators had full access to the data.

Statistical Analysis

Sample size was based on the number of evaluable patients based on the CONSORT diagrams (figure 1a and 1b). Power analysis was conducted based on the left-truncated Cox model with Monte Carlo simulation size 2,000. We assumed 80% OS at 5 years for the transplant arm from when the clock starts. The censoring distribution and the late entry time distribution were chosen based on the current CIBMTR/NLCS dataset. The median censoring time was assumed to be 8 years based on the data. We have 85.5% power to detect a 15% difference (that is, 65% for NLCS at 5 years, HR=1.93). For the subset analysis (of patients undergoing autoHCT within 1 year), we have 82.25% power to detect a 16% difference (that is, 64% for NLCS at 5 years, HR=2).

Table S1. Variables included in Multivariate analysis

Main Effect :
·  No autoHCT consolidation (NLCS cohort) vs. AutoHCT consolidation (CIBMTR cohort)
Patient related:
·  Age at diagnosis: 18-39, 40-49, 50-59, ≥60 years and continuous
·  Gender: male vs. female
·  Race: White vs. Others vs Unknown
Disease related:
·  Disease stage at diagnosis: I/II vs III/IV
·  WHO Histological grade: grade I vs grade II vs grade III vs unknown
·  Bone marrow involvement at diagnosis: yes vs no vs. missing
·  Extranodal involvements at diagnosis: yes vs no vs. missing
·  Type of first line therapy: R-CHOP vs. R-CVP vs. R-Flu containing vs. R + others
·  Response to first line therapy (Pattern of first line therapy failure): CR/PR vs. SD vs. Progression vs. unknown
·  Time from diagnosis to 1st therapy failure: continuous

Table S2. Baseline characteristics of non-HCT patients and subgroup of auto-HCT cohort patients receiving transplantation within 1 year of early therapy failure.

Variable / No AutoHCT-cohort
(NLCS) (%) / AutoHCT-Cohort
(CIBMTR) (%) / p-value /
Number of patients / 174 / 123
Median age at diagnosis, years (range) / 55 (31-69) / 51 (22-69) / <0.0001
18-39 / 12 (7) / 19 (16)
40-49 / 33 (19) / 36 (29)
50-59 / 66 (38) / 41 (33)
60-64 / 31 (18) / 18 (15)
65-70 / 32 (18) / 9 (7)
Median Age at autoHCT, years (range) / N/A / 53 (23-71) / N/A
≥65 / 13 (11)
Male sex / 100 (57) / 73 (59) / 0.75
KPS at diagnosis
90-100% / 58 (33) / Not available / N/A
<90% / 52 (30)
Missing / 64 (37)
Karnofsky Score at autoHCT
90-100% / N/A / 77 (63)
<90% / 35 (28)
Missing / 11 (9)
Race / 0.04
Caucasian / 149 (86) / 114 (93)
African-American / 12 (7) / 3 (2)
Other / 13 (7) / 3 (2)
Unknown / 0 / 3 (2)
Advanced stage (III/IV) at diagnosis / 155 (89) / 102 (83) / 0.17
Missing / 0 / 1 (<1)
WHO Grade / 0.01
1/2 / 117 (67) / 69 (57)
3 / 40 (23) / 47 (38)
Unknown / 17 (10) / 7 (6)
Median time from diagnosis to HCT, months (range) / N/A / 18 (6-35)
Median time from treatment failure to HCT, months (range) / N/A / 6 (2-12)
Bone marrow positive at diagnosis / 86 (49) / 51 (41) / 0.75
Missing / 15 (8) / 7 (6)
Extranodal sites involved at diagnosis / 120 (69) / 80 (65) / 0.60
Missing / 7 (4) / 7 (6)
Normal LDH at diagnosis / 80 (46) / 25 (20) / 0.04
Unknown / 51 (29) / 71 (58)
FLIPI Score at diagnosis
Good (0-1) / 24 (14) / Not available
Intermediate (2) / 48 (28)
Poor (3-5) / 65 (37)
Unknown / 37 (21)
First line therapy / <0.0001
R-CHOP / 100 (57) / 102 (83)
R-CVP / 36 (21) / 11 (9)
R-Fludarabine-containing / 32 (18) / 4 (3)
R-Other / 6 (3) / 6 (5)
Response to first line therapy / 0.003
CR/PR / 116 (67) / 101 (82)
Progressive disease / 15 (9) / 3 (2)
Stable disease / 38 (21) / 14 (11)
Unknown / 5 (3) / 5 (4)
Median lines of therapy prior to HCT (range) / N/A / 2 (1-6)
Remission status before HCT
Complete remission / N/A / 54 (44)
Partial remission / 47 (38)
Chemorefractory / 20 (16)
Untreated / 1 (<1)
Missing / 1 (<1)
Conditioning Regimen
TBI-based / N/A / 13 (11)
BEAM and similar / 84 (68)
CBV or similar / 23 (19)
BuMEL/BuCy / 3 (1)
Year of transplant
2002-2003 / N/A / 2 (1)
2004-2005 / 26 (21)
2006-2007 / 46 (37)
2008-2009 / 46 (37)
2010-2011 / 3 (2)
Median follow-up of survivors (range), months-from 4 months after failure / 76 (2-130) / 69 (3-120)

