- Email: [email protected]
Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell Transplantation for Multiple Myeloma
Accepted Manuscript Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell Transplantation for Multiple Myeloma Sanghee Hong, MD, Lisa Rybicki, MS, Donna Abounader, Brian J. Bolwell, MD, Robert Dean, MD, Aaron T. Gerds, MD, MS, Betty K. Hamilton, MD, Brian T. Hill, MD, PhD, Deepa Jagadeesh, MD, MPH, Matt Kalaycio, MD, Hien D. Liu, MD, Brad Pohlman, MD, Ronald Sobecks, MD, Navneet S. Majhail, MD, MS PII:
S1083-8791(16)00160-9
DOI:
10.1016/j.bbmt.2016.03.011
Reference:
YBBMT 54224
To appear in:
Biology of Blood and Marrow Transplantation
Received Date: 9 February 2016 Accepted Date: 9 March 2016
Please cite this article as: Hong S, Rybicki L, Abounader D, Bolwell BJ, Dean R, Gerds AT, Hamilton BK, Hill BT, Jagadeesh D, Kalaycio M, Liu HD, Pohlman B, Sobecks R, Majhail NS, Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell Transplantation for Multiple Myeloma, Biology of Blood and Marrow Transplantation (2016), doi: 10.1016/j.bbmt.2016.03.011. 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
Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell
1
RI PT
Transplantation for Multiple Myeloma
Sanghee Hong, MD, 2Lisa Rybicki, MS, 3Donna Abounader, 3,4Brian J Bolwell, MD, 3,4Robert
Dean, MD, Aaron T Gerds, MD, MS, 3,4Betty K Hamilton, MD, 3,4Brian T Hill, MD, PhD,
Deepa Jagadeesh, MD, MPH, 3,4Matt Kalaycio, MD, 3,4Hien D Liu, MD 3,4Brad Pohlman, MD,
3,4
Ronald Sobecks, MD, 3,4Navneet S Majhail, MD, MS
M AN U
1
SC
3,4
Department of Internal Medicine, Cleveland Clinic, Cleveland, OH; 2Department of
Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH; 3Blood & Marrow Transplant Program, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH; 4Department of
Corresponding Author:
Cleveland Clinic
EP
Navneet Majhail, MD, MS
TE D
Hematology and Oncology, Cleveland Clinic, Cleveland, OH
AC C
9500 Euclid Ave, R35 Cleveland, OH 44195 Phone: 216-444-2199 Fax: 216-444-9464
Email: [email protected]
Short Title: SES and myeloma AHCT outcomes 1
ACCEPTED MANUSCRIPT
Financial Disclosure: None of the authors has a financial conflict of interest to disclose in relation to this study. Word Count: Abstract – 182 words, Text – 1493 words, References – 16, Tables – 2, Figures –
RI PT
1
Key Words: Autologous stem cell transplantation, multiple myeloma, socioeconomic status,
AC C
EP
TE D
M AN U
SC
healthcare disparities, ethnicity, income
2
ACCEPTED MANUSCRIPT
ABSTRACT: Autologous hematopoietic cell transplantation (AHCT) is standard therapy for eligible patients with multiple myeloma. Healthcare disparities can influence transplantation outcomes. However,
RI PT
the association of socioeconomic status (SES), a major indicator of healthcare disparities, with outcomes in patients with myeloma after AHCT has not been previously described. We analyzed 346 consecutive AHCT recipients with myeloma transplanted between 2003 and 2013 in this
SC
retrospective cohort study. ZIP code of residence at the time of AHCT was obtained to assess annual household income based on 2010 US Census data (median $49,054; range $16,546-
M AN U
127,313). SES groups were divided into <$45,000 (low; N=120), $45-60,000 (middle; N=116), and >$60,000 (high; N=110). The low income cohort had smallest portion of Caucasians (69% vs. 89% vs. 91%), otherwise patient, disease, and transplant characteristics were comparable or different without significant patterns found among cohorts. Median follow-up was 49 months.
TE D
There was no difference among SES groups in overall survival, progression-free survival, nonrelapse mortality or relapse in univariate and multivariable analysis. Similarly, SES was not associated with survival in a subset analysis of 303 patients who had survived for 1 year post-
AC C
EP
transplant.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Healthcare disparities have been described in the epidemiology and outcomes of multiple myeloma. African-Amercians have a two-fold higher incidence of multiple myeloma (MM)
RI PT
compared to Caucasians.1-3 However, their survival is similar or even superior compared with Caucasian myeloma patients.4-6 Race/ethnicity was not found to be associated with access to transplantation or outcomes after autologous hematopoietic cell transplantation (AHCT) for
SC
myeloma, although African-American patients were younger and at the later myeloma stage at the time of transplantation.4,7,8 The impact of socioeconomic status (SES) on myeloma outcomes
M AN U
is controversial.5,6,9-11 A recent retrospective study by Fiala et al. reviewed a single center and the Surveillance, Epidemiology, and End Results-18 database and found that low SES was an independent risk factor for poor overall survival (OS) in MM.12 This study also showed that low SES by median household income limits access to AHCT.12 However, the association of SES
TE D
with survival after AHCT for patients with myeloma has not been previously described. We investigated the association of SES on outcomes after AHCT for MM.
