- Email: [email protected]
Small Lymphocytic Lymphoma
Accepted Manuscript Time to second-line treatment and subsequent relative survival in older patients with relapsed CLL/SLL Eric M. Ammann, PhD, Tait D. Shanafelt, MD, Melissa C. Larson, MS, Kara B. Wright, MS, Bradley D. McDowell, PhD, Brian K. Link, MD, Elizabeth A. Chrischilles, PhD PII:
S2152-2650(17)30829-7
DOI:
10.1016/j.clml.2017.07.004
Reference:
CLML 975
To appear in:
Clinical Lymphoma, Myeloma and Leukemia
Received Date: 1 June 2017 Revised Date:
2152-2650 2152-2650
Accepted Date: 10 July 2017
Please cite this article as: Ammann EM, Shanafelt TD, Larson MC, Wright KB, McDowell BD, Link BK, Chrischilles EA, Time to second-line treatment and subsequent relative survival in older patients with relapsed CLL/SLL, Clinical Lymphoma, Myeloma and Leukemia (2017), doi: 10.1016/j.clml.2017.07.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Short title: Time to retreatment and survival in CLL/SLL
RI PT
Authors: Eric M. Ammann PhDa,b, Tait D. Shanafelt MDc, Melissa C. Larson MSc, Kara B. Wright MSa,b, Bradley D. McDowell PhDb, Brian K. Link MDb,d, Elizabeth A. Chrischilles PhDa,b
TE D
Email: [email protected] Phone: 319-384-1574
M AN U
Corresponding Author: Eric M. Ammann, PhD Department of Epidemiology College of Public Health Building S414 The University of Iowa Iowa City, IA 52242
SC
Affiliations: a. Department of Epidemiology, College of Public Health, University of Iowa, Iowa City, IA b. Holden Comprehensive Cancer Center, University of Iowa, Iowa City, IA c. Department of Internal Medicine, Mayo Clinic, Rochester, MN d. Department of Internal Medicine, Carver College of Medicine, University of Iowa, Iowa City, IA
Abstract word count: 250 Text word count: 3016 Figure/table count: 5 Reference count: 28 Data supplement for online posting included.
EP
32
Title: Time to second-line treatment and subsequent relative survival in older patients with relapsed CLL/SLL
AC C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
ACCEPTED MANUSCRIPT 2 MicroAbstract
34
We investigated whether stratifying by the interval between first-line (T1) and second-line (T2)
35
treatments could identify a subgroup of older patients with relapsed CLL/SLL with an expectation of
36
normal overall survival. Longer time-to-T2 was associated with a modestly improved prognosis;
37
however, even among those retreated ≥3 years after T1, survival was poor compared to the general
38
population (5-year relative survival=50%).
39
Abstract
40
Novel targeted therapies offer excellent short-term outcomes in patients with chronic lymphocytic
41
leukemia and small lymphocytic lymphoma (CLL/SLL). However, there is disagreement over how widely
42
these therapies should be used in place of standard chemo-immunotherapy (CIT). We investigated
43
whether stratification on the length of the interval between first-line (T1) and second-line (T2)
44
treatments could identify a subgroup of older patients with relapsed CLL/SLL with an expectation of
45
normal overall survival, and for whom CIT could be an acceptable treatment choice. Relapsed CLL/SLL
46
patients who received T2 were identified from the SEER-Medicare Linked Database. Five-year relative
47
survival (RS5, i.e., the ratio of observed survival to expected survival based on population life tables) was
48
assessed after stratifying patients on the interval between T1 and T2. We then validated our findings in
49
the Mayo Clinic CLL Database. Among 1,974 SEER-Medicare patients (median age = 77 years) who
50
received T2 for relapsed CLL/SLL, longer time-to-retreatment was associated with a modestly improved
51
prognosis (p = 0.01). However, even among those retreated ≥3 years after T1, survival was poor
52
compared to the general population (RS5 = 0.50 or lower in SEER-Medicare). Similar patterns were
53
observed in the younger Mayo validation cohort, though prognosis was better overall among the Mayo
54
patients, and patients with favorable FISH retreated ≥3 years after first-line treatment had close to
55
normal expected survival (RS5 = 0.87). Further research is needed to quantify the degree to which
AC C
EP
TE D
M AN U
SC
RI PT
33
ACCEPTED MANUSCRIPT 3 targeted therapies provide meaningful improvements over CIT in long-term outcomes for older patients
57
with relapsed CLL/SLL.
58 59 60 61 62 63 64 65
Keywords • Chronic lymphocytic leukemia • Small lymphocytic lymphoma • Relapse • Prognosis • Survival • Second-line treatment
AC C
EP
TE D
M AN U
SC
RI PT
56
ACCEPTED MANUSCRIPT 4 66
Introduction Novel therapies available targeting Bruton’s tyrosine kinase, PI-3 kinase, and bcl-2 have
68
demonstrated excellent short-term outcomes in patients with chronic lymphocytic leukemia and small
69
lymphocytic lymphoma (CLL/SLL).[1-3] Particularly for patients with an otherwise poor prognosis, these
70
treatments are likely to be transformative. However, these agents must be taken indefinitely, are costly
71
($100,000-150,000 annually),[4, 5] and carry risks of significant side effects (e.g., atrial fibrillation, colitis,
72
infection, and tumor lysis syndrome).[6] For these reasons, it is important to identify the CLL/SLL
73
patients for whom these new agents are most likely to provide a meaningful benefit compared to
74
standard chemo-immunotherapy (CIT) regimens. Additionally, it would be useful to identify subgroups
75
of patients with relapsed or refractory CLL/SLL who are at low risk for premature mortality with
76
standard CIT options, and thus would have a suitable alternative to long-term use of newer agents.
SC
M AN U
77
RI PT
67
In patients with relapsed CLL/SLL, the time interval from initial therapy until relapse is recognized as an important prognostic marker. Patients who relapse sooner are less responsive to
79
subsequent treatments and have shorter overall survival (OS).[7-9] This association has been
80
characterized in a younger patient cohort (median age <60 years) who received first-line treatment with
81
fludarabine, cyclophosphamide and rituximab (FCR). Median OS following relapse was 13 months among
82
patients who relapsed in less than 3 years, as compared with 63 months who relapsed in ≥3 years.[10]
83
However, since the median age at diagnosis for CLL/SLL is 71 years,[11] these results may not be
84
generalizable to the majority of CLL/SLL patients. Older patients typically receive less intensive
85
antineoplastic therapies, and OS patterns following relapse are likely to be different in older adults due
86
to competing risks of mortality. A more precise characterization of this relationship in older adults will
87
inform clinicians’ efforts to risk-stratify and appropriately treat older patients with relapsed CLL/SLL.
88
In this study we assessed the prognostic value of time-until-retreatment in predicting
89
AC C
EP
TE D
78
subsequent OS among older CLL/SLL cases identified from the Surveillance, Epidemiology, and End
ACCEPTED MANUSCRIPT 5 Results (SEER)-Medicare Linked Database (SEER-Medicare) and an independent validation cohort
91
identified from the Mayo Clinic CLL Database. We hypothesized that—prior to the era of novel targeted
92
therapies—patients retreated ≥3 years after their first course of antineoplastic therapy would have OS
93
close to that expected in the U.S. general population after controlling for age, sex and calendar year.
94
Methods
95
Data Sources
The study data for the SEER-Medicare patient cohort were obtained from the 2014 SEER-
SC
96
RI PT
90
Medicare linkage, which included SEER cancer cases diagnosed from 1973-2011 and their Medicare
98
claims from 1992-2013. Within SEER registry catchment areas, 93% of patients aged 65+ years
99
diagnosed with cancer have been linked to their Medicare claims data.[12] Participating registries
M AN U
97
collect data for all cancer patients diagnosed within their defined geographic area. Registry data include
101
month and year of diagnosis, age at diagnosis, race, tumor stage, and histology. Medicare files from the
102
Centers for Medicare and Medicaid Services (CMS) include demographic and enrollment information,
103
date of death, and all bills submitted for inpatient hospital care, outpatient hospital care, physician
104
services, and prescription fills.
Normative mortality rates based on the U.S. general population were obtained from The Human
EP
105
TE D
100
Mortality Database,[13] which assembles historical U.S. life tables based on data published by the U.S.
107
Census Bureau[14] and National Center for Health Statistics.[15] As described below in the statistical
108
methods section, these data were used for the relative survival estimates used to describe CLL/SLL
109
patients’ prognosis compared to age- and sex-matched controls.
110
Institutional Review Board Review and Research Ethics
111
AC C
106
This research project was approved by SEER-Medicare Program staff and the University of Iowa
112
Institutional Review Board (IRB). The University of Iowa IRB granted a waiver of informed consent
113
because the project consisted of a secondary analysis of existing data. SEER-Medicare Program staff
ACCEPTED MANUSCRIPT 6 reviewed the manuscript to ensure that it met reporting requirements to protect patient confidentiality.
115
The validation study in the Mayo Clinic CLL Database, described at the end of the methods section, was
116
performed with the approval of the Mayo Clinic IRB.
117
Cohort Definition
118
RI PT
114
Our study population consisted of patients who at age ≥66 years were diagnosed with CLL/SLL in 1992-2011. Cases were excluded if they had no specific diagnosis month recorded by SEER, had a prior
120
malignancy, had inconsistent birth or death dates recorded by SEER and CMS, were diagnosed at death,
121
did not have microscopic or laboratory diagnostic confirmation, were not enrolled in traditional fee-for-
122
service Medicare Parts A & B at diagnosis and the prior 365 days, or had claims for antineoplastic
123
therapy prior to the diagnosis date recorded by SEER (Figure 1). The age and Medicare eligibility
124
restrictions ensured that we could evaluate patients’ receipt of antineoplastic therapy, comorbidity
125
burden and proxies for CLL/SLL progression using Medicare claims data.
M AN U
Patients in this CLL/SLL inception cohort became eligible for inclusion in the present study upon
TE D
126
SC
119
receiving first- and second-line courses of antineoplastic therapy. Follow-up began on the date that
128
second-line therapy was initiated. Patients were excluded if prior to second-line treatment any of the
129
following censoring events occurred: death, the end of the study period (December 31, 2013), loss of
130
traditional Medicare enrollment, diagnosis with a second primary malignancy, or initiation of ibrutinib (a
131
novel Bruton’s tyrosine kinase inhibitor initially approved in 2013). Cohort identification steps are shown
132
in Figure 1.
AC C
133
EP
127
We restricted our study sample to patients whose first course of antineoplastic therapy was no
134
more than 365 days. For most CLL/SLL chemotherapy regimens, the treatment plan would consist of six
135
cycles of treatment at four-week intervals. Some patients may discontinue treatment early or take
136
breaks between treatment cycles, and treatment with some agents (e.g., chlorambucil or rituximab
137
monotherapy) may continue for 12 months. We further restricted to patients where the start of the
ACCEPTED MANUSCRIPT 7 second treatment course was at least 183 days after the end of the first treatment course. This
139
requirement was motivated by concerns that parsing shorter gaps between treatment episodes into
140
distinct treatment courses would be difficult based solely on claims data. In addition, patients who
141
relapse fewer than six months after their last antineoplastic treatment are considered to have refractory
142
CLL/SLL, and are known to have a poor prognosis.[9, 16]
143
Exposure Definition
The main prognostic factor of interest was time-until-retreatment, defined as the difference in
SC
144
RI PT
138
years between the end of the first course of antineoplastic therapy and the beginning of the second
146
course of antineoplastic therapy. Receipt of antineoplastic treatment was assessed using procedure,
147
diagnosis and drug codes recorded in Medicare inpatient, outpatient, and prescription claims. See
148
Appendix Tables A1 and A2 for details on the codes and code ranges that were used. Corticosteroids
149
were not included in our study definition of antineoplastic therapy because they are frequently used for
150
other indications, and rarely used as monotherapy for CLL/SLL. Drug-specific HCPCS and National Drug
151
Codes (NDCs) were used to identify specific antineoplastic agents administered. Patients were classified
152
as having received chemo-immunotherapy (CIT), chemotherapy, or immunotherapy. More details on the
153
specific treatment regimens that patients received are provided in Appendix Table A3.