Abbreviations: AutoHCT=autologous hematopoietic cell transplantation; BEAM=carmustine, etoposide, cytarabine, melphalan; BuCy=busulfan, cyclophosphamide; BuMel=busulfan, melpahlan; CBV=cyclophosphamide, carmustine; etoposide; CHOP=cyclophosphamide, doxorubicin, vincristine, prednisone; CIBMTR: Center for International Blood and Marrow Transplant Research; CVP= cyclophosphamide, vincristine, prednisone; FLIPI=follicular lymphoma international prognostic index; KPS=Karnofsky Performance Score; LDH=lactate dehydrogenase; NLCS=National LymphoCare study; R=rituximab; TBI=total body irradiation.

Table S3. Characteristics Propensity Score Matched Patients.

Variable / NLCS / CIBMTR / P-value /
Number of patients / 118 / 118
Age at diagnosis, median (range) / 57 (31-69) / 53 (22-70) / 0.11
30-39 / 6 (5) / 13 (11)
40-49 / 25 (21) / 35 (30)
50-59 / 46 (39) / 37 (31)
60-70 / 41 (35) / 33 (28)
Sex / 0.89
Male / 70 (59) / 71 (60)
Female / 48 (41) / 47 (40)
Race (propensity score matched) / 0.80
Caucasian / 109 (92) / 110 (93)
Other / 9 (8) / 8 (7)
Disease stage at diagnosis / 0.05
I/II / 12 (10) / 21 (18)
III/IV / 106 (90) / 94 (80)
Missing / 0 / 3 (3)
WHO Grade classification / 0.31
1 / 37 (31) / 32 (27)
2 / 39 (33) / 40 (34)
3 / 30 (25) / 43 (36)
Unknown / 12 (10) / 3 (3)
Time from diagnosis to early therapy failure / 0.36
≤1year / 59 (50) / 66 (56)
>1year / 59 (50) / 52 (44)
Time from diagnosis to HCT, months (range) / 16 (2-71)
Time from treatment failure to HCT, months (range) / 3 (<1-63)
Bone marrow involvement at diagnosis / 0.06
No / 45 (38) / 63 (53)
Yes / 64 (54) / 49 (42)
Missing / 9 (8) / 6 (5)
Extranodal involvement at diagnosis / 0.35
No / 30 (25) / 30 (25)
Yes / 86 (73) / 82 (69)
Missing / 2 (2) / 6 (5)
Normal LDH at diagnosis / 0.30
Yes / 56 (47) / 25 (21)
No / 34 (29) / 22 (19)
Missing / 28 (24) / 71 (60)
Front line therapy (propensity score matched) / 1.00
R-CHOP / 87 (74) / 88 (75)
R-CVP / 20 (17) / 19 (16)
R-Flu-containing / 6 (5) / 6 (5)
R-other / 5 (4) / 5 (4)
Response to front line therapy (propensity score matched) / 0.41
CR/PR / 90 (76) / 85 (72)
Progressive/stable disease / 25 (21) / 26 (22)
Missing / 3 (3) / 7 (6)

IMPACT OF PATTERN OF EARLY THERAPY FAILURE ON OUTCOMES:

Patients included in this analysis (from the NLCS and CIBMTR database) experienced early therapy failure (ETF) following frontline rituximab-based chemoimmunotherapies by either displaying primary refractory disease (i.e. stable disease [SD] or progressive disease [PD]) in response to first line therapies, or by exhibiting early relapse/progression after an initial partial remission (PR) or complete remission (CR) in response to the first line therapies. The pattern of ETF (primary refractory disease vs. early relapse/progression) was not predictive of overall survival (OS) on MVA analysis. Tables S3-S4 below, summarize univariate analyses assessing impact of the pattern of ETF on OS of patients in the NLSC and CIBMTR cohorts.

Table S4. Impact of the pattern of ETF on the OS of CIBMTR cohort (Univariate analysis; clock starts at 4 months after treatment failure)

/ Early Relapse/Progression / Primary Refractory Disease /
Outcomes / N Eval / Prob (95% CI) / N Eval / Prob (95% CI) / p-value /
Overall survival / 136 / 27
1-year / 91 (85-96)% / 88 (78-98)% / 0.61
2-year / 85 (78-92)% / 80 (67-95)% / 0.55
3-year / 79 (72-87)% / 73 (58-92)% / 0.51
5-year / 70 (62-78)% / 62 (44-86)% / 0.47

Table S5. . Impact of the pattern of ETF on the OS of NLCS cohort (Univariate analysis; clock starts at 4 months after treatment failure)

/ Early Relapse/Progression / Primary Refractory Disease /
Outcomes / N Eval / Prob (95% CI) / N Eval / Prob (95% CI) / p-value /
Overall survival / 116 / 53
1-year / 83 (75-89)% / 88 (75-94)% / 0.45
2-year / 77 (68-84)% / 77 (62-86)% / 0.98
3-year / 71 (62-78)% / 72 (57-83)% / 0.89
5-year / 58 (48-67)% / 65 (49-77)% / 0.46

Table S6. Baseline characteristics of auto-HCT patients receiving transplantation within 1 year of early therapy failure or after one year of early therapy failure.