EP
METHODS Patient Population
AC C
We identified 354 consecutive adult patients with MM who received AHCT at our institution from 2003-2013. Data were obtained from our institutional BMT database, which prospectively captures clinical and outcomes data on all transplants. The study was approved by our institutional IRB. We excluded 8 patients who did not provide consent for using data for research or had a history of prior transplantation. Therefore, our analysis cohort consisted of 346 patients. We also performed a subset analysis on 303 patients who survived at least one year after AHCT,
4
ACCEPTED MANUSCRIPT
based on an observation that transition of care of allogeneic transplant recipients from transplant center to community was found to impact long-term transplant outcomes.13
RI PT
ZIP code of residence at the time of transplantation was used to obtain median annual household income based on 2010 US Census data14 and to calculate distance from our center.
American Society of Blood and Marrow Transplant (ASBMT) Request for Information criteria
SC
were used to assign disease risk (low risk MM includes first complete remission, very good
M AN U
partial remission or partial remission; all other disease status are considered as high risk).15
Statistical Analysis
The primary outcome was OS. Secondary endpoints included progression-free survival (PFS), non-relapse mortality (NRM) and relapse. Recursive partitioning analysis was used to evaluate
TE D
the best cutpoint in median household income to predict OS. No significant cutpoints were identified, and hence SES was grouped into approximate tertiles for the analysis.12 Annual income <$45,000, $45,000-60,000, and >$60,000 were categorized as low, middle, and high SES,
EP
respectively. In addition, median household income was analyzed as a continuous variable to
AC C
assess outcome trends.
Distance from transplant center was calculated by a driving time program and categorized as ≤30, 31-60, 61-90, 91-120, and >120 miles. International Staging System (ISS) score was summarized but not used in risk factor analysis due to missing data in 27% patients.
5
ACCEPTED MANUSCRIPT
Patient-, disease-, and transplant- related characteristics were compared among SES groups with Chi-square or Kruskal-Wallis test.
RI PT
Outcomes were estimated using Kaplan-Meier or cumulative incidence methods and compared among SES groups using log-rank or Gray test. Prognostic factors were identified with Cox or Fine and Gray regression. Automated stepwise variable selection was performed on 1000
SC
bootstrap samples; variables occurring in >50% of these 1000 models were considered important prognostic factors. SES and race/ethnicity were included in all multivariable models even if they
M AN U
were not significant, while all other variables appearing in the models were significant. Variables that were assessed as potential prognostic factors in multivariable analysis included distance of residence from transplant center, year of AHCT, gender, age at transplant, history of prior cancer, performance status, number of prior chemotherapy regimens, prior radiation therapy, time from
TE D
diagnosis to transplantation and disease risk at transplantation. Analyses were performed using SAS software (version 9.4) and all P-values are two-sided.
EP
RESULTS Patient Characteristics
AC C
Table 1 describes patient characteristics. There were differences in race/ethnicity and distance from the transplant center, otherwise demographic features did not significantly differ among the SES cohorts.
Granulocyte colony stimulating factor (G-CSF) with or without plerixafor was the most common regimen for mobilization and 18% of patients received chemotherapy mobilization. Melphalan
6
ACCEPTED MANUSCRIPT
(63%) was the most commonly used conditioning regimen followed by busulfan and cyclophosphamide (37%). There were no statistical differences among SES groups by
RI PT
conditioning regimen and cell doses given for both CD34+ cells and total nucleated cells.
Outcomes by SES
Median follow-up was 49 months (range, 3-129) and was comparable among three SES groups.
SC
There were no differences in hospital stay and neutrophil/platelet engraftment among SES
groups (all with p>0.05; results not shown). Relapse (74% overall) was the most common cause
M AN U
of death for all three groups. Incidence of 100-day mortality was 1%, and included 2 patients in middle SES and 3 in high SES categories. No patients from low SES died within 100 days post AHCT.
TE D
Table 2 shows 5-year estimates of OS, PFS, relapse, and NRM by SES, with OS through 10 years in Figure 1. We did not observe any significant differences in these endpoints based on recipient SES at AHCT in univariate analyses, orin multivariable analysis after adjusting for
EP
significant patient, disease, and transplant related variables. Compared to low SES recipients, the hazard ratios (HR) for OS were 1.40 (95% CI, 0.93-2.10, P=0.11) for middle SES and 1.08 (95%
AC C
CI, 0.71-1.64, P=0.72) for high SES recipients. For PFS, the corresponding HR’s were 1.04 (95% CI, 0.74-1.47, P=0.81) and 0.69 (0.48-1.00, P=0.05), respectively. The risk of relapse was also comparable (HR 1.20, P=0.32 for middle SES and 0.78, P=0.22 for high SES compared to low SES), as was the risk for NRM (HR 0.46, P=0.06 and 0.70, P=0.35, respectively). Race/ethnicity was not found to be prognostic for any outcome.