TE D
EP
154
M AN U
145
A course of treatment was defined as beginning on the start date associated with the first claim where antineoplastic therapy was recorded, continuing for as long as additional claims were observed
156
(allowing for gaps of no more than 90 days), and ending on the end date of the last claim in the series. A
157
prior chart validation study performed in lymphoma patients found that Medicare claims data generally
158
provide reliable information on whether and when patients receive chemotherapy.[17]
159
Study Endpoint
AC C
155
160
The study endpoint was all-cause mortality. The SEER-Medicare dataset includes each patient’s
161
date of death as recorded in the Social Security Death Master File. In the files used in the present study
ACCEPTED MANUSCRIPT 8 162
from the 2014 SEER-Medicare linkage, mortality data were available through the end of 2013. Patients
163
who were alive on December 31, 2013, were right-censored in our survival analyses.
164
Statistical Methods Five-year relative survival (RS5) was used to characterize CLL/SLL patients’ prognosis. RS5—the
RI PT
165
ratio of five-year observed overall survival relative to overall survival in the general population after
167
conditioning on age, sex and calendar year—was estimated with the Ederer II method.[18-20] RS
168
measures are frequently used in cancer epidemiology as indirect estimates of the burden of cancer-
169
specific mortality in defined patient populations. The statistical significance of differences and trends in
170
RS across patient subgroups was assessed using the additive hazards model endorsed by Dickman et al.
171
for analyses of relative survival data within a generalized linear models framework.[19]
172
Covariates
M AN U
173
SC
166
A number of demographic and clinical characteristics were assessed in order to characterize our study sample (see Table I), and to explore possible modification of the relationship between time-to-
175
retreatment and relative survival. As a summary measure of comorbidity burden, we report a count of
176
the following 12 major comorbidities included in the National Cancer Institute and Charlson comorbidity
177
indices:[21-23] cerebrovascular disease, chronic pulmonary disease, congestive heart failure, dementia,
178
diabetes, hemiplegia/paraplegia, liver disease, myocardial infarction, peptic ulcer disease, renal disease,
179
rheumatic disease, and human immunodeficiency virus (HIV) infection. In addition, we provide data on
180
the frequency of several health conditions associated with advanced CLL/SLL disease, e.g., anemia and
181
infection.
EP
AC C
182
TE D
174
Health conditions and healthcare utilization were assessed based on administrative diagnosis
183
codes recorded during the year prior to retreatment. Following the approach of Klabunde et al.,[23] a
184
condition was considered present if a corresponding inpatient diagnosis code or two outpatient
185
diagnosis codes 30+ days apart were observed. A higher standard was required for outpatient diagnosis
ACCEPTED MANUSCRIPT 9 186
codes, because they can sometimes reflect diagnoses that were ruled out or merely considered as part
187
of a differential diagnosis.
188
Validation Study in Mayo Clinic CLL Database Data for the validation cohort were obtained from the Mayo Clinic CLL Database with the
RI PT
189
approval of the Mayo Clinic IRB. The database includes patients with a diagnosis of CLL/SLL seen in the
191
Division of Hematology at the Mayo Clinic since 1995. Clinical information regarding date of diagnosis,
192
baseline evaluation, prognostic parameters, treatment history, and disease-related complications was
193
abstracted from clinical records on all patients and maintained in the database on an ongoing
194
prospective basis. Additional details about the Mayo Clinic CLL Database may be found in prior
195
publications.[24, 25]
M AN U
SC
190
The validation study cohort included patients diagnosed with CLL/SLL in 1995-2013, and
197
longitudinal follow-up data through 2016. All CLL/SLL patients who received second-line therapy during
198
this period were identified, classified based on the time-to-retreatment, and followed until death or
199
censoring. Five-year relative survival (RS5) from the initiation of second-line therapy was estimated with
200
the relsurv package for R.[26, 27] RS5 was estimated for patient subgroups defined by time-to-
201
retreatment, type of first-line antineoplastic therapy received (CIT, chemotherapy or immunotherapy),
202
IGHV mutation status, and whether 11q or 17p chromosome deletions were detected through
203
fluorescence in situ hybridization (FISH).
204
Results
EP
AC C
205
TE D
196
During 1992-2011, there were 6044 CLL/SLL cases identified from SEER-Medicare who initiated
206
first-line therapy and were eligible for continued follow-up. Of this group, 1974 initiated second-line
207
therapy prior to censoring and were included in our analyses (Figure 1). As first-line treatment, 693
208
patients had received CIT (35%), 824 chemotherapy alone (42%) and 457 immunotherapy alone (23%).
209
Frequencies of specific treatment regimens are provided in Appendix Table A3. At retreatment, the
ACCEPTED MANUSCRIPT 10 210
patients had a median age of 77 years (interquartile range: 73, 82); 44% were female. Additional patient
211
characteristics are shown in Table I. Five-year relative survival (RS5) following second-line treatment was modestly higher in patients
213
retreated ≥3 years after first-line treatment (0.49; 95% CI: 0.40, 0.58) compared to patients retreated in
214
6 months to <2 years (0.42, 95% CI: 0.38, 0.45) or 2 to <3 years (0.42, 95% CI: 0.35, 0.50; test for trend: p
215
= 0.01). The relationship between time-until-retreatment and subsequent RS5 varied by the type of first-
216
line treatment received (Figure 2a). A protective association between longer time-until-retreatment and
217
RS5 was seen among patients who received chemotherapy alone (p = 0.004) or CIT (p = 0.054). No clear
218
pattern was observed in patients who received immunotherapy alone (p = 0.71) as first-line treatment.
SC
M AN U
219
RI PT
212
In additional subgroup analyses, we explored whether the association between time-untilretreatment and subsequent RS5 varied by number of treatment cycles received during first-line
221
therapy, patient age, or comorbidity burden. The relationship between longer time-until-retreatment
222
and improved subsequent RS5 was most pronounced among those whose first-line treatment course
223
included 3+ treatment cycles (Appendix Figure A1), who were age 80+ years or older (Appendix Figure
224
A2), and who had no major comorbidities (Appendix Figure A3). However, across all subgroups of
225
retreated patients, survival was meaningfully reduced compared to appropriate age and gender
226
matched cohort from the U.S. general population.
227
Mayo Clinic Validation Study
EP
AC C
228
TE D
220
The validation cohort consisted of 508 patients identified from the Mayo Clinic CLL Database
229
who received second-line antineoplastic therapy. The median patient age was 66 years (interquartile
230
range: 59, 74); 28% were female; 58%, 29%, and 12% had received CIT, chemotherapy alone, and
231
immunotherapy alone as first-line treatment, respectively (see Table II). RS5 estimates were 0.52, 0.60,
232
and 0.70 for patients retreated in 6 months to <2 years, 2 to <3 years, and 3+ years, respectively (test for
233
trend: p = 0.09). Similar patterns were observed when patients were stratified by type of first-line
ACCEPTED MANUSCRIPT 11 antineoplastic therapy received (Figure 2a). In subgroup analyses, we stratified patients by IGHV
235
mutation and FISH status. Time-to-retreatment was more strongly predictive of survival in patients with
236
favorable FISH status (i.e., no 11q or 17p deletions; Figure 3a) and in patients with unmutated IGHV
237
(Figure 3b), though these trends did not reach statistical significance.
238
Discussion
RI PT
234
In our primary population-based SEER-Medicare study of mortality among older patients with
240
relapsed CLL/SLL who received second-line therapy, we found that longer time-until-retreatment was
241
associated with moderately improved subsequent RS5. Contrary to our initial study hypothesis, and in
242
contradistinction to previous findings in follicular lymphoma – another chronic B-cell malignancy—
243
stratification on time-to-retreatment did not allow us to identify a patient subgroup with normal
244
longevity. Even among patients retreated 3+ years after the end of first-line therapy, mortality was high
245
among the SEER-Medicare patients relative to the general population (RS5 ≈ 0.50). These results
246
indicate that a significant number of older patients with relapsed CLL/SLL retreated with rituximab-era
247
CIT regimens are likely to have an unmet therapeutic need.
M AN U
TE D
248
SC
239
Overall, similar patterns were observed among patients in the Mayo Clinic CLL Database. Prognosis was better across the board for retreated patients in the younger Mayo cohort, with RS5
250
estimates ranging from 0.52 for patients retreated in 6 months to <2 years to 0.70 for patients retreated
251
in 3+ years. It is possible that the better prognosis observed in the Mayo cohort relative to the SEER-
252
Medicare cohort is attributable to its younger age (median age of 66 years versus 77 years in SEER-
253
Medicare cohort) and other case-mix differences, as well as differences in the selection of patients for
254
treatment.
255
AC C
EP
249
Because clinical prognostic markers were available in the Mayo Clinic CLL Database, we were
256
able to stratify the Mayo cohort by IGHV and FISH status. These subgroup analyses suggested that time-
257
to-retreatment had greater prognostic value among patients with unmutated IGHV and favorable FISH.
ACCEPTED MANUSCRIPT 12 258
Notably, among patients with time-to-retreatment ≥3 years and favorable FISH, subsequent survival was
259
close to normal (RS5 = 0.86). Survival was also close to normal for patients with time-to-retreatment ≥3
260
years who had received immunotherapy as first-line treatment (RS5 = 0.90). Our results differ from those reported in prior work in follicular lymphoma (FL), another typically
RI PT
261
chronic and indolent lymphoproliferative disease. In FL, it was demonstrated that early events following
263
initial therapy are strongly predictive of subsequent outcomes, independent of baseline clinical
264
prognostic scores.[28] In FL patients who received CIT as first-line treatment, disease progression within
265
24 months is predictive of meaningfully reduced overall survival. However, in FL patients without an
266
event in the 24 months following CIT (or 12 months following less aggressive treatment), overall survival
267
is comparable to expected rates based on the age and sex matched general population. Our differing
268
results may be explained by differences in the biology of FL and CLL/SLL, and in the typical trajectory of
269
disease progression following first-line treatment. In addition, the research questions in the FL study and
270
ours were slightly different: we focused on survival outcomes among patients who initiated second-line
271
treatment, whereas the FL study sought to characterize subsequent survival trajectories among patients
272
who at 24 months had not been retreated or experienced a relapse.
M AN U
TE D
Our primary study relied on data available from SEER registries and Medicare claims, and thus
EP
273
SC
262
had several limitations. Treatment dates and the type of antineoplastic therapy received were
275
determined from procedure and diagnosis codes recorded by healthcare providers for reimbursement
276
and Part D pharmacy claims. A prior validation study of billing codes for antineoplastic treatment in
277
Medicare beneficiaries with lymphoma indicated that claims data generally provide valid information on
278
chemotherapeutic agents and services dates.[17] We would expect that modest misclassification of the
279
timing and type of treatment would mean that the observed differences in RS5 across patient subgroups
280
may underestimate the true between-group differences in survival outcomes. In addition, clinical
281
prognostic markers were unavailable for the SEER-Medicare cohort.
AC C
274
ACCEPTED MANUSCRIPT 13 282
Since our RS5 estimates reflect comparisons between observed and expected survival based on mortality rates observed in the U.S. general population, it is important to consider the possibility of
284
treatment selection bias. The patients in our SEER-Medicare and Mayo cohorts were all determined to
285
be fit enough for first-and second-line antineoplastic therapy. For these reasons, our sample is likely to
286
be slightly healthier than the general population in terms of comorbidity burden and functional status.