Variable / Late AutoHCT
(%) / Early AutoHCT
(%) / p-value /
Number of patients / 52 / 123
Median age at diagnosis, years (range) / 57 (32-70) / 51 (22-69) / 0.01
Median Age at autoHCT, years (range) / 60 (35-75) / 53 (23-70) / <0.001
≥65 / 16 (31) / 13 (11) / 0.001
Male sex / 32 (62) / 73 (59) / 0.79
Karnofsky Score at autoHCT / 0.54
90-100% / 37 (71) / 77 (63)
<90% / 11 (21) / 35 (28)
Missing / 4 (8) / 11 (9)
Race / 0.67
Caucasian / 49 (94) / 114 (93)
African-American / 2 (4) / 3 (2)
Other / 0 / 3 (2)
Unknown / 1 (2) / 3 (2)
Advanced stage (III/IV) at diagnosis / 38 (73) / 102 (83) / 0.09
Missing / 3 (6) / 1 (<1)
WHO Grade / 0.59
1/2 / 35 (67) / 69 (57)
3 / 15 (29) / 47 (38)
Unknown / 2 (4) / 7 (6)
Median time from diagnosis to HCT, months (range) / 29 (14-82) / 14 (2-31) / <0.001
Bone marrow positive at diagnosis / 24 (46) / 51 (41) / 0.78
Missing / 2 (4) / 7 (6)
Extranodal sites involved at diagnosis / 37 (71) / 80 (65) / 0.71
Missing / 2 (4) / 7 (6)
Normal LDH at diagnosis / 9 (17) / 25 (20) / 0.78
Unknown / 33 (63) / 71 (58)
First line therapy / 0.33
R-CHOP / 37 (71) / 102 (83)
R-CVP / 9 (17) / 11 (9)
R-Fludarabine-containing / 2 (4) / 4 (3)
R-Other / 4 (8) / 6 (5)
Response to first line therapy / 0.04
CR/PR / 35 (67) / 101 (82)
Progressive disease / stable disease / 10 (19) / 17 (14)
Unknown / 7 (13) / 5 (4)
Median lines of therapy prior to HCT (range) / 3 (1-5) / 2 (1-4) / <0.001
Remission status before HCT / 0.08
Complete remission / 16 (31) / 54 (44)
Partial remission / 21 (40) / 47 (38)
Chemorefractory / 10 (19) / 20 (16)
Untreated / 1 (2) / 1 (<1)
Missing / 4 (8) / 1 (<1)
Conditioning Regimen / 0.11
TBI-based / 10 (19) / 13 (11)
BEAM and similar / 35 (67) / 84 (68)
CBV or similar / 4 (8) / 23 (19)
BuMEL/BuCy / 2 (4) / 3 (2)
Others / 1 (2) / 0
Year of transplant / 0.40
2002-2003 / 0 / 2 (2)
2004-2005 / 7 (13) / 26 (21)
2006-2007 / 19 (37) / 46 (37)
2008-2009 / 24 (46) / 46 (37)
2010-2012 / 2 (4) / 3 (2)
Median follow-up of survivors (range), months-from 4 months after failure / 87 (22-127) / 69 (3-120)

Abbreviations: AutoHCT=autologous hematopoietic cell transplantation; BEAM=carmustine, etoposide, cytarabine, melphalan; BuCy=busulfan, cyclophosphamide; BuMel=busulfan, melpahlan; CBV=cyclophosphamide, carmustine; etoposide; CHOP=cyclophosphamide, doxorubicin, vincristine, prednisone; CVP= cyclophosphamide, vincristine, prednisone; FLIPI=follicular lymphoma international prognostic index; KPS=Karnofsky Performance Score; LDH=lactate dehydrogenase; R=rituximab; TBI=total body irradiation.

Multivariate Analysis

We also examined MVA at the significance level 0.1. Using that significance level allowed inclusion of two additional variables for the MVA based on the entire data set: time from diagnosis to first therapy failure and gender. After adjusting for age and these two additional variables, the HR of the main effect was 0.82 with p-value 0.28. This result is still very similar to our original analysis (Table 3).

Multivariate analysis for overall survival (using a significance level of 0.1).

Parameter / p-value / HR / 95% CI
autoHCT vs. NLCS / 0.2754 / 0.816 / 0.566 / 1.176
Age 40-49 vs. 18-39 / 0.9937 / 1.003 / 0.445 / 2.262
Age 50-59 vs. 18-39 / 0.4419 / 1.350 / 0.629 / 2.897
Age ≥60 vs. 18-39 / 0.0238 / 2.367 / 1.121 / 4.997
Female vs. male / 0.0337 / 0.676 / 0.471 / 0.970
Interval from diagnosis to 1st therapy failure / 0.0506 / 2.529 / 0.998 / 6.413