7
ACCEPTED MANUSCRIPT
Subset Analysis of 1-year Survivors In the subset analysis of 303 1-year survivors, patient characteristics were similar among the three SES groups (data not shown). Median follow-up was 51 (range, 12-129) months. Table 2
RI PT
shows 5-year outcome estimates, which were comparable among the three groups. Causes of deaths were similar and relapse (65-78%) was the most common cause of death for all SES
groups. In multivariable analyses, SES again was not found to be associated with OS, relapse, or
SC
NRM. There was no difference in PFS between low and middle SES groups, but high SES
patients had a significant improvement in PFS compared to low SES patients (HR 0.64, 95% CI,
M AN U
0.42-0.98, P=0.04). Race/ethnicity again was not found to be associated with any outcome.
DISCUSSION
We did not observe an association between SES, as determined by median household income
TE D
based on Zip Code of residence, and outcomes after AHCT in myeloma patients. Our study adds information to the growing body of literature on healthcare disparities in myeloma,5,6,9-11 and highlights that for patients who undergo AHCT, income disparities or socioeconomic status does
EP
not directly impact outcomes. We also did not find an association of race/ethnicity with any
AC C
outcome measure after AHCT, which agrees with current literature.4-6
Although our study indicates lack of association between SES and AHCT outcomes for myeloma, we were not able to evaluate the role of socioeconomic disparities in referral for AHCT. A recent retrospective study by Fiala et al. reviewed a single center and the Surveillance, Epidemiology, and End Results-18 database and found that low SES was an independent risk factor for poor overall survival (OS) in MM.12 This study also showed that low SES by median household
8
ACCEPTED MANUSCRIPT
income limits patients access to AHCT.12 Similarly, Al-Hamadani et al. found that a lower income was negatively associated with consideration for AHCT upfront.16
RI PT
There are some limitations of our study that have to be considered. First, our population size was relatively small. We were not able to account for health disparity measures other than SES and race/ethnicity (e.g., educational status), and SES was based on median household income based
SC
on ZIP code rather than actual household income. Our study was desgined to focus on patients who had received a transplant; we did not address access to transplantation for myeloma patients.
M AN U
Also, we did not have complete information on some disease specific characteristics, such as cytogenetics and myeloma stage. Given the low number of non-Caucasians in our study, we had to group all other race/ethniticies together. Additionally, maintenance therapy after AHCT may also impact outcomes especially given the financial aspects of using expensive medications long-
TE D
term. However, use and type of maintenance therapy were not captured in our study. Furthermore, quality of life is an important of outcome of AHCT which can be impacted by SES; however, it was not analyzed in this study. Overall, these limitations emphasize the importance
EP
of futures studies with larger populations addressing this issue.
AC C
In conclusion, we did not see an association between SES and outcomes of AHCT for myeloma. However, our findings need to be validated in a large multi-center or registry based cohorts. In the meantime, transplant providers should continue to be aware of special populations including the socioeconomically disadvantaged that are risk for healthcare disparities.
ACKNOWLEDGEMENT:
9
ACCEPTED MANUSCRIPT
This study was partly supported by American Society of Hematology HONORS (Hematology
RI PT
Opportunities for the Next Generation of Research Scientists) Award to Sanghee Hong.
REFERENCES:
1. Kosary CL, Altekruse SF, Ruhl J, Lee R, Dickie L. Clinical and prognostic factors for
M AN U
1 and version 2. Cancer. 2014;120 Suppl 23:3807-3814.
SC
melanoma of the skin using SEER registries: Collaborative stage data collection system, version
2. Smith GD, Wentworth D, Neaton JD, Stamler R, Stamler J. Socioeconomic differentials in mortality risk among men screened for the multiple risk factor intervention trial: II. black men. Am J Public Health. 1996;86(4):497-504.
TE D
3. Waxman AJ, Mink PJ, Devesa SS, et al. Racial disparities in incidence and outcome in multiple myeloma: A population-based study. Blood. 2010;116(25):5501-5506.
4. Hari PN, Majhail NS, Zhang MJ, et al. Race and outcomes of autologous hematopoietic cell
EP
transplantation for multiple myeloma. Biol Blood Marrow Transplant. 2010;16(3):395-402.
AC C
5. Kaya H, Peressini B, Jawed I, et al. Impact of age, race and decade of treatment on overall survival in a critical population analysis of 40,000 multiple myeloma patients. Int J Hematol. 2012;95(1):64-70.
6. Abou-Jawde RM, Baz R, Walker E, et al. The role of race, socioeconomic status, and distance traveled on the outcome of african-american patients with multiple myeloma. Haematologica. 2006;91(10):1410-1413.
10
ACCEPTED MANUSCRIPT
7. Costa LJ, Huang JX, Hari PN. Disparities in utilization of autologous hematopoietic cell transplantation for treatment of multiple myeloma. Biol Blood Marrow Transplant.
RI PT
2015;21(4):701-706.
8. Verma PS, Howard RS, Weiss BM. The impact of race on outcomes of autologous
transplantation in patients with multiple myeloma. Am J Hematol. 2008;83(5):355-358.
SC
9. Kristinsson SY, Derolf AR, Edgren G, Dickman PW, Bjorkholm M. Socioeconomic
differences in patient survival are increasing for acute myeloid leukemia and multiple myeloma
M AN U
in sweden. J Clin Oncol. 2009;27(12):2073-2080.
10. Pasqualetti P, Colantonio D, Collacciani A, Casale R. Socioeconomic status and survival in multiple myeloma. Minerva Med. 1990;81(10):713-716.