287
Despite this possible source of bias toward improved survival outcomes in the CLL/SLL patients who
288
received antineoplastic therapy, survival rates were still meaningfully reduced among the majority of
289
patients receiving second-line therapy.
SC
RI PT
283
Our findings have several implications. During the rituximab era, the majority of older patients
291
with relapsed and retreated CLL/SLL had meaningfully reduced survival and a clear unmet therapeutic
292
need. Possible exceptions, based on subgroup analyses in the younger Mayo cohort, were patients with
293
favorable FISH and time-to-retreatment ≥3 years, or immunotherapy as first-line treatment and time-to-
294
retreatment ≥3 years. Further research is needed to quantify the degree to which targeted therapies
295
provide meaningful improvements over CIT in long-term outcomes for older patients with relapsed
296
CLL/SLL.
TE D
EP AC C
297
M AN U
290
ACCEPTED MANUSCRIPT 14 Acknowledgements:
RI PT
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
M AN U
SC
The collection of cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute's Surveillance, Epidemiology and End Results Program under contract HHSN261201000140C awarded to the Cancer Prevention Institute of California, contract HHSN261201000035C awarded to the University of Southern California, and contract HHSN261201000034C awarded to the Public Health Institute; and the Centers for Disease Control and Prevention's National Program of Cancer Registries, under agreement # U58DP003862-01 awarded to the California Department of Public Health. The ideas and opinions expressed herein are those of the author(s) and endorsement by the State of California Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their Contractors and Subcontractors is not intended nor should be inferred. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
TE D
Funding:
EP
Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number P50CA097274, by the University of Iowa Holden Comprehensive Cancer Center (HCCC) Population Research Core, which is supported in part by P30 CA086862, and by a Cancer & Aging Pilot Project Award from the University of Iowa HCCC and Center on Aging.
AC C
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330
ACCEPTED MANUSCRIPT 15 Authorship and conflict of interest statements: EMA contributed to the research design, analysis and interpretation of the data, and manuscript writing. He is now employed in Johnson & Johnson’s Medical Device Epidemiology research division. The analyses for this paper were completed and the manuscript was drafted prior to the start of that role.
RI PT
TDS contributed to the research design, analysis and interpretation of the data, and manuscript revisions. He reports no disclosures.
MCL contributed to the analysis and interpretation of the data and manuscript revisions. She reports no disclosures.
SC
KBW contributed to the analysis and interpretation of the data and manuscript revisions. She reports no disclosures.
M AN U
BDM contributed to the analysis and interpretation of the data and manuscript revisions. He reports no disclosures. BKL contributed to the research design, analysis and interpretation of the data, and manuscript revisions. He has received compensation as a consultant for Gilead and AbbVie.
EP
TE D
EAC contributed to the research design and manuscript revisions. She reports no disclosures.
AC C
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354
ACCEPTED MANUSCRIPT 16 References
EP
TE D
M AN U
SC
RI PT
1. Byrd JC, Brown JR, O'Brien S, et al. Ibrutinib versus ofatumumab in previously treated chronic lymphoid leukemia. N Engl J Med 2014;371:213-223. 2. Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in relapsed chronic lymphocytic leukemia. N Engl J Med 2014;370:997-1007. 3. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with Venetoclax in Relapsed Chronic Lymphocytic Leukemia. N Engl J Med 2016;374:311-322. 4. Shanafelt TD, Borah BJ, Finnes HD, et al. Impact of ibrutinib and idelalisib on the pharmaceutical cost of treating chronic lymphocytic leukemia at the individual and societal levels. J Oncol Pract 2015;11:252-258. 5. Johnson LA. FDA approves drug for tough-to-treat type of leukemia. Associated Press: Financial News; 2016. 6. McMullen JR, Boey EJ, Ooi JY, et al. Ibrutinib increases the risk of atrial fibrillation, potentially through inhibition of cardiac PI3K-Akt signaling. Blood 2014;124:3829-3830. 7. Rai KR, Stilgenbauer S. Treatment of relapsed or refractory chronic lymphocytic leukemia. UpToDate. Waltham, MA: UpToDate; 2016. 8. Brown JR. The treatment of relapsed refractory chronic lymphocytic leukemia. Hematology Am Soc Hematol Educ Program 2011;2011:110-118. 9. Hallek M. Chronic lymphocytic leukemia: 2015 Update on diagnosis, risk stratification, and treatment. Am J Hematol 2015;90:446-460. 10. Tam CS, O'Brien S, Plunkett W, et al. Long-term results of first salvage treatment in CLL patients treated initially with FCR (fludarabine, cyclophosphamide, rituximab). Blood 2014;124:3059-3064. 11. SEER Stat Fact Sheets: Chronic Lymphocytic Leukemia (CLL). National Cancer Institute: Surveillance, Epidemiology, and End Results Program; 2015. 12. Warren JL, Klabunde CN, Schrag D, et al. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Medical care 2002;40:IV-3-18. 13. Human Mortality Database. University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany); 2016. 14. U.S. Census Bureau. Population Estimates and Projection, Annual Estimates of the Resident Population by Single Year of Age and Sex for the United States. Washington, D.C.; , various years. 15. National Center for Health Statistics. Vital Statistics of the United States, Volume II: Mortality, Part A. Washington, D.C.: Government Printing Office; various years. 16. Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008;111:5446-5456. 17. Chen-Hardee S, Chrischilles EA, Voelker MD, et al. Population-based assessment of hospitalizations for neutropenia from chemotherapy in older adults with non-Hodgkin's lymphoma (United States). Cancer Causes Control 2006;17:647-654. 18. Cho H, Howlader N, Mariotto AB, et al. Technical Report #2011-01: Estimating relative survival for cancer patients from the SEER Program using expected rates based on Ederer I versus Ederer II method. National Cancer Institute Surveillance Research Program; 2011. 19. Dickman PW, Sloggett A, Hills M, et al. Regression models for relative survival. Stat Med 2004;23:51-64. 20. Ederer F, Heise H. Methodological note No. 10: Instructions to IBM 650 programmers in processing survival computations. Bethesda, MD: National Cancer Institute End Results Evaluation Section; 1959.
AC C
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
ACCEPTED MANUSCRIPT 17
EP
TE D
M AN U
SC
RI PT
21. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-383. 22. Klabunde CN, Legler JM, Warren JL, et al. A refined comorbidity measurement algorithm for claims-based studies of breast, prostate, colorectal, and lung cancer patients. Ann Epidemiol 2007;17:584-590. 23. Klabunde CN, Potosky AL, Legler JM, et al. Development of a comorbidity index using physician claims data. J Clin Epidemiol 2000;53:1258-1267. 24. Thurmes P, Call T, Slager S, et al. Comorbid conditions and survival in unselected, newly diagnosed patients with chronic lymphocytic leukemia. Leuk Lymphoma 2008;49:49-56. 25. Shanafelt TD, Jenkins G, Call TG, et al. Validation of a new prognostic index for patients with chronic lymphocytic leukemia. Cancer 2009;115:363-372. 26. Pohar M, Stare J. Relative survival analysis in R. Comput Methods Programs Biomed 2006;81:272-278. 27. Pohar M, Stare J. Making relative survival analysis relatively easy. Comput Biol Med 2007;37:1741-1749. 28. Casulo C, Byrtek M, Dawson KL, et al. Early Relapse of Follicular Lymphoma After Rituximab Plus Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone Defines Patients at High Risk for Death: An Analysis From the National LymphoCare Study. J Clin Oncol 2015;33:2516-2522.
AC C
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
ACCEPTED MANUSCRIPT 18
EP
TE D
M AN U
SC
RI PT
Figure 1. Flow diagram showing identification of eligible SEER-Medicare patients with chronic lymphocytic leukemia or small lymphocytic lymphoma (CLL/SLL).
AC C
422 423
ACCEPTED MANUSCRIPT 19
EP
TE D
M AN U
SC
RI PT
Figure 2. Estimated five-year relative survival following initiation of second-line treatment in the SEERMedicare CLL/SLL patient cohort and Mayo validation cohort.
AC C
424 425 426 427
ACCEPTED MANUSCRIPT 20
EP
TE D
M AN U
SC
RI PT
Figure 3. Estimated five-year relative survival following initiation of second-line treatment in Mayo validation cohort, stratified by FISH and IGHV mutation status
AC C
428 429
ACCEPTED MANUSCRIPT 21 Table I. Characteristics of eligible patients who received second-line treatment stratified by first-line therapy. ChemoChemoImmunoAll immunotherapy therapy Characteristic treatments therapy alone alone (N=1974) (N=693) (N=824) (N=457) Age at second-line treatment in years 653 (33%) 277 (40%) 268 (33%) 108 (24%) • 66-74 577 (29%) 202 (29%) 261 (32%) 114 (25%) • 75-79 744 (38%) 214 (31%) 295 (36%) 235 (51%) • 80+ Sex 874 (44%) 266 (38%) 388 (47%) 220 (48%) • Female 1100 (56%) 427 (62%) 436 (53%) 237 (52%) • Male Number of treatment cycles included in first-line treatment course 650 (33%) 109 (16%) 231 (28%) 310 (68%) • 1-2 1324 (67%) 584 (84%) 593 (72%) 147 (32%) • 3+ Interval between first- and secondline antineoplastic treatments 1325 (67%) 434 (63%) 531 (64%) 360 (79%) • 6 months to <2 years 355 (18%) 155 (22%) 142 (17%) 58 (13%) • 2 years to <3 years 294 (15%) 104 (15%) 151 (18%) 39 (9%) • 3+ years Type of second-line antineoplastic treatment 576 (29%) 298 (43%) 183 (22%) 95 (21%) • Chemo-immunotherapy 486 (25%) 80 (12%) 375 (46%) 31 (7%) • Chemotherapy alone 609 (31%) 207 (30%) 122 (15%) 280 (61%) • Immunotherapy alone 303 (15%) 108 (16%) 144 (17%) 51 (11%) • Unknown (non-specific billing codes) Indicators of CLL/SLL progression 916 (46%) 331 (48%) 345 (42%) 240 (53%) • Anemia 358 (18%) 145 (21%) 135 (16%) 78 (17%) • Thrombocytopenia 282 (14%) 147 (21%) 93 (11%) 42 (9%) • Immune deficiency Count of major comorbidities* 964 (49%) 330 (48%) 420 (51%) 214 (47%) • None 566 (29%) 197 (28%) 236 (29%) 133 (29%) • One 444 (22%) 166 (24%) 168 (20%) 110 (24%) • Two or more Hospitalized in prior year 777 (39%) 269 (39%) 335 (41%) 173 (38%)
432 433 434 435 436 437
*The major comorbidities covariate, adapted from the National Cancer Institute and Charlson comorbidity indices, is a count of the following 12 comorbidities: cerebrovascular disease, chronic pulmonary disease, congestive heart failure, dementia, diabetes, hemiplegia/paraplegia, liver disease, myocardial infarction, peptic ulcer disease, renal disease, rheumatic disease, and HIV infection.