TE D
11. Savage D, Lindenbaum J, Van Ryzin J, Struening E, Garrett TJ. Race, poverty, and survival in multiple myeloma. Cancer. 1984;54(12):3085-3094.
EP
12. Fiala MA, Finney JD, Liu J, et al. Socioeconomic status is independently associated with overall survival in patients with multiple myeloma. Leuk Lymphoma. 2015:1-7.
AC C
13. Fu S, Rybicki L, Abounader D, et al. Association of socioeconomic status with long-term outcomes in 1-year survivors of allogeneic hematopoietic cell transplantation. Bone Marrow Transplant. 2015;50(10):1326-1330.
14. US Census Bureau. American FactFinder. http://factfinder.census.gov/faces/nav/jsf/pages/index.html. Updated 2010. Accessed October 15, 2015, 2015. 11
ACCEPTED MANUSCRIPT
15. American Society of Blood and Marrow Transplantation. ASBMT RFI 2015 - disease classifications corresponding to CIBMTR classification. https://www.asbmt.org/resource/resmgr/RFI/RFI_2015_-_CIBMTR_Disease_Cl.pdf. Updated
RI PT
2015. Accessed October, 15 2015, 2015.
16. Al-Hamadani M, Hashmi SK, Go RS. Use of autologous hematopoietic cell transplantation
AC C
EP
TE D
M AN U
era of novel agents. Am J Hematol. 2014;89(8):825-830.
SC
as initial therapy in multiple myeloma and the impact of socio-geo-demographic factors in the
12
ACCEPTED MANUSCRIPT
TABLES AND FIGURES: Table 1. Patient Characteristics High SES N (%) 110 $68,501 (60,763-127,313) 22 (1.3-410)
P-value
64 (53) 37 (31) 57 (22-70) 107/118 (91)
70 (60) 13 (11) 57 (36-74) 106/112 (95)
65 (59) 10 (9) 57 (35-76) 101/104 (97)
0.51 <0.001 0.78 0.12
1 (1-5)
1 (1-7)
1 (1-5)
0.57
10 (4-330)
0.19
EP
AC C
SC
RI PT
Middle SES N (%) 116 $49,421 (45,139-59,973) 64 (3.4-1768)
11 (4-115)
9 (2-272)
-
<0.001
0.44
86 (72) 34 (28)
88 (76) 28 (24)
75 (68) 35 (32)
31 (38) 27 (33) 24 (29) 38
33 (38) 30 (34) 25 (28) 28
36 (44) 26 (32) 19 (24) 29
71 (59) 26 (22) 23 (19)
48 (41) 51 (44) 17 (15)
52 (47) 36 (33) 22 (20)
71 (59) 49 (41)* 5 (2-22)
72 (62) 44 (38) 5 (2-15)
75 (68) 35 (32) 5 (2-14)
0.86
7 (6) 57 (3-121)
5 (4) 42 (4-122)
12 (11) 48 (9-129)
0.13 0.23
TE D
N Median household income, median (range) Distance from transplant center in miles, median (range) Male sex Non-Caucasians Age, median (range) Good performance status (ECOG ≤1 or KPS ≥90) Number of prior chemotherapy regimens, median (range) Time from diagnosis to AHCT in months, median (range) Disease risk Low High ISS stage I II III Missing Mobilizing regimen G-CSF only G-CSF + Plerixafor Chemotherapy mobilization Conditioning regimen Melphalan Busulfan+cyclophosphamide CD34+ cell dose x 106/kg, median (range) Tandem transplant Followup in months, median (range)
Low SES N (%) 120 $38,409 (16,546-44,809) 54 (0.7-1036)
M AN U
Variable
0.86
0.007
0.36
Abbreviations: SES – socioeconomic status; AHCT – autologous hematopoietic cell transplantation; ECOG – Eastern Cooperative Oncology Group; KPS – Karnofsky Performance Status; ISS – International Staging System; G-CSF – granulocyte- colony stimulating factor. * Includes 2 patients who received busulfan + cyclophosphamide + etoposide
13
ACCEPTED MANUSCRIPT
Table 2. 5-year outcomes after AHCT
Overall survival Progressionfree survival Relapse Non-relapse mortality
62% (52-71%) 33% (23-43%) 57% (46-67%) 13% (8-21%)
PLow SES value 0.36 68% (57-77%) 0.29 41% (30-51%) 0.51 52% (40-62%) 0.15 10% (5-17%)
AC C
EP
TE D
M AN U
* Numbers in brackets indicate 95% confidence intervals
1-year survivors Middle High SES SES 54% 62% (41-65%) (50-73%) 42% 48% (31-53%) (36-59%) 54% 51% (42-65%) (39-62%) 4% 5% (1-10%) (2-12%)
14
Pvalue 0.27
RI PT
Low SES
All patients Middle High SES SES 49% 57% (37-60%) (45-67%) 41% 43% (30-50%) (32-53%) 54% 52% (43-64%) (41-62%) 5% 8% (2-11%) (4-15%)
SC
Outcome*
0.37 0.73 0.26
ACCEPTED MANUSCRIPT
Figure Legends Figure 1. Overall survival after autologous hematopoietic cell transplantation based on
AC C
EP
TE D
M AN U
SC
RI PT
socioeconomic status
15
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
EP
TE D
M AN U
SC
RI PT
We studied role of socioeconomic status with myeloma autologous transplant outcomes Socioeconomic status had no association with outcomes for all patients Socioeconomic status had no association with outcomes in 1-year survivors
AC C
• • •
S1083-8791(16)00160-9
DOI:
10.1016/j.bbmt.2016.03.011
Reference:
YBBMT 54224
To appear in:
Biology of Blood and Marrow Transplantation
Received Date: 9 February 2016 Accepted Date: 9 March 2016
Please cite this article as: Hong S, Rybicki L, Abounader D, Bolwell BJ, Dean R, Gerds AT, Hamilton BK, Hill BT, Jagadeesh D, Kalaycio M, Liu HD, Pohlman B, Sobecks R, Majhail NS, Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell Transplantation for Multiple Myeloma, Biology of Blood and Marrow Transplantation (2016), doi: 10.1016/j.bbmt.2016.03.011. 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
Association of Socioeconomic Status with Outcomes of Autologous Hematopoietic Cell
1
RI PT
Transplantation for Multiple Myeloma
Sanghee Hong, MD, 2Lisa Rybicki, MS, 3Donna Abounader, 3,4Brian J Bolwell, MD, 3,4Robert
Dean, MD, Aaron T Gerds, MD, MS, 3,4Betty K Hamilton, MD, 3,4Brian T Hill, MD, PhD,
Deepa Jagadeesh, MD, MPH, 3,4Matt Kalaycio, MD, 3,4Hien D Liu, MD 3,4Brad Pohlman, MD,
3,4
Ronald Sobecks, MD, 3,4Navneet S Majhail, MD, MS
M AN U
1
SC
3,4
Department of Internal Medicine, Cleveland Clinic, Cleveland, OH; 2Department of
Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH; 3Blood & Marrow Transplant Program, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH; 4Department of
Corresponding Author:
Cleveland Clinic
EP
Navneet Majhail, MD, MS
TE D
Hematology and Oncology, Cleveland Clinic, Cleveland, OH
AC C
9500 Euclid Ave, R35 Cleveland, OH 44195 Phone: 216-444-2199 Fax: 216-444-9464
Email: [email protected]
Short Title: SES and myeloma AHCT outcomes 1
ACCEPTED MANUSCRIPT
Financial Disclosure: None of the authors has a financial conflict of interest to disclose in relation to this study. Word Count: Abstract – 182 words, Text – 1493 words, References – 16, Tables – 2, Figures –
RI PT
1
Key Words: Autologous stem cell transplantation, multiple myeloma, socioeconomic status,
AC C
EP
TE D
M AN U
SC
healthcare disparities, ethnicity, income
2
ACCEPTED MANUSCRIPT
ABSTRACT: Autologous hematopoietic cell transplantation (AHCT) is standard therapy for eligible patients with multiple myeloma. Healthcare disparities can influence transplantation outcomes. However,
RI PT
the association of socioeconomic status (SES), a major indicator of healthcare disparities, with outcomes in patients with myeloma after AHCT has not been previously described. We analyzed 346 consecutive AHCT recipients with myeloma transplanted between 2003 and 2013 in this
SC
retrospective cohort study. ZIP code of residence at the time of AHCT was obtained to assess annual household income based on 2010 US Census data (median $49,054; range $16,546-
M AN U
127,313). SES groups were divided into <$45,000 (low; N=120), $45-60,000 (middle; N=116), and >$60,000 (high; N=110). The low income cohort had smallest portion of Caucasians (69% vs. 89% vs. 91%), otherwise patient, disease, and transplant characteristics were comparable or different without significant patterns found among cohorts. Median follow-up was 49 months.
TE D
There was no difference among SES groups in overall survival, progression-free survival, nonrelapse mortality or relapse in univariate and multivariable analysis. Similarly, SES was not associated with survival in a subset analysis of 303 patients who had survived for 1 year post-
AC C
EP
transplant.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Healthcare disparities have been described in the epidemiology and outcomes of multiple myeloma. African-Amercians have a two-fold higher incidence of multiple myeloma (MM)
RI PT
compared to Caucasians.1-3 However, their survival is similar or even superior compared with Caucasian myeloma patients.4-6 Race/ethnicity was not found to be associated with access to transplantation or outcomes after autologous hematopoietic cell transplantation (AHCT) for
SC
myeloma, although African-American patients were younger and at the later myeloma stage at the time of transplantation.4,7,8 The impact of socioeconomic status (SES) on myeloma outcomes
M AN U
is controversial.5,6,9-11 A recent retrospective study by Fiala et al. reviewed a single center and the Surveillance, Epidemiology, and End Results-18 database and found that low SES was an independent risk factor for poor overall survival (OS) in MM.12 This study also showed that low SES by median household income limits access to AHCT.12 However, the association of SES
TE D
with survival after AHCT for patients with myeloma has not been previously described. We investigated the association of SES on outcomes after AHCT for MM.