AC C
EP
TE D
M AN U
SC
RI PT
430 431
ACCEPTED MANUSCRIPT 22 Table II. Baseline characteristics of eligible patients in Mayo validation cohort who received secondline treatment. Characteristic Patient count (N = 508) Age at second-line treatment in years 245 (48%) • ≤65 148 (29%) • 66-74 77 (15%) • 75-79 38 (7%) • 80+ Sex – N (%) 144 (28%) • Female 364 (72%) • Male Type of first-line treatment received – N (%) 295 (58%) • Chemo-immunotherapy (CIT) 148 (29%) • Chemotherapy alone 59 (12%) • Immunotherapy alone 6 (1%) • Missing Interval between end of first-line therapy and start of second-line therapy – N (%) 275 (54%) • 6 months to <2 years 105 (21%) • 2 years to <3 years 128 (25%) • 3+ years FISH status – N (%) 52 (10%) • High risk (11q- or 17p-) 187 (37%) • Other/favorable 269 (53%) • Missing IGHV mutation status – N (%) 217 (43%) • Unmutated 84 (17%) • Mutated 207 (41%) • Missing
EP AC C
440 441 442
TE D
M AN U
SC
RI PT
438 439
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
EP
TE D
M AN U
SC
RI PT
Appendix Table A1: Administrative diagnosis and procedure codes used to identify receipt of chemotherapy and immunotherapy. Valid interval Valid interval Definition Code Type Code start date end date alemtuzumab CPT/HCPCS J9010 01/01/2003 . bendamustine CPT/HCPCS J9033 01/01/2009 . chemotherapy DRG 410 . . chemotherapy CITD-9 V5811 . . DIAGNOSIS immunotherapy CITD-9 V5812 . . DIAGNOSIS chemotherapy CITD-9 9925 . . PROCEDURE chemotherapy* CPT/HCPCS 964xx . . chemotherapy CPT/HCPCS 9650x . . chemotherapy CPT/HCPCS 9651x . . chemotherapy CPT/HCPCS 9652x . . chemotherapy CPT/HCPCS 9653x . . chemotherapy CPT/HCPCS 9654x . . chemotherapy CPT/HCPCS G0355 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0356 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0357 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0358 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0359 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0360 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0361 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0362 01/01/2005 12/31/2005 chemotherapy* CPT/HCPCS J9xxx . . chemotherapy CPT/HCPCS Q0083 . . chemotherapy CPT/HCPCS Q0084 . . chemotherapy CPT/HCPCS Q0085 . . chemotherapy Revenue 0331 . . chemotherapy Revenue 0332 . . chemotherapy Revenue 0335 . . chlorambucil CPT/HCPCS S0172 01/01/2002 . cladribine CPT/HCPCS J9065 01/01/1995 . cyclophosphamide CPT/HCPCS J8530 01/01/1995 . cyclophosphamide CPT/HCPCS J9070 01/01/1984 . cyclophosphamide CPT/HCPCS J9080 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9090 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9091 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9092 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9093 01/01/1988 12/31/2010 cyclophosphamide CPT/HCPCS J9094 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9095 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9096 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9097 01/01/1994 12/31/2010
AC C
1 2
1
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
2
TE D
M AN U
SC
RI PT
Valid interval Valid interval Definition Code Type Code start date end date doxorubicin CPT/HCPCS J9000 01/01/1984 . doxorubicin CPT/HCPCS J9001 01/01/2000 12/31/2012 doxorubicin CPT/HCPCS J9002 01/01/2013 12/31/2013 doxorubicin CPT/HCPCS Q2048 07/01/2012 12/31/2012 doxorubicin CPT/HCPCS Q2049 07/01/2012 . doxorubicin CPT/HCPCS Q2050 07/01/2013 . fludarabine CPT/HCPCS C9262 04/01/2010 06/30/2010 fludarabine CPT/HCPCS J8562 01/01/2011 . fludarabine CPT/HCPCS J9185 01/01/1994 . fludarabine CPT/HCPCS Q2025 07/01/2010 12/31/2010 mitoxantrone CPT/HCPCS J9293 01/01/1990 . obinutuzumab CPT/HCPCS J9301 . . ofatumumab CPT/HCPCS C9260 04/01/2010 12/31/2010 ofatumumab CPT/HCPCS J9302 01/01/2011 . pentostatin CPT/HCPCS J9268 01/01/1994 . rituximab CPT/HCPCS J9310 01/01/1999 . vincristine CPT/HCPCS J9370 01/01/1986 . vincristine CPT/HCPCS J9371 01/01/2014 . vincristine CPT/HCPCS J9375 01/01/1988 12/31/2010 vincristine CPT/HCPCS J9380 01/01/1994 12/31/2010 * A subset of codes in the CPT/HCPCS ranges 964xx and J9xxx were excluded as being unlikely to represent antineoplastic therapy for CLL/SLL. These codes are listed in Appendix Table A2.
AC C
3 4 5 6
EP
In addition to the procedure and diagnosis codes listed above, National Drug Codes (NDCs) obtained from the 2015 Cerner Multum Lexicon were used to define receipt of the following drugs: alemtuzumab, bendamustine, chlorambucil, cladribine, cyclophosphamide, doxorubicin, fludarabine, ibrutinib, idelalisib, lenalidomide, mitoxantrone, obinutuzumab, ofatumumab, pentostatin, rituximab, vincristine.
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
EP
TE D
M AN U
SC
RI PT
Appendix Table A2: Specific chemotherapy, immunotherapy and related billing codes excluded as unlikely to represent treatment for CLL/SLL. Definition Code Type Code Percentage of all chemo-related codes observed in cohort Aldesleukin injection CPT/HCPCS J9015 0.00 Arsenic trioxide injection CPT/HCPCS J9017 0.02 Asparaginase, nos CPT/HCPCS J9020 0.00 Azacitidine injection CPT/HCPCS J9025 0.21 Bcg live intravesical vac CPT/HCPCS J9031 0.10 Bevacizumab injection CPT/HCPCS J9035 0.32 Bleomycin sulfate injection CPT/HCPCS J9040 0.06 Carmustine injection CPT/HCPCS J9050 0.00 Cetuximab injection CPT/HCPCS J9055 0.04 CHEMO HORMON ANTINEOPL SQ/IM CPT/HCPCS 96402 0.19 chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9240 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9295 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9376 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9381 0.00 range but definition unavailable) CHEMOTHERAPY INTO CNS CPT/HCPCS 96450 0.04 Clofarabine injection CPT/HCPCS J9027 0.00 DACARBAZINE 100 MG CPT/HCPCS J9130 0.03 Dacarbazine 200 mg inj CPT/HCPCS J9140 0.02 Dactinomycin injection CPT/HCPCS J9120 0.00 Daunorubicin injection CPT/HCPCS J9150 0.00 Degarelix injection CPT/HCPCS J9155 0.00 Denileukin diftitox inj CPT/HCPCS J9160 0.00 Diethylstilbestrol injection CPT/HCPCS J9165 0.00 Docetaxel injection CPT/HCPCS J9171 0.03 Docetaxel injection CPT/HCPCS J9170 0.05 Elliotts b solution per ml CPT/HCPCS J9175 0.00 Epirubicin hydrochloride, 50 mg CPT/HCPCS J9180 0.00 Floxuridine injection CPT/HCPCS J9200 0.00 Fluorouracil injection CPT/HCPCS J9190 0.23 Gemcitabine hcl injection CPT/HCPCS J9201 0.19 Gemtuzumab ozogamicin inj CPT/HCPCS J9300 0.00 Goserelin acetate implant CPT/HCPCS J9202 0.10 HISTRELIN IMPLANT VANTAS 50 MG CPT/HCPCS J9225 0.00 Inj melphalan hydrochl 50 mg CPT/HCPCS J9245 0.00 INJ PEGASPARGASE SINGLE DOSE V CPT/HCPCS J9266 0.00 Inj, epirubicin hcl, 2 mg CPT/HCPCS J9178 0.00 INJECTION IDARUBCITIN HCL 5 MG CPT/HCPCS J9211 0.00 INJECTION IRINOTECAN 20 MG CPT/HCPCS J9206 0.03
AC C
7 8
3
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS
TE D
EP
AC C
9 10 11
Percentage of all chemo-related codes observed in cohort J9209 0.04 J9280 0.02 J9263 0.05 J9265 0.19 J9350 0.01 J9360 0.07 J9395 0.00 J9213 0.05 J9214 0.17 J9212 0.02 J9228 0.00 96523 1.29 J9217 0.21 J9219 0.00 J9218 0.00 J9230 0.00 J9290 0.00 J9291 0.01 J9264 0.02 J9303 0.00 J9305 0.03 J9270 0.00 96521 0.07 96522 0.07 J9315 0.00 J9320 0.00 J9330 0.01 J9340 0.00 J9351 0.00 J9355 0.03 J9357 0.00 J9390 0.02
RI PT
INJECTION MESNA 200 MG INJECTION MITOMYCIN 5 MG INJECTION OXALIPLATIN 0.5 MG INJECTION PACLITAXEL 30 MG INJECTION TOPOTECAN 4 MG INJECTION VINBLASTINE SULFATE Injection, fulvestrant Interferon alfa-2a inj Interferon alfa-2b inj Interferon alfacon-1 inj Ipilimumab injection IRRIG DRUG DELIVERY DEVCITE LEUPROLIDE ACETATE 7.5 MG Leuprolide acetate implant Leuprolide acetate injeciton Mechlorethamine hcl inj MITOMYCIN 20 MG Mitomycin 40 mg inj Paclitaxel protein bound Panitumumab injection Pemetrexed injection Plicamycin (mithramycin) inj REFILL/MAINT PORTABLE PUMP REFILL/MAINT PUMP/RESVR SYST Romidepsin injection Streptozocin injection Temsirolimus injection Thiotepa injection Topotecan injection Trastuzumab injection Valrubicin injection Vinorelbine tartrate inj
Code
SC
Code Type
M AN U
Definition
4
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement Appendix Table A3. Categorization of patients’ first-line treatment regimens. Type of antineoplastic treatment regimen
Specific regimen as determined from drug-specific HCPCS and NDC codes
Frequency Percent
Chemoimmunotherapy (N = 693)
Fludarabine, cyclophosphamide and immunotherapy
179
Fludarabine and immunotherapy (Pentostatin or cladribine) and immunotherapy Bendamustine and immunotherapy
254
12.87
21
1.06
50
2.53
48
2.43
SC
Cyclophosphamide, vincristine, doxorubicin and immunotherapy
None of the above. Immunotherapy and at least one of the following: chlorambucil, cyclophosphamide, vincristine, doxorubicin or mitoxantrone
39
1.97
Fludarabine and cyclophosphamide
64
3.24
Fludarabine, pentostatin or cladribine*
427
21.64
Bendamustine
16
0.81
Cyclophosphamide, vincristine & doxorubicin
37
1.87
Cyclophosphamide & vincristine
137
6.94
None of the above. At least one of the following: chlorambucil, cyclophosphamide, vincristine, doxorubicin or mitoxantrone
143
7.25
Rituximab, ofatumumab or alemtuzumab*
457
23.15
M AN U
5.17
* Cells with fewer than 11 patients combined due to SEER-Medicare reporting requirements. Of the 427 patients who received fludarabine, pentostatin, or cladribine, fewer than 11 received pentostatin or cladribine. Of the 457 patients who received rituximab, ofatumumab or alemtuzumab, fewer than 11 patients received ofatumumab or alemtuzumab.
AC C
13 14 15 16 17 18
102
EP
Immunotherapy alone (N = 457)
Cyclophosphamide, vincristine and immunotherapy
TE D
Chemotherapy alone (N = 824)
9.07
RI PT
12
5
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
6
Appendix Figure A1. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by number of treatment cycles received during first-line treatment and the time interval between first- and second-line treatments.
Trend: p = 0.67
AC C
EP
TE D
M AN U
SC
RI PT
Trend: p < 0.001
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
7
Appendix Figure A2. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by patient age and time interval between first- and second-line treatments.
Trend: p = 0.004
RI PT
Trend: p = 0.099
AC C
EP
TE D
M AN U
SC
Trend: p = 0.107
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
8
Appendix Figure A3. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by major comorbidity count and time interval between first- and second-line treatments. Trend: p = 0.001 Trend: p = 0.555
AC C
EP
TE D
M AN U
SC
RI PT
Trend: p = 0.262
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S2152-2650(17)30829-7
DOI:
10.1016/j.clml.2017.07.004
Reference:
CLML 975
To appear in:
Clinical Lymphoma, Myeloma and Leukemia
Received Date: 1 June 2017 Revised Date:
2152-2650 2152-2650
Accepted Date: 10 July 2017
Please cite this article as: Ammann EM, Shanafelt TD, Larson MC, Wright KB, McDowell BD, Link BK, Chrischilles EA, Time to second-line treatment and subsequent relative survival in older patients with relapsed CLL/SLL, Clinical Lymphoma, Myeloma and Leukemia (2017), doi: 10.1016/j.clml.2017.07.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Short title: Time to retreatment and survival in CLL/SLL
RI PT
Authors: Eric M. Ammann PhDa,b, Tait D. Shanafelt MDc, Melissa C. Larson MSc, Kara B. Wright MSa,b, Bradley D. McDowell PhDb, Brian K. Link MDb,d, Elizabeth A. Chrischilles PhDa,b
TE D
Email: [email protected] Phone: 319-384-1574
M AN U
Corresponding Author: Eric M. Ammann, PhD Department of Epidemiology College of Public Health Building S414 The University of Iowa Iowa City, IA 52242
SC
Affiliations: a. Department of Epidemiology, College of Public Health, University of Iowa, Iowa City, IA b. Holden Comprehensive Cancer Center, University of Iowa, Iowa City, IA c. Department of Internal Medicine, Mayo Clinic, Rochester, MN d. Department of Internal Medicine, Carver College of Medicine, University of Iowa, Iowa City, IA
Abstract word count: 250 Text word count: 3016 Figure/table count: 5 Reference count: 28 Data supplement for online posting included.
EP
32
Title: Time to second-line treatment and subsequent relative survival in older patients with relapsed CLL/SLL
AC C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
ACCEPTED MANUSCRIPT 2 MicroAbstract
34
We investigated whether stratifying by the interval between first-line (T1) and second-line (T2)
35
treatments could identify a subgroup of older patients with relapsed CLL/SLL with an expectation of
36
normal overall survival. Longer time-to-T2 was associated with a modestly improved prognosis;
37
however, even among those retreated ≥3 years after T1, survival was poor compared to the general
38
population (5-year relative survival=50%).
39
Abstract
40
Novel targeted therapies offer excellent short-term outcomes in patients with chronic lymphocytic
41
leukemia and small lymphocytic lymphoma (CLL/SLL). However, there is disagreement over how widely
42
these therapies should be used in place of standard chemo-immunotherapy (CIT). We investigated
43
whether stratification on the length of the interval between first-line (T1) and second-line (T2)
44
treatments could identify a subgroup of older patients with relapsed CLL/SLL with an expectation of
45
normal overall survival, and for whom CIT could be an acceptable treatment choice. Relapsed CLL/SLL
46
patients who received T2 were identified from the SEER-Medicare Linked Database. Five-year relative
47
survival (RS5, i.e., the ratio of observed survival to expected survival based on population life tables) was
48
assessed after stratifying patients on the interval between T1 and T2. We then validated our findings in
49
the Mayo Clinic CLL Database. Among 1,974 SEER-Medicare patients (median age = 77 years) who
50
received T2 for relapsed CLL/SLL, longer time-to-retreatment was associated with a modestly improved
51
prognosis (p = 0.01). However, even among those retreated ≥3 years after T1, survival was poor
52
compared to the general population (RS5 = 0.50 or lower in SEER-Medicare). Similar patterns were
53
observed in the younger Mayo validation cohort, though prognosis was better overall among the Mayo
54
patients, and patients with favorable FISH retreated ≥3 years after first-line treatment had close to
55
normal expected survival (RS5 = 0.87). Further research is needed to quantify the degree to which
AC C
EP
TE D
M AN U
SC
RI PT
33
ACCEPTED MANUSCRIPT 3 targeted therapies provide meaningful improvements over CIT in long-term outcomes for older patients
57
with relapsed CLL/SLL.
58 59 60 61 62 63 64 65
Keywords • Chronic lymphocytic leukemia • Small lymphocytic lymphoma • Relapse • Prognosis • Survival • Second-line treatment
AC C
EP
TE D
M AN U
SC
RI PT
56
ACCEPTED MANUSCRIPT 4 66
Introduction Novel therapies available targeting Bruton’s tyrosine kinase, PI-3 kinase, and bcl-2 have
68
demonstrated excellent short-term outcomes in patients with chronic lymphocytic leukemia and small
69
lymphocytic lymphoma (CLL/SLL).[1-3] Particularly for patients with an otherwise poor prognosis, these
70
treatments are likely to be transformative. However, these agents must be taken indefinitely, are costly
71
($100,000-150,000 annually),[4, 5] and carry risks of significant side effects (e.g., atrial fibrillation, colitis,
72
infection, and tumor lysis syndrome).[6] For these reasons, it is important to identify the CLL/SLL
73
patients for whom these new agents are most likely to provide a meaningful benefit compared to
74
standard chemo-immunotherapy (CIT) regimens. Additionally, it would be useful to identify subgroups
75
of patients with relapsed or refractory CLL/SLL who are at low risk for premature mortality with
76
standard CIT options, and thus would have a suitable alternative to long-term use of newer agents.
SC
M AN U
77
RI PT
67
In patients with relapsed CLL/SLL, the time interval from initial therapy until relapse is recognized as an important prognostic marker. Patients who relapse sooner are less responsive to
79
subsequent treatments and have shorter overall survival (OS).[7-9] This association has been
80
characterized in a younger patient cohort (median age <60 years) who received first-line treatment with
81
fludarabine, cyclophosphamide and rituximab (FCR). Median OS following relapse was 13 months among
82
patients who relapsed in less than 3 years, as compared with 63 months who relapsed in ≥3 years.[10]
83
However, since the median age at diagnosis for CLL/SLL is 71 years,[11] these results may not be
84
generalizable to the majority of CLL/SLL patients. Older patients typically receive less intensive
85
antineoplastic therapies, and OS patterns following relapse are likely to be different in older adults due
86
to competing risks of mortality. A more precise characterization of this relationship in older adults will
87
inform clinicians’ efforts to risk-stratify and appropriately treat older patients with relapsed CLL/SLL.
88
In this study we assessed the prognostic value of time-until-retreatment in predicting
89
AC C
EP
TE D
78
subsequent OS among older CLL/SLL cases identified from the Surveillance, Epidemiology, and End
ACCEPTED MANUSCRIPT 5 Results (SEER)-Medicare Linked Database (SEER-Medicare) and an independent validation cohort
91
identified from the Mayo Clinic CLL Database. We hypothesized that—prior to the era of novel targeted
92
therapies—patients retreated ≥3 years after their first course of antineoplastic therapy would have OS
93
close to that expected in the U.S. general population after controlling for age, sex and calendar year.
94
Methods
95
Data Sources
The study data for the SEER-Medicare patient cohort were obtained from the 2014 SEER-
SC
96
RI PT
90
Medicare linkage, which included SEER cancer cases diagnosed from 1973-2011 and their Medicare
98
claims from 1992-2013. Within SEER registry catchment areas, 93% of patients aged 65+ years
99
diagnosed with cancer have been linked to their Medicare claims data.[12] Participating registries
M AN U
97
collect data for all cancer patients diagnosed within their defined geographic area. Registry data include
101
month and year of diagnosis, age at diagnosis, race, tumor stage, and histology. Medicare files from the
102
Centers for Medicare and Medicaid Services (CMS) include demographic and enrollment information,
103
date of death, and all bills submitted for inpatient hospital care, outpatient hospital care, physician
104
services, and prescription fills.
Normative mortality rates based on the U.S. general population were obtained from The Human
EP
105
TE D
100
Mortality Database,[13] which assembles historical U.S. life tables based on data published by the U.S.
107
Census Bureau[14] and National Center for Health Statistics.[15] As described below in the statistical
108
methods section, these data were used for the relative survival estimates used to describe CLL/SLL
109
patients’ prognosis compared to age- and sex-matched controls.
110
Institutional Review Board Review and Research Ethics
111
AC C
106
This research project was approved by SEER-Medicare Program staff and the University of Iowa
112
Institutional Review Board (IRB). The University of Iowa IRB granted a waiver of informed consent
113
because the project consisted of a secondary analysis of existing data. SEER-Medicare Program staff
ACCEPTED MANUSCRIPT 6 reviewed the manuscript to ensure that it met reporting requirements to protect patient confidentiality.
115
The validation study in the Mayo Clinic CLL Database, described at the end of the methods section, was
116
performed with the approval of the Mayo Clinic IRB.
117
Cohort Definition
118
RI PT
114
Our study population consisted of patients who at age ≥66 years were diagnosed with CLL/SLL in 1992-2011. Cases were excluded if they had no specific diagnosis month recorded by SEER, had a prior
120
malignancy, had inconsistent birth or death dates recorded by SEER and CMS, were diagnosed at death,
121
did not have microscopic or laboratory diagnostic confirmation, were not enrolled in traditional fee-for-
122
service Medicare Parts A & B at diagnosis and the prior 365 days, or had claims for antineoplastic
123
therapy prior to the diagnosis date recorded by SEER (Figure 1). The age and Medicare eligibility
124
restrictions ensured that we could evaluate patients’ receipt of antineoplastic therapy, comorbidity
125
burden and proxies for CLL/SLL progression using Medicare claims data.
M AN U
Patients in this CLL/SLL inception cohort became eligible for inclusion in the present study upon
TE D
126
SC
119
receiving first- and second-line courses of antineoplastic therapy. Follow-up began on the date that
128
second-line therapy was initiated. Patients were excluded if prior to second-line treatment any of the
129
following censoring events occurred: death, the end of the study period (December 31, 2013), loss of
130
traditional Medicare enrollment, diagnosis with a second primary malignancy, or initiation of ibrutinib (a
131
novel Bruton’s tyrosine kinase inhibitor initially approved in 2013). Cohort identification steps are shown
132
in Figure 1.
AC C
133
EP
127
We restricted our study sample to patients whose first course of antineoplastic therapy was no
134
more than 365 days. For most CLL/SLL chemotherapy regimens, the treatment plan would consist of six
135
cycles of treatment at four-week intervals. Some patients may discontinue treatment early or take
136
breaks between treatment cycles, and treatment with some agents (e.g., chlorambucil or rituximab
137
monotherapy) may continue for 12 months. We further restricted to patients where the start of the
ACCEPTED MANUSCRIPT 7 second treatment course was at least 183 days after the end of the first treatment course. This
139
requirement was motivated by concerns that parsing shorter gaps between treatment episodes into
140
distinct treatment courses would be difficult based solely on claims data. In addition, patients who
141
relapse fewer than six months after their last antineoplastic treatment are considered to have refractory
142
CLL/SLL, and are known to have a poor prognosis.[9, 16]
143
Exposure Definition
The main prognostic factor of interest was time-until-retreatment, defined as the difference in
SC
144
RI PT
138
years between the end of the first course of antineoplastic therapy and the beginning of the second
146
course of antineoplastic therapy. Receipt of antineoplastic treatment was assessed using procedure,
147
diagnosis and drug codes recorded in Medicare inpatient, outpatient, and prescription claims. See
148
Appendix Tables A1 and A2 for details on the codes and code ranges that were used. Corticosteroids
149
were not included in our study definition of antineoplastic therapy because they are frequently used for
150
other indications, and rarely used as monotherapy for CLL/SLL. Drug-specific HCPCS and National Drug
151
Codes (NDCs) were used to identify specific antineoplastic agents administered. Patients were classified
152
as having received chemo-immunotherapy (CIT), chemotherapy, or immunotherapy. More details on the
153
specific treatment regimens that patients received are provided in Appendix Table A3.