EP
METHODS Patient Population
AC C
We identified 354 consecutive adult patients with MM who received AHCT at our institution from 2003-2013. Data were obtained from our institutional BMT database, which prospectively captures clinical and outcomes data on all transplants. The study was approved by our institutional IRB. We excluded 8 patients who did not provide consent for using data for research or had a history of prior transplantation. Therefore, our analysis cohort consisted of 346 patients. We also performed a subset analysis on 303 patients who survived at least one year after AHCT,
4
ACCEPTED MANUSCRIPT
based on an observation that transition of care of allogeneic transplant recipients from transplant center to community was found to impact long-term transplant outcomes.13
RI PT
ZIP code of residence at the time of transplantation was used to obtain median annual household income based on 2010 US Census data14 and to calculate distance from our center.
American Society of Blood and Marrow Transplant (ASBMT) Request for Information criteria
SC
were used to assign disease risk (low risk MM includes first complete remission, very good
M AN U
partial remission or partial remission; all other disease status are considered as high risk).15
Statistical Analysis
The primary outcome was OS. Secondary endpoints included progression-free survival (PFS), non-relapse mortality (NRM) and relapse. Recursive partitioning analysis was used to evaluate
TE D
the best cutpoint in median household income to predict OS. No significant cutpoints were identified, and hence SES was grouped into approximate tertiles for the analysis.12 Annual income <$45,000, $45,000-60,000, and >$60,000 were categorized as low, middle, and high SES,
EP
respectively. In addition, median household income was analyzed as a continuous variable to
AC C
assess outcome trends.
Distance from transplant center was calculated by a driving time program and categorized as ≤30, 31-60, 61-90, 91-120, and >120 miles. International Staging System (ISS) score was summarized but not used in risk factor analysis due to missing data in 27% patients.
5
ACCEPTED MANUSCRIPT
Patient-, disease-, and transplant- related characteristics were compared among SES groups with Chi-square or Kruskal-Wallis test.
RI PT
Outcomes were estimated using Kaplan-Meier or cumulative incidence methods and compared among SES groups using log-rank or Gray test. Prognostic factors were identified with Cox or Fine and Gray regression. Automated stepwise variable selection was performed on 1000
SC
bootstrap samples; variables occurring in >50% of these 1000 models were considered important prognostic factors. SES and race/ethnicity were included in all multivariable models even if they
M AN U
were not significant, while all other variables appearing in the models were significant. Variables that were assessed as potential prognostic factors in multivariable analysis included distance of residence from transplant center, year of AHCT, gender, age at transplant, history of prior cancer, performance status, number of prior chemotherapy regimens, prior radiation therapy, time from
TE D
diagnosis to transplantation and disease risk at transplantation. Analyses were performed using SAS software (version 9.4) and all P-values are two-sided.
EP
RESULTS Patient Characteristics
AC C
Table 1 describes patient characteristics. There were differences in race/ethnicity and distance from the transplant center, otherwise demographic features did not significantly differ among the SES cohorts.
Granulocyte colony stimulating factor (G-CSF) with or without plerixafor was the most common regimen for mobilization and 18% of patients received chemotherapy mobilization. Melphalan
6
ACCEPTED MANUSCRIPT
(63%) was the most commonly used conditioning regimen followed by busulfan and cyclophosphamide (37%). There were no statistical differences among SES groups by
RI PT
conditioning regimen and cell doses given for both CD34+ cells and total nucleated cells.
Outcomes by SES
Median follow-up was 49 months (range, 3-129) and was comparable among three SES groups.
SC
There were no differences in hospital stay and neutrophil/platelet engraftment among SES
groups (all with p>0.05; results not shown). Relapse (74% overall) was the most common cause
M AN U
of death for all three groups. Incidence of 100-day mortality was 1%, and included 2 patients in middle SES and 3 in high SES categories. No patients from low SES died within 100 days post AHCT.
TE D
Table 2 shows 5-year estimates of OS, PFS, relapse, and NRM by SES, with OS through 10 years in Figure 1. We did not observe any significant differences in these endpoints based on recipient SES at AHCT in univariate analyses, orin multivariable analysis after adjusting for
EP
significant patient, disease, and transplant related variables. Compared to low SES recipients, the hazard ratios (HR) for OS were 1.40 (95% CI, 0.93-2.10, P=0.11) for middle SES and 1.08 (95%
AC C
CI, 0.71-1.64, P=0.72) for high SES recipients. For PFS, the corresponding HR’s were 1.04 (95% CI, 0.74-1.47, P=0.81) and 0.69 (0.48-1.00, P=0.05), respectively. The risk of relapse was also comparable (HR 1.20, P=0.32 for middle SES and 0.78, P=0.22 for high SES compared to low SES), as was the risk for NRM (HR 0.46, P=0.06 and 0.70, P=0.35, respectively). Race/ethnicity was not found to be prognostic for any outcome.
7
ACCEPTED MANUSCRIPT
Subset Analysis of 1-year Survivors In the subset analysis of 303 1-year survivors, patient characteristics were similar among the three SES groups (data not shown). Median follow-up was 51 (range, 12-129) months. Table 2
RI PT
shows 5-year outcome estimates, which were comparable among the three groups. Causes of deaths were similar and relapse (65-78%) was the most common cause of death for all SES
groups. In multivariable analyses, SES again was not found to be associated with OS, relapse, or
SC
NRM. There was no difference in PFS between low and middle SES groups, but high SES
patients had a significant improvement in PFS compared to low SES patients (HR 0.64, 95% CI,
M AN U
0.42-0.98, P=0.04). Race/ethnicity again was not found to be associated with any outcome.