TE D
EP
154
M AN U
145
A course of treatment was defined as beginning on the start date associated with the first claim where antineoplastic therapy was recorded, continuing for as long as additional claims were observed
156
(allowing for gaps of no more than 90 days), and ending on the end date of the last claim in the series. A
157
prior chart validation study performed in lymphoma patients found that Medicare claims data generally
158
provide reliable information on whether and when patients receive chemotherapy.[17]
159
Study Endpoint
AC C
155
160
The study endpoint was all-cause mortality. The SEER-Medicare dataset includes each patient’s
161
date of death as recorded in the Social Security Death Master File. In the files used in the present study
ACCEPTED MANUSCRIPT 8 162
from the 2014 SEER-Medicare linkage, mortality data were available through the end of 2013. Patients
163
who were alive on December 31, 2013, were right-censored in our survival analyses.
164
Statistical Methods Five-year relative survival (RS5) was used to characterize CLL/SLL patients’ prognosis. RS5—the
RI PT
165
ratio of five-year observed overall survival relative to overall survival in the general population after
167
conditioning on age, sex and calendar year—was estimated with the Ederer II method.[18-20] RS
168
measures are frequently used in cancer epidemiology as indirect estimates of the burden of cancer-
169
specific mortality in defined patient populations. The statistical significance of differences and trends in
170
RS across patient subgroups was assessed using the additive hazards model endorsed by Dickman et al.
171
for analyses of relative survival data within a generalized linear models framework.[19]
172
Covariates
M AN U
173
SC
166
A number of demographic and clinical characteristics were assessed in order to characterize our study sample (see Table I), and to explore possible modification of the relationship between time-to-
175
retreatment and relative survival. As a summary measure of comorbidity burden, we report a count of
176
the following 12 major comorbidities included in the National Cancer Institute and Charlson comorbidity
177
indices:[21-23] cerebrovascular disease, chronic pulmonary disease, congestive heart failure, dementia,
178
diabetes, hemiplegia/paraplegia, liver disease, myocardial infarction, peptic ulcer disease, renal disease,
179
rheumatic disease, and human immunodeficiency virus (HIV) infection. In addition, we provide data on
180
the frequency of several health conditions associated with advanced CLL/SLL disease, e.g., anemia and
181
infection.
EP
AC C
182
TE D
174
Health conditions and healthcare utilization were assessed based on administrative diagnosis
183
codes recorded during the year prior to retreatment. Following the approach of Klabunde et al.,[23] a
184
condition was considered present if a corresponding inpatient diagnosis code or two outpatient
185
diagnosis codes 30+ days apart were observed. A higher standard was required for outpatient diagnosis
ACCEPTED MANUSCRIPT 9 186
codes, because they can sometimes reflect diagnoses that were ruled out or merely considered as part
187
of a differential diagnosis.
188
Validation Study in Mayo Clinic CLL Database Data for the validation cohort were obtained from the Mayo Clinic CLL Database with the
RI PT
189
approval of the Mayo Clinic IRB. The database includes patients with a diagnosis of CLL/SLL seen in the
191
Division of Hematology at the Mayo Clinic since 1995. Clinical information regarding date of diagnosis,
192
baseline evaluation, prognostic parameters, treatment history, and disease-related complications was
193
abstracted from clinical records on all patients and maintained in the database on an ongoing
194
prospective basis. Additional details about the Mayo Clinic CLL Database may be found in prior
195
publications.[24, 25]
M AN U
SC
190
The validation study cohort included patients diagnosed with CLL/SLL in 1995-2013, and
197
longitudinal follow-up data through 2016. All CLL/SLL patients who received second-line therapy during
198
this period were identified, classified based on the time-to-retreatment, and followed until death or
199
censoring. Five-year relative survival (RS5) from the initiation of second-line therapy was estimated with
200
the relsurv package for R.[26, 27] RS5 was estimated for patient subgroups defined by time-to-
201
retreatment, type of first-line antineoplastic therapy received (CIT, chemotherapy or immunotherapy),
202
IGHV mutation status, and whether 11q or 17p chromosome deletions were detected through
203
fluorescence in situ hybridization (FISH).
204
Results
EP
AC C
205
TE D
196
During 1992-2011, there were 6044 CLL/SLL cases identified from SEER-Medicare who initiated
206
first-line therapy and were eligible for continued follow-up. Of this group, 1974 initiated second-line
207
therapy prior to censoring and were included in our analyses (Figure 1). As first-line treatment, 693
208
patients had received CIT (35%), 824 chemotherapy alone (42%) and 457 immunotherapy alone (23%).
209
Frequencies of specific treatment regimens are provided in Appendix Table A3. At retreatment, the
ACCEPTED MANUSCRIPT 10 210
patients had a median age of 77 years (interquartile range: 73, 82); 44% were female. Additional patient
211
characteristics are shown in Table I. Five-year relative survival (RS5) following second-line treatment was modestly higher in patients
213
retreated ≥3 years after first-line treatment (0.49; 95% CI: 0.40, 0.58) compared to patients retreated in
214
6 months to <2 years (0.42, 95% CI: 0.38, 0.45) or 2 to <3 years (0.42, 95% CI: 0.35, 0.50; test for trend: p
215
= 0.01). The relationship between time-until-retreatment and subsequent RS5 varied by the type of first-
216
line treatment received (Figure 2a). A protective association between longer time-until-retreatment and
217
RS5 was seen among patients who received chemotherapy alone (p = 0.004) or CIT (p = 0.054). No clear
218
pattern was observed in patients who received immunotherapy alone (p = 0.71) as first-line treatment.
SC
M AN U
219
RI PT
212
In additional subgroup analyses, we explored whether the association between time-untilretreatment and subsequent RS5 varied by number of treatment cycles received during first-line
221
therapy, patient age, or comorbidity burden. The relationship between longer time-until-retreatment
222
and improved subsequent RS5 was most pronounced among those whose first-line treatment course
223
included 3+ treatment cycles (Appendix Figure A1), who were age 80+ years or older (Appendix Figure
224
A2), and who had no major comorbidities (Appendix Figure A3). However, across all subgroups of
225
retreated patients, survival was meaningfully reduced compared to appropriate age and gender
226
matched cohort from the U.S. general population.
227
Mayo Clinic Validation Study
EP
AC C
228
TE D
220
The validation cohort consisted of 508 patients identified from the Mayo Clinic CLL Database
229
who received second-line antineoplastic therapy. The median patient age was 66 years (interquartile
230
range: 59, 74); 28% were female; 58%, 29%, and 12% had received CIT, chemotherapy alone, and
231
immunotherapy alone as first-line treatment, respectively (see Table II). RS5 estimates were 0.52, 0.60,
232
and 0.70 for patients retreated in 6 months to <2 years, 2 to <3 years, and 3+ years, respectively (test for
233
trend: p = 0.09). Similar patterns were observed when patients were stratified by type of first-line
ACCEPTED MANUSCRIPT 11 antineoplastic therapy received (Figure 2a). In subgroup analyses, we stratified patients by IGHV
235
mutation and FISH status. Time-to-retreatment was more strongly predictive of survival in patients with
236
favorable FISH status (i.e., no 11q or 17p deletions; Figure 3a) and in patients with unmutated IGHV
237
(Figure 3b), though these trends did not reach statistical significance.
238
Discussion
RI PT
234
In our primary population-based SEER-Medicare study of mortality among older patients with
240
relapsed CLL/SLL who received second-line therapy, we found that longer time-until-retreatment was
241
associated with moderately improved subsequent RS5. Contrary to our initial study hypothesis, and in
242
contradistinction to previous findings in follicular lymphoma – another chronic B-cell malignancy—
243
stratification on time-to-retreatment did not allow us to identify a patient subgroup with normal
244
longevity. Even among patients retreated 3+ years after the end of first-line therapy, mortality was high
245
among the SEER-Medicare patients relative to the general population (RS5 ≈ 0.50). These results
246
indicate that a significant number of older patients with relapsed CLL/SLL retreated with rituximab-era
247
CIT regimens are likely to have an unmet therapeutic need.
M AN U
TE D
248
SC
239
Overall, similar patterns were observed among patients in the Mayo Clinic CLL Database. Prognosis was better across the board for retreated patients in the younger Mayo cohort, with RS5
250
estimates ranging from 0.52 for patients retreated in 6 months to <2 years to 0.70 for patients retreated
251
in 3+ years. It is possible that the better prognosis observed in the Mayo cohort relative to the SEER-
252
Medicare cohort is attributable to its younger age (median age of 66 years versus 77 years in SEER-
253
Medicare cohort) and other case-mix differences, as well as differences in the selection of patients for
254
treatment.
255
AC C
EP
249
Because clinical prognostic markers were available in the Mayo Clinic CLL Database, we were
256
able to stratify the Mayo cohort by IGHV and FISH status. These subgroup analyses suggested that time-
257
to-retreatment had greater prognostic value among patients with unmutated IGHV and favorable FISH.
ACCEPTED MANUSCRIPT 12 258
Notably, among patients with time-to-retreatment ≥3 years and favorable FISH, subsequent survival was
259
close to normal (RS5 = 0.86). Survival was also close to normal for patients with time-to-retreatment ≥3
260
years who had received immunotherapy as first-line treatment (RS5 = 0.90). Our results differ from those reported in prior work in follicular lymphoma (FL), another typically
RI PT
261
chronic and indolent lymphoproliferative disease. In FL, it was demonstrated that early events following
263
initial therapy are strongly predictive of subsequent outcomes, independent of baseline clinical
264
prognostic scores.[28] In FL patients who received CIT as first-line treatment, disease progression within
265
24 months is predictive of meaningfully reduced overall survival. However, in FL patients without an
266
event in the 24 months following CIT (or 12 months following less aggressive treatment), overall survival
267
is comparable to expected rates based on the age and sex matched general population. Our differing
268
results may be explained by differences in the biology of FL and CLL/SLL, and in the typical trajectory of
269
disease progression following first-line treatment. In addition, the research questions in the FL study and
270
ours were slightly different: we focused on survival outcomes among patients who initiated second-line
271
treatment, whereas the FL study sought to characterize subsequent survival trajectories among patients
272
who at 24 months had not been retreated or experienced a relapse.
M AN U
TE D
Our primary study relied on data available from SEER registries and Medicare claims, and thus
EP
273
SC
262
had several limitations. Treatment dates and the type of antineoplastic therapy received were
275
determined from procedure and diagnosis codes recorded by healthcare providers for reimbursement
276
and Part D pharmacy claims. A prior validation study of billing codes for antineoplastic treatment in
277
Medicare beneficiaries with lymphoma indicated that claims data generally provide valid information on
278
chemotherapeutic agents and services dates.[17] We would expect that modest misclassification of the
279
timing and type of treatment would mean that the observed differences in RS5 across patient subgroups
280
may underestimate the true between-group differences in survival outcomes. In addition, clinical
281
prognostic markers were unavailable for the SEER-Medicare cohort.
AC C
274
ACCEPTED MANUSCRIPT 13 282
Since our RS5 estimates reflect comparisons between observed and expected survival based on mortality rates observed in the U.S. general population, it is important to consider the possibility of
284
treatment selection bias. The patients in our SEER-Medicare and Mayo cohorts were all determined to
285
be fit enough for first-and second-line antineoplastic therapy. For these reasons, our sample is likely to
286
be slightly healthier than the general population in terms of comorbidity burden and functional status.
287
Despite this possible source of bias toward improved survival outcomes in the CLL/SLL patients who
288
received antineoplastic therapy, survival rates were still meaningfully reduced among the majority of
289
patients receiving second-line therapy.