DISCUSSION
We did not observe an association between SES, as determined by median household income
TE D
based on Zip Code of residence, and outcomes after AHCT in myeloma patients. Our study adds information to the growing body of literature on healthcare disparities in myeloma,5,6,9-11 and highlights that for patients who undergo AHCT, income disparities or socioeconomic status does
EP
not directly impact outcomes. We also did not find an association of race/ethnicity with any
AC C
outcome measure after AHCT, which agrees with current literature.4-6
Although our study indicates lack of association between SES and AHCT outcomes for myeloma, we were not able to evaluate the role of socioeconomic disparities in referral for AHCT. A recent retrospective study by Fiala et al. reviewed a single center and the Surveillance, Epidemiology, and End Results-18 database and found that low SES was an independent risk factor for poor overall survival (OS) in MM.12 This study also showed that low SES by median household
8
ACCEPTED MANUSCRIPT
income limits patients access to AHCT.12 Similarly, Al-Hamadani et al. found that a lower income was negatively associated with consideration for AHCT upfront.16
RI PT
There are some limitations of our study that have to be considered. First, our population size was relatively small. We were not able to account for health disparity measures other than SES and race/ethnicity (e.g., educational status), and SES was based on median household income based
SC
on ZIP code rather than actual household income. Our study was desgined to focus on patients who had received a transplant; we did not address access to transplantation for myeloma patients.
M AN U
Also, we did not have complete information on some disease specific characteristics, such as cytogenetics and myeloma stage. Given the low number of non-Caucasians in our study, we had to group all other race/ethniticies together. Additionally, maintenance therapy after AHCT may also impact outcomes especially given the financial aspects of using expensive medications long-
TE D
term. However, use and type of maintenance therapy were not captured in our study. Furthermore, quality of life is an important of outcome of AHCT which can be impacted by SES; however, it was not analyzed in this study. Overall, these limitations emphasize the importance
EP
of futures studies with larger populations addressing this issue.
AC C
In conclusion, we did not see an association between SES and outcomes of AHCT for myeloma. However, our findings need to be validated in a large multi-center or registry based cohorts. In the meantime, transplant providers should continue to be aware of special populations including the socioeconomically disadvantaged that are risk for healthcare disparities.
ACKNOWLEDGEMENT:
9
ACCEPTED MANUSCRIPT
This study was partly supported by American Society of Hematology HONORS (Hematology
RI PT
Opportunities for the Next Generation of Research Scientists) Award to Sanghee Hong.
REFERENCES:
1. Kosary CL, Altekruse SF, Ruhl J, Lee R, Dickie L. Clinical and prognostic factors for
M AN U
1 and version 2. Cancer. 2014;120 Suppl 23:3807-3814.
SC
melanoma of the skin using SEER registries: Collaborative stage data collection system, version
2. Smith GD, Wentworth D, Neaton JD, Stamler R, Stamler J. Socioeconomic differentials in mortality risk among men screened for the multiple risk factor intervention trial: II. black men. Am J Public Health. 1996;86(4):497-504.
TE D
3. Waxman AJ, Mink PJ, Devesa SS, et al. Racial disparities in incidence and outcome in multiple myeloma: A population-based study. Blood. 2010;116(25):5501-5506.
4. Hari PN, Majhail NS, Zhang MJ, et al. Race and outcomes of autologous hematopoietic cell
EP
transplantation for multiple myeloma. Biol Blood Marrow Transplant. 2010;16(3):395-402.
AC C
5. Kaya H, Peressini B, Jawed I, et al. Impact of age, race and decade of treatment on overall survival in a critical population analysis of 40,000 multiple myeloma patients. Int J Hematol. 2012;95(1):64-70.
6. Abou-Jawde RM, Baz R, Walker E, et al. The role of race, socioeconomic status, and distance traveled on the outcome of african-american patients with multiple myeloma. Haematologica. 2006;91(10):1410-1413.
10
ACCEPTED MANUSCRIPT
7. Costa LJ, Huang JX, Hari PN. Disparities in utilization of autologous hematopoietic cell transplantation for treatment of multiple myeloma. Biol Blood Marrow Transplant.
RI PT
2015;21(4):701-706.
8. Verma PS, Howard RS, Weiss BM. The impact of race on outcomes of autologous
transplantation in patients with multiple myeloma. Am J Hematol. 2008;83(5):355-358.
SC
9. Kristinsson SY, Derolf AR, Edgren G, Dickman PW, Bjorkholm M. Socioeconomic
differences in patient survival are increasing for acute myeloid leukemia and multiple myeloma
M AN U
in sweden. J Clin Oncol. 2009;27(12):2073-2080.
10. Pasqualetti P, Colantonio D, Collacciani A, Casale R. Socioeconomic status and survival in multiple myeloma. Minerva Med. 1990;81(10):713-716.
TE D
11. Savage D, Lindenbaum J, Van Ryzin J, Struening E, Garrett TJ. Race, poverty, and survival in multiple myeloma. Cancer. 1984;54(12):3085-3094.