SC
RI PT
283
Our findings have several implications. During the rituximab era, the majority of older patients
291
with relapsed and retreated CLL/SLL had meaningfully reduced survival and a clear unmet therapeutic
292
need. Possible exceptions, based on subgroup analyses in the younger Mayo cohort, were patients with
293
favorable FISH and time-to-retreatment ≥3 years, or immunotherapy as first-line treatment and time-to-
294
retreatment ≥3 years. Further research is needed to quantify the degree to which targeted therapies
295
provide meaningful improvements over CIT in long-term outcomes for older patients with relapsed
296
CLL/SLL.
TE D
EP AC C
297
M AN U
290
ACCEPTED MANUSCRIPT 14 Acknowledgements:
RI PT
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
M AN U
SC
The collection of cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute's Surveillance, Epidemiology and End Results Program under contract HHSN261201000140C awarded to the Cancer Prevention Institute of California, contract HHSN261201000035C awarded to the University of Southern California, and contract HHSN261201000034C awarded to the Public Health Institute; and the Centers for Disease Control and Prevention's National Program of Cancer Registries, under agreement # U58DP003862-01 awarded to the California Department of Public Health. The ideas and opinions expressed herein are those of the author(s) and endorsement by the State of California Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their Contractors and Subcontractors is not intended nor should be inferred. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
TE D
Funding:
EP
Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number P50CA097274, by the University of Iowa Holden Comprehensive Cancer Center (HCCC) Population Research Core, which is supported in part by P30 CA086862, and by a Cancer & Aging Pilot Project Award from the University of Iowa HCCC and Center on Aging.
AC C
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330
ACCEPTED MANUSCRIPT 15 Authorship and conflict of interest statements: EMA contributed to the research design, analysis and interpretation of the data, and manuscript writing. He is now employed in Johnson & Johnson’s Medical Device Epidemiology research division. The analyses for this paper were completed and the manuscript was drafted prior to the start of that role.
RI PT
TDS contributed to the research design, analysis and interpretation of the data, and manuscript revisions. He reports no disclosures.
MCL contributed to the analysis and interpretation of the data and manuscript revisions. She reports no disclosures.
SC
KBW contributed to the analysis and interpretation of the data and manuscript revisions. She reports no disclosures.
M AN U
BDM contributed to the analysis and interpretation of the data and manuscript revisions. He reports no disclosures. BKL contributed to the research design, analysis and interpretation of the data, and manuscript revisions. He has received compensation as a consultant for Gilead and AbbVie.
EP
TE D
EAC contributed to the research design and manuscript revisions. She reports no disclosures.
AC C
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354
ACCEPTED MANUSCRIPT 16 References
EP
TE D
M AN U
SC
RI PT
1. Byrd JC, Brown JR, O'Brien S, et al. Ibrutinib versus ofatumumab in previously treated chronic lymphoid leukemia. N Engl J Med 2014;371:213-223. 2. Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in relapsed chronic lymphocytic leukemia. N Engl J Med 2014;370:997-1007. 3. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with Venetoclax in Relapsed Chronic Lymphocytic Leukemia. N Engl J Med 2016;374:311-322. 4. Shanafelt TD, Borah BJ, Finnes HD, et al. Impact of ibrutinib and idelalisib on the pharmaceutical cost of treating chronic lymphocytic leukemia at the individual and societal levels. J Oncol Pract 2015;11:252-258. 5. Johnson LA. FDA approves drug for tough-to-treat type of leukemia. Associated Press: Financial News; 2016. 6. McMullen JR, Boey EJ, Ooi JY, et al. Ibrutinib increases the risk of atrial fibrillation, potentially through inhibition of cardiac PI3K-Akt signaling. Blood 2014;124:3829-3830. 7. Rai KR, Stilgenbauer S. Treatment of relapsed or refractory chronic lymphocytic leukemia. UpToDate. Waltham, MA: UpToDate; 2016. 8. Brown JR. The treatment of relapsed refractory chronic lymphocytic leukemia. Hematology Am Soc Hematol Educ Program 2011;2011:110-118. 9. Hallek M. Chronic lymphocytic leukemia: 2015 Update on diagnosis, risk stratification, and treatment. Am J Hematol 2015;90:446-460. 10. Tam CS, O'Brien S, Plunkett W, et al. Long-term results of first salvage treatment in CLL patients treated initially with FCR (fludarabine, cyclophosphamide, rituximab). Blood 2014;124:3059-3064. 11. SEER Stat Fact Sheets: Chronic Lymphocytic Leukemia (CLL). National Cancer Institute: Surveillance, Epidemiology, and End Results Program; 2015. 12. Warren JL, Klabunde CN, Schrag D, et al. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Medical care 2002;40:IV-3-18. 13. Human Mortality Database. University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany); 2016. 14. U.S. Census Bureau. Population Estimates and Projection, Annual Estimates of the Resident Population by Single Year of Age and Sex for the United States. Washington, D.C.; , various years. 15. National Center for Health Statistics. Vital Statistics of the United States, Volume II: Mortality, Part A. Washington, D.C.: Government Printing Office; various years. 16. Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008;111:5446-5456. 17. Chen-Hardee S, Chrischilles EA, Voelker MD, et al. Population-based assessment of hospitalizations for neutropenia from chemotherapy in older adults with non-Hodgkin's lymphoma (United States). Cancer Causes Control 2006;17:647-654. 18. Cho H, Howlader N, Mariotto AB, et al. Technical Report #2011-01: Estimating relative survival for cancer patients from the SEER Program using expected rates based on Ederer I versus Ederer II method. National Cancer Institute Surveillance Research Program; 2011. 19. Dickman PW, Sloggett A, Hills M, et al. Regression models for relative survival. Stat Med 2004;23:51-64. 20. Ederer F, Heise H. Methodological note No. 10: Instructions to IBM 650 programmers in processing survival computations. Bethesda, MD: National Cancer Institute End Results Evaluation Section; 1959.
AC C
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
ACCEPTED MANUSCRIPT 17
EP
TE D
M AN U
SC
RI PT
21. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-383. 22. Klabunde CN, Legler JM, Warren JL, et al. A refined comorbidity measurement algorithm for claims-based studies of breast, prostate, colorectal, and lung cancer patients. Ann Epidemiol 2007;17:584-590. 23. Klabunde CN, Potosky AL, Legler JM, et al. Development of a comorbidity index using physician claims data. J Clin Epidemiol 2000;53:1258-1267. 24. Thurmes P, Call T, Slager S, et al. Comorbid conditions and survival in unselected, newly diagnosed patients with chronic lymphocytic leukemia. Leuk Lymphoma 2008;49:49-56. 25. Shanafelt TD, Jenkins G, Call TG, et al. Validation of a new prognostic index for patients with chronic lymphocytic leukemia. Cancer 2009;115:363-372. 26. Pohar M, Stare J. Relative survival analysis in R. Comput Methods Programs Biomed 2006;81:272-278. 27. Pohar M, Stare J. Making relative survival analysis relatively easy. Comput Biol Med 2007;37:1741-1749. 28. Casulo C, Byrtek M, Dawson KL, et al. Early Relapse of Follicular Lymphoma After Rituximab Plus Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone Defines Patients at High Risk for Death: An Analysis From the National LymphoCare Study. J Clin Oncol 2015;33:2516-2522.
AC C
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
ACCEPTED MANUSCRIPT 18
EP
TE D
M AN U
SC
RI PT
Figure 1. Flow diagram showing identification of eligible SEER-Medicare patients with chronic lymphocytic leukemia or small lymphocytic lymphoma (CLL/SLL).
AC C
422 423
ACCEPTED MANUSCRIPT 19
EP
TE D
M AN U
SC
RI PT
Figure 2. Estimated five-year relative survival following initiation of second-line treatment in the SEERMedicare CLL/SLL patient cohort and Mayo validation cohort.
AC C
424 425 426 427
ACCEPTED MANUSCRIPT 20
EP
TE D
M AN U
SC
RI PT
Figure 3. Estimated five-year relative survival following initiation of second-line treatment in Mayo validation cohort, stratified by FISH and IGHV mutation status
AC C
428 429
ACCEPTED MANUSCRIPT 21 Table I. Characteristics of eligible patients who received second-line treatment stratified by first-line therapy. ChemoChemoImmunoAll immunotherapy therapy Characteristic treatments therapy alone alone (N=1974) (N=693) (N=824) (N=457) Age at second-line treatment in years 653 (33%) 277 (40%) 268 (33%) 108 (24%) • 66-74 577 (29%) 202 (29%) 261 (32%) 114 (25%) • 75-79 744 (38%) 214 (31%) 295 (36%) 235 (51%) • 80+ Sex 874 (44%) 266 (38%) 388 (47%) 220 (48%) • Female 1100 (56%) 427 (62%) 436 (53%) 237 (52%) • Male Number of treatment cycles included in first-line treatment course 650 (33%) 109 (16%) 231 (28%) 310 (68%) • 1-2 1324 (67%) 584 (84%) 593 (72%) 147 (32%) • 3+ Interval between first- and secondline antineoplastic treatments 1325 (67%) 434 (63%) 531 (64%) 360 (79%) • 6 months to <2 years 355 (18%) 155 (22%) 142 (17%) 58 (13%) • 2 years to <3 years 294 (15%) 104 (15%) 151 (18%) 39 (9%) • 3+ years Type of second-line antineoplastic treatment 576 (29%) 298 (43%) 183 (22%) 95 (21%) • Chemo-immunotherapy 486 (25%) 80 (12%) 375 (46%) 31 (7%) • Chemotherapy alone 609 (31%) 207 (30%) 122 (15%) 280 (61%) • Immunotherapy alone 303 (15%) 108 (16%) 144 (17%) 51 (11%) • Unknown (non-specific billing codes) Indicators of CLL/SLL progression 916 (46%) 331 (48%) 345 (42%) 240 (53%) • Anemia 358 (18%) 145 (21%) 135 (16%) 78 (17%) • Thrombocytopenia 282 (14%) 147 (21%) 93 (11%) 42 (9%) • Immune deficiency Count of major comorbidities* 964 (49%) 330 (48%) 420 (51%) 214 (47%) • None 566 (29%) 197 (28%) 236 (29%) 133 (29%) • One 444 (22%) 166 (24%) 168 (20%) 110 (24%) • Two or more Hospitalized in prior year 777 (39%) 269 (39%) 335 (41%) 173 (38%)
432 433 434 435 436 437
*The major comorbidities covariate, adapted from the National Cancer Institute and Charlson comorbidity indices, is a count of the following 12 comorbidities: cerebrovascular disease, chronic pulmonary disease, congestive heart failure, dementia, diabetes, hemiplegia/paraplegia, liver disease, myocardial infarction, peptic ulcer disease, renal disease, rheumatic disease, and HIV infection.