EP
12. Fiala MA, Finney JD, Liu J, et al. Socioeconomic status is independently associated with overall survival in patients with multiple myeloma. Leuk Lymphoma. 2015:1-7.
AC C
13. Fu S, Rybicki L, Abounader D, et al. Association of socioeconomic status with long-term outcomes in 1-year survivors of allogeneic hematopoietic cell transplantation. Bone Marrow Transplant. 2015;50(10):1326-1330.
14. US Census Bureau. American FactFinder. http://factfinder.census.gov/faces/nav/jsf/pages/index.html. Updated 2010. Accessed October 15, 2015, 2015. 11
ACCEPTED MANUSCRIPT
15. American Society of Blood and Marrow Transplantation. ASBMT RFI 2015 - disease classifications corresponding to CIBMTR classification. https://www.asbmt.org/resource/resmgr/RFI/RFI_2015_-_CIBMTR_Disease_Cl.pdf. Updated
RI PT
2015. Accessed October, 15 2015, 2015.
16. Al-Hamadani M, Hashmi SK, Go RS. Use of autologous hematopoietic cell transplantation
AC C
EP
TE D
M AN U
era of novel agents. Am J Hematol. 2014;89(8):825-830.
SC
as initial therapy in multiple myeloma and the impact of socio-geo-demographic factors in the
12
ACCEPTED MANUSCRIPT
TABLES AND FIGURES: Table 1. Patient Characteristics High SES N (%) 110 $68,501 (60,763-127,313) 22 (1.3-410)
P-value
64 (53) 37 (31) 57 (22-70) 107/118 (91)
70 (60) 13 (11) 57 (36-74) 106/112 (95)
65 (59) 10 (9) 57 (35-76) 101/104 (97)
0.51 <0.001 0.78 0.12
1 (1-5)
1 (1-7)
1 (1-5)
0.57
10 (4-330)
0.19
EP
AC C
SC
RI PT
Middle SES N (%) 116 $49,421 (45,139-59,973) 64 (3.4-1768)
11 (4-115)
9 (2-272)
-
<0.001
0.44
86 (72) 34 (28)
88 (76) 28 (24)
75 (68) 35 (32)
31 (38) 27 (33) 24 (29) 38
33 (38) 30 (34) 25 (28) 28
36 (44) 26 (32) 19 (24) 29
71 (59) 26 (22) 23 (19)
48 (41) 51 (44) 17 (15)
52 (47) 36 (33) 22 (20)
71 (59) 49 (41)* 5 (2-22)
72 (62) 44 (38) 5 (2-15)
75 (68) 35 (32) 5 (2-14)
0.86
7 (6) 57 (3-121)
5 (4) 42 (4-122)
12 (11) 48 (9-129)
0.13 0.23
TE D
N Median household income, median (range) Distance from transplant center in miles, median (range) Male sex Non-Caucasians Age, median (range) Good performance status (ECOG ≤1 or KPS ≥90) Number of prior chemotherapy regimens, median (range) Time from diagnosis to AHCT in months, median (range) Disease risk Low High ISS stage I II III Missing Mobilizing regimen G-CSF only G-CSF + Plerixafor Chemotherapy mobilization Conditioning regimen Melphalan Busulfan+cyclophosphamide CD34+ cell dose x 106/kg, median (range) Tandem transplant Followup in months, median (range)
Low SES N (%) 120 $38,409 (16,546-44,809) 54 (0.7-1036)
M AN U
Variable
0.86
0.007
0.36
Abbreviations: SES – socioeconomic status; AHCT – autologous hematopoietic cell transplantation; ECOG – Eastern Cooperative Oncology Group; KPS – Karnofsky Performance Status; ISS – International Staging System; G-CSF – granulocyte- colony stimulating factor. * Includes 2 patients who received busulfan + cyclophosphamide + etoposide
13
ACCEPTED MANUSCRIPT
Table 2. 5-year outcomes after AHCT
Overall survival Progressionfree survival Relapse Non-relapse mortality
62% (52-71%) 33% (23-43%) 57% (46-67%) 13% (8-21%)
PLow SES value 0.36 68% (57-77%) 0.29 41% (30-51%) 0.51 52% (40-62%) 0.15 10% (5-17%)
AC C
EP
TE D
M AN U
* Numbers in brackets indicate 95% confidence intervals
1-year survivors Middle High SES SES 54% 62% (41-65%) (50-73%) 42% 48% (31-53%) (36-59%) 54% 51% (42-65%) (39-62%) 4% 5% (1-10%) (2-12%)
14
Pvalue 0.27
RI PT
Low SES
All patients Middle High SES SES 49% 57% (37-60%) (45-67%) 41% 43% (30-50%) (32-53%) 54% 52% (43-64%) (41-62%) 5% 8% (2-11%) (4-15%)
SC
Outcome*
0.37 0.73 0.26
ACCEPTED MANUSCRIPT
Figure Legends Figure 1. Overall survival after autologous hematopoietic cell transplantation based on
AC C
EP
TE D
M AN U
SC
RI PT
socioeconomic status
15
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
EP
TE D
M AN U
SC
RI PT
We studied role of socioeconomic status with myeloma autologous transplant outcomes Socioeconomic status had no association with outcomes for all patients Socioeconomic status had no association with outcomes in 1-year survivors
AC C
• • •