AC C
EP
TE D
M AN U
SC
RI PT
430 431
ACCEPTED MANUSCRIPT 22 Table II. Baseline characteristics of eligible patients in Mayo validation cohort who received secondline treatment. Characteristic Patient count (N = 508) Age at second-line treatment in years 245 (48%) • ≤65 148 (29%) • 66-74 77 (15%) • 75-79 38 (7%) • 80+ Sex – N (%) 144 (28%) • Female 364 (72%) • Male Type of first-line treatment received – N (%) 295 (58%) • Chemo-immunotherapy (CIT) 148 (29%) • Chemotherapy alone 59 (12%) • Immunotherapy alone 6 (1%) • Missing Interval between end of first-line therapy and start of second-line therapy – N (%) 275 (54%) • 6 months to <2 years 105 (21%) • 2 years to <3 years 128 (25%) • 3+ years FISH status – N (%) 52 (10%) • High risk (11q- or 17p-) 187 (37%) • Other/favorable 269 (53%) • Missing IGHV mutation status – N (%) 217 (43%) • Unmutated 84 (17%) • Mutated 207 (41%) • Missing
EP AC C
440 441 442
TE D
M AN U
SC
RI PT
438 439
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
EP
TE D
M AN U
SC
RI PT
Appendix Table A1: Administrative diagnosis and procedure codes used to identify receipt of chemotherapy and immunotherapy. Valid interval Valid interval Definition Code Type Code start date end date alemtuzumab CPT/HCPCS J9010 01/01/2003 . bendamustine CPT/HCPCS J9033 01/01/2009 . chemotherapy DRG 410 . . chemotherapy CITD-9 V5811 . . DIAGNOSIS immunotherapy CITD-9 V5812 . . DIAGNOSIS chemotherapy CITD-9 9925 . . PROCEDURE chemotherapy* CPT/HCPCS 964xx . . chemotherapy CPT/HCPCS 9650x . . chemotherapy CPT/HCPCS 9651x . . chemotherapy CPT/HCPCS 9652x . . chemotherapy CPT/HCPCS 9653x . . chemotherapy CPT/HCPCS 9654x . . chemotherapy CPT/HCPCS G0355 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0356 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0357 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0358 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0359 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0360 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0361 01/01/2005 12/31/2005 chemotherapy CPT/HCPCS G0362 01/01/2005 12/31/2005 chemotherapy* CPT/HCPCS J9xxx . . chemotherapy CPT/HCPCS Q0083 . . chemotherapy CPT/HCPCS Q0084 . . chemotherapy CPT/HCPCS Q0085 . . chemotherapy Revenue 0331 . . chemotherapy Revenue 0332 . . chemotherapy Revenue 0335 . . chlorambucil CPT/HCPCS S0172 01/01/2002 . cladribine CPT/HCPCS J9065 01/01/1995 . cyclophosphamide CPT/HCPCS J8530 01/01/1995 . cyclophosphamide CPT/HCPCS J9070 01/01/1984 . cyclophosphamide CPT/HCPCS J9080 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9090 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9091 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9092 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9093 01/01/1988 12/31/2010 cyclophosphamide CPT/HCPCS J9094 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9095 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9096 01/01/1994 12/31/2010 cyclophosphamide CPT/HCPCS J9097 01/01/1994 12/31/2010
AC C
1 2
1
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
2
TE D
M AN U
SC
RI PT
Valid interval Valid interval Definition Code Type Code start date end date doxorubicin CPT/HCPCS J9000 01/01/1984 . doxorubicin CPT/HCPCS J9001 01/01/2000 12/31/2012 doxorubicin CPT/HCPCS J9002 01/01/2013 12/31/2013 doxorubicin CPT/HCPCS Q2048 07/01/2012 12/31/2012 doxorubicin CPT/HCPCS Q2049 07/01/2012 . doxorubicin CPT/HCPCS Q2050 07/01/2013 . fludarabine CPT/HCPCS C9262 04/01/2010 06/30/2010 fludarabine CPT/HCPCS J8562 01/01/2011 . fludarabine CPT/HCPCS J9185 01/01/1994 . fludarabine CPT/HCPCS Q2025 07/01/2010 12/31/2010 mitoxantrone CPT/HCPCS J9293 01/01/1990 . obinutuzumab CPT/HCPCS J9301 . . ofatumumab CPT/HCPCS C9260 04/01/2010 12/31/2010 ofatumumab CPT/HCPCS J9302 01/01/2011 . pentostatin CPT/HCPCS J9268 01/01/1994 . rituximab CPT/HCPCS J9310 01/01/1999 . vincristine CPT/HCPCS J9370 01/01/1986 . vincristine CPT/HCPCS J9371 01/01/2014 . vincristine CPT/HCPCS J9375 01/01/1988 12/31/2010 vincristine CPT/HCPCS J9380 01/01/1994 12/31/2010 * A subset of codes in the CPT/HCPCS ranges 964xx and J9xxx were excluded as being unlikely to represent antineoplastic therapy for CLL/SLL. These codes are listed in Appendix Table A2.
AC C
3 4 5 6
EP
In addition to the procedure and diagnosis codes listed above, National Drug Codes (NDCs) obtained from the 2015 Cerner Multum Lexicon were used to define receipt of the following drugs: alemtuzumab, bendamustine, chlorambucil, cladribine, cyclophosphamide, doxorubicin, fludarabine, ibrutinib, idelalisib, lenalidomide, mitoxantrone, obinutuzumab, ofatumumab, pentostatin, rituximab, vincristine.
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
EP
TE D
M AN U
SC
RI PT
Appendix Table A2: Specific chemotherapy, immunotherapy and related billing codes excluded as unlikely to represent treatment for CLL/SLL. Definition Code Type Code Percentage of all chemo-related codes observed in cohort Aldesleukin injection CPT/HCPCS J9015 0.00 Arsenic trioxide injection CPT/HCPCS J9017 0.02 Asparaginase, nos CPT/HCPCS J9020 0.00 Azacitidine injection CPT/HCPCS J9025 0.21 Bcg live intravesical vac CPT/HCPCS J9031 0.10 Bevacizumab injection CPT/HCPCS J9035 0.32 Bleomycin sulfate injection CPT/HCPCS J9040 0.06 Carmustine injection CPT/HCPCS J9050 0.00 Cetuximab injection CPT/HCPCS J9055 0.04 CHEMO HORMON ANTINEOPL SQ/IM CPT/HCPCS 96402 0.19 chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9240 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9295 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9376 0.00 range but definition unavailable) chemotherapy (in SEER-Medicare chemo CPT/HCPCS J9381 0.00 range but definition unavailable) CHEMOTHERAPY INTO CNS CPT/HCPCS 96450 0.04 Clofarabine injection CPT/HCPCS J9027 0.00 DACARBAZINE 100 MG CPT/HCPCS J9130 0.03 Dacarbazine 200 mg inj CPT/HCPCS J9140 0.02 Dactinomycin injection CPT/HCPCS J9120 0.00 Daunorubicin injection CPT/HCPCS J9150 0.00 Degarelix injection CPT/HCPCS J9155 0.00 Denileukin diftitox inj CPT/HCPCS J9160 0.00 Diethylstilbestrol injection CPT/HCPCS J9165 0.00 Docetaxel injection CPT/HCPCS J9171 0.03 Docetaxel injection CPT/HCPCS J9170 0.05 Elliotts b solution per ml CPT/HCPCS J9175 0.00 Epirubicin hydrochloride, 50 mg CPT/HCPCS J9180 0.00 Floxuridine injection CPT/HCPCS J9200 0.00 Fluorouracil injection CPT/HCPCS J9190 0.23 Gemcitabine hcl injection CPT/HCPCS J9201 0.19 Gemtuzumab ozogamicin inj CPT/HCPCS J9300 0.00 Goserelin acetate implant CPT/HCPCS J9202 0.10 HISTRELIN IMPLANT VANTAS 50 MG CPT/HCPCS J9225 0.00 Inj melphalan hydrochl 50 mg CPT/HCPCS J9245 0.00 INJ PEGASPARGASE SINGLE DOSE V CPT/HCPCS J9266 0.00 Inj, epirubicin hcl, 2 mg CPT/HCPCS J9178 0.00 INJECTION IDARUBCITIN HCL 5 MG CPT/HCPCS J9211 0.00 INJECTION IRINOTECAN 20 MG CPT/HCPCS J9206 0.03
AC C
7 8
3
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS CPT/HCPCS
TE D
EP
AC C
9 10 11
Percentage of all chemo-related codes observed in cohort J9209 0.04 J9280 0.02 J9263 0.05 J9265 0.19 J9350 0.01 J9360 0.07 J9395 0.00 J9213 0.05 J9214 0.17 J9212 0.02 J9228 0.00 96523 1.29 J9217 0.21 J9219 0.00 J9218 0.00 J9230 0.00 J9290 0.00 J9291 0.01 J9264 0.02 J9303 0.00 J9305 0.03 J9270 0.00 96521 0.07 96522 0.07 J9315 0.00 J9320 0.00 J9330 0.01 J9340 0.00 J9351 0.00 J9355 0.03 J9357 0.00 J9390 0.02
RI PT
INJECTION MESNA 200 MG INJECTION MITOMYCIN 5 MG INJECTION OXALIPLATIN 0.5 MG INJECTION PACLITAXEL 30 MG INJECTION TOPOTECAN 4 MG INJECTION VINBLASTINE SULFATE Injection, fulvestrant Interferon alfa-2a inj Interferon alfa-2b inj Interferon alfacon-1 inj Ipilimumab injection IRRIG DRUG DELIVERY DEVCITE LEUPROLIDE ACETATE 7.5 MG Leuprolide acetate implant Leuprolide acetate injeciton Mechlorethamine hcl inj MITOMYCIN 20 MG Mitomycin 40 mg inj Paclitaxel protein bound Panitumumab injection Pemetrexed injection Plicamycin (mithramycin) inj REFILL/MAINT PORTABLE PUMP REFILL/MAINT PUMP/RESVR SYST Romidepsin injection Streptozocin injection Temsirolimus injection Thiotepa injection Topotecan injection Trastuzumab injection Valrubicin injection Vinorelbine tartrate inj
Code
SC
Code Type
M AN U
Definition
4
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement Appendix Table A3. Categorization of patients’ first-line treatment regimens. Type of antineoplastic treatment regimen
Specific regimen as determined from drug-specific HCPCS and NDC codes
Frequency Percent
Chemoimmunotherapy (N = 693)
Fludarabine, cyclophosphamide and immunotherapy
179
Fludarabine and immunotherapy (Pentostatin or cladribine) and immunotherapy Bendamustine and immunotherapy
254
12.87
21
1.06
50
2.53
48
2.43
SC
Cyclophosphamide, vincristine, doxorubicin and immunotherapy
None of the above. Immunotherapy and at least one of the following: chlorambucil, cyclophosphamide, vincristine, doxorubicin or mitoxantrone
39
1.97
Fludarabine and cyclophosphamide
64
3.24
Fludarabine, pentostatin or cladribine*
427
21.64
Bendamustine
16
0.81
Cyclophosphamide, vincristine & doxorubicin
37
1.87
Cyclophosphamide & vincristine
137
6.94
None of the above. At least one of the following: chlorambucil, cyclophosphamide, vincristine, doxorubicin or mitoxantrone
143
7.25
Rituximab, ofatumumab or alemtuzumab*
457
23.15
M AN U
5.17
* Cells with fewer than 11 patients combined due to SEER-Medicare reporting requirements. Of the 427 patients who received fludarabine, pentostatin, or cladribine, fewer than 11 received pentostatin or cladribine. Of the 457 patients who received rituximab, ofatumumab or alemtuzumab, fewer than 11 patients received ofatumumab or alemtuzumab.
AC C
13 14 15 16 17 18
102
EP
Immunotherapy alone (N = 457)
Cyclophosphamide, vincristine and immunotherapy
TE D
Chemotherapy alone (N = 824)
9.07
RI PT
12
5
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
6
Appendix Figure A1. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by number of treatment cycles received during first-line treatment and the time interval between first- and second-line treatments.
Trend: p = 0.67
AC C
EP
TE D
M AN U
SC
RI PT
Trend: p < 0.001
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
7
Appendix Figure A2. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by patient age and time interval between first- and second-line treatments.
Trend: p = 0.004
RI PT
Trend: p = 0.099
AC C
EP
TE D
M AN U
SC
Trend: p = 0.107
ACCEPTED MANUSCRIPT Time to retreatment and survival in CLL/SLL: Data Supplement
8
Appendix Figure A3. SEER-Medicare cohort: Estimated five-year relative survival from the initiation of second-line treatment, stratified by major comorbidity count and time interval between first- and second-line treatments. Trend: p = 0.001 Trend: p = 0.555
AC C
EP
TE D
M AN U
SC
RI PT
Trend: p = 0.262
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT