- Email: [email protected]
Characteristics and Outcomes of Patients With Nodular Lymphocyte-Predominant Hodgkin Lymphoma Versus Those With Classical Hodgkin Lymphoma: A Population-Based Analysis
International Journal of
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
Characteristics and Outcomes of Patients With Nodular Lymphocyte-Predominant Hodgkin Lymphoma Versus Those With Classical Hodgkin Lymphoma: A Population-Based Analysis Naamit K. Gerber, MD,* Coral L. Atoria, MPH,y Elena B. Elkin, PhD,y and Joachim Yahalom, MD* *Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, New York; and yDepartment of Epidemiology and Biostatistics, Health Outcomes Research Group, New York, New York Received Dec 15, 2014, and in revised form Jan 27, 2015. Accepted for publication Feb 5, 2015.
Summary This population-based analysis compared patients with nodular lymphocytepredominant Hodgkin lymphoma (NLPHL) to those with classical Hodgkin lymphoma (CHL). We found that controlling for earlier stage and more favorable presenting characteristics resulted in a better prognosis for patients with NLPHL than for CHL patients. We also found an association between radiation therapy and improved survival in NLPHL patients, whereas we showed a decline in the use of radiation therapy over time among NLPHL patients.
Purpose: Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is rare, comprising approximately 5% of all Hodgkin lymphoma (HL) cases. Patients with NLPHL tend to have better prognoses than those with classical HL (CHL). Our goal was to assess differences in survival between NLPHL and CHL patients, controlling for differences in patient and disease characteristics. Methods and Materials: Using data from the population-based Surveillance, Epidemiology and End Results (SEER) cancer registry program, we identified patients diagnosed with pathologically confirmed HL between 1988 and 2010. Results: We identified 1,162 patients with NLPHL and 29,083 patients with CHL. With a median follow-up of 7 years, 5- and 10-year overall survival (OS) rates were 91% and 83% for NLPHL, respectively, and 81% and 74% for CHL, respectively. After adjusting for all available characteristics, NLPHL (vs CHL) was associated with higher OS (hazard ratio [HR]: 0.62, P<.01) and disease-specific survival (DSS; HR: 0.48, P<.01). The male predominance of NLPHL, compared to CHL, as well as the more favorable prognostic features in NLPHL patients are most pronounced in NLPHL patients <20 years old. Among all NLPHL patients, younger patients were less likely to receive radiation, and radiation use has declined by 40% for all patients from 1988 to 2010. Receipt of radiation was associated with better OS (HR: 0.64, PZ.03) and DSS (HR: 0.45, PZ.01) in NLPHL patients after controlling for available baseline characteristics. Other factors associated with OS and DSS in NLPHL patients are younger age and early stage. Conclusions: Our results in a large population dataset demonstrated that NLPHL patients have improved prognosis compared to CHL patients, even after accounting for
Reprint requests to: Joachim Yahalom, MD, Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 10065. Tel: (212) 639-5999; E-mail: [email protected]
Conflict of interest: none. Supplementary material www.redjournal.org.
for
this
article
can
be
found
at
Int J Radiation Oncol Biol Phys, Vol. 92, No. 1, pp. 76e83, 2015 0360-3016/Ó 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-ncnd/4.0/). http://dx.doi.org/10.1016/j.ijrobp.2015.02.012
Volume 92 Number 1 2015
A SEER analysis of NLPHL
77
stage and baseline characteristics. Use of radiation is declining among NLPHL patients despite an association in this series between radiation and better DSS and OS. Unique treatment strategies for NLPHL are warranted in both early and advanced stage disease. Ó 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is a rare histological subtype, comprising approximately 5% of all HL cases (1). The clinical characteristics, treatment approach, and outcomes of patients with NLPHL differ from those of classic HL (CHL) (1-4). NLPHL patients tend to present at an earlier stage and have an improved prognosis compared to patients with CHL (2, 5). A study from the German Hodgkin Study Group (GHSG) comparing 394 patients with NLPHL to 7,904 patients with CHL found that patients with NLPHL had lower rates of adverse prognostic features such as advanced stage at presentation, B-symptoms, involvement of 3 nodal areas, elevated erythrocyte sedimentation rate (ESR), elevated lactate dehydrogenase level, mediastinal bulky tumor or extranodal involvement. This study found improved overall survival (OS) and freedom from treatment failure (FFTF) in NLPHL patients compared with CHL patients (2). Whether this survival advantage of NLPHL patients persists after controlling for the more favorable presentation and earlier stage of NLPHL patients is not known. In the GHSG data, there were no differences in FFTF after controlling for early favorable, early unfavorable, and advanced stage at presentation (2). Similarly, a comparison of patients with stage IA or IIA NLPHL and those with CHL without mediastinal involvement found no differences in OS between the 2 groups and, in contrast to the GHSG data, similar presenting characteristics between the 2 groups (6). A matched-pairs analysis from the British Columbia Cancer Agency of advanced stage patients with NLPHL and CHL patients treated with chemotherapy revealed shorter time to progression among NLPHL patients compared to CHL patients, although OS and FFTF were similar between the 2 groups (7). Among patients with NLPHL alone, there is a paucity of data regarding prognostic factors as much of the data are extrapolated from studies that include all HL patients, most of whom have CHL. Furthermore, although differences between adult and pediatric patients with HL have been described (8), it is unclear which of these differences applies specifically to NLPHL patients. Finally, there is evidence that the use of radiation is declining among patients with CHL and it is unclear if this trend applies to NLPHL patients as well (9). Our objective was to examine differences between patients with NLPHL and CHL in a large population-based cohort. Specifically, we examined differences between the 2
groups in their demographic and disease characteristics and survival. We also studied NLPHL patients alone to evaluate trends in the use of radiation, and associations between receipt of radiation and demographic and disease characteristics.
Methods and Materials Data We analyzed data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) cancer registry program, a consortium of population-based cancer registries. We identified cases of HL diagnosed between 1988 and 2000 from 13 registries and cases diagnosed between 2000 and 2010 from 18 registries covering 14% and 28% of the US population in each time period, respectively, reflecting an expansion of the SEER program after 2000. SEER registries collect information regarding site and extent of disease, first course of cancer-directed therapy, and sociodemographic characteristics, with active follow-up for date and cause of death for all incident cancers. The SEER data are publicly available, and the study was considered exempt from informed consent requirements.
Study cohort We identified patients with a pathologically confirmed diagnosis of HL. We excluded patients with a diagnosis of Hodgkin sarcoma or Hodgkin granuloma. We also excluded patients with a previous cancer diagnosis in SEER and those whose condition was diagnosed only at the time of death. We excluded patients who were missing information about stage or radiation.
Outcomes and covariates The outcomes were OS and disease-specific survival (DSS). HL-specific deaths were obtained from the SEER CauseSpecific Death Classification, where deaths attributed to the cancer of interest are treated as events, and deaths from other causes are censored. Death records were complete through December 2011. The predictor of interest was type of HL, classified as CHL and NLPHL. Demographic covariates included patient age at diagnosis, sex, race and geographic region. Age at diagnosis was classified as under age 20, 20-45, and over 45. Disease characteristics included histologic subtype, stage,
78
Gerber et al.
extra-nodal involvement, B-symptoms, receipt of radiation and diagnosis year. SEER staging is based on the Ann Arbor Staging system and was classified as early (stages I and II) and late (stages III and IV) stage. Diagnosis years were grouped as 1988 to 1994 and 1995 to 2010.
International Journal of Radiation Oncology Biology Physics Table 1 Characteristics of nodular lymphocyte-predominant versus those of classical Hodgkin lymphoma patients
Characteristic
Statistical analysis Unadjusted associations between Hodgkin type (CHL vs NLPHL) and demographic and disease characteristics were assessed using c2 statistics. Kaplan-Meier methods were used to estimate survival functions, and the log-rank test was used to compare the resulting unadjusted survival curves. Cox proportional-hazards regression was used to estimate the impact of type of HL (CHL vs NLPHL) on overall and HL-specific survival, controlling for demographic and disease characteristics. All variables were entered into the multivariate model. Hazard ratios, 95% confidence intervals, and 2-sided P values were estimated for each covariate. Competing-risks regression models were also used to assess the impact of characteristics on outcomes, with observations censored at the end of follow-up and death from other causes treated as competing events. We performed stratified analyses to assess whether the relationship between types of HL and death varied by stage at diagnosis. Similar analyses were also performed in the subset of patients with a diagnosis of NLPHL to assess the impact of demographic and disease characteristics on the risk of overall and HL-specific death. In addition, we assessed unadjusted associations of age at diagnosis with demographic and disease characteristics in this subset. We estimated trends in radiation use over time and use of radiation by age and stage. Analysis was performed using SAS version 9.2 software (SAS Institute, Cary, NC) and R software (R Foundation for Statistical Computing, Vienna, Austria).
Results We identified 1,162 patients with NLPHL and 29,083 patients with CHL. Median age at diagnosis of NLPHL and CHL was 38 and 35 years of age, respectively. NLPHL patients were more likely to be male (69% vs 54%, respectively, P<.01) and black (22% vs 11%, respectively, P<.01) (Table 1). NLPHL patients were more likely to present with stage I disease (50% vs 20%, respectively, P<.01) and less likely to have B symptoms (11% vs 37%, respectively, P<.01) or extranodal extension (11% vs 24%, respectively, P<.01). NLPHL patients were more likely to have received radiation therapy (RT; 49% vs 40%, P<.01). Among early-stage patients, 59% of NLPHL versus 52% of CHL patients received radiation. NLPHL patients were more prevalent among the highest income quartile whereas CHL patients were more prevalent in the lowest income quartile.
NLPHL
Classic HL
nZ1162
nZ29,083
n
%
n
%
P
Median age at 38 (25-53) 35 (24-51) .0562* diagnosis (IQR) Age at diagnosis <20 186 16% 3936 14% <.0001 20-45 441 38% 9308 32% >45 535 46% 15,839 54% Sex Male 806 69% 15,723 54% <.0001 Female 356 31% 13,360 46% Race White 853 73% 24,682 85% <.0001 Black 258 22% 3119 11% Other 51 4% 1282 4% Geographic region Midwest 190 16% 4132 14% .02 Northeast 171 15% 4984 17% South 213 18% 4832 17% West 588 51% 15,135 52% Urban/rural Metropolitan 1056 91% 25,931 89% NS Nonmetropolitan 106 9% 3152 11% Income quartile Lowest quartile 245 21% 7370 25% <.0001 Second quartile 312 27% 8113 28% Third quartile 253 22% 6458 22% Fourth quartile 352 30% 7142 25% Stage I 578 50% 5735 20% <.0001 II 303 26% 12,079 42% III 204 18% 5921 20% IV 77 7% 5348 18% Extranodal extension Yes 129 11% 7011 24% <.0001 No 1033 89% 22,072 76% B symptoms Absent 765 66% 12,128 42% <.0001 Present 131 11% 10,815 37% Unknown 266 23% 6140 21% Radiation Received 564 49% 11,585 40% <.0001 Did not receive 598 51% 17,498 60% Diagnosis year 1988-1994 113 10% 5064 17% <.0001 1995-2010 1049 90% 24,019 83% Abbreviations: IQR Z interquartile range; NLPHL Z nodular lymphocyte-predominant Hodgkin lymphoma; NS Z not significant. * Kruskal-Wallis test was used for differences in medians.
With a median follow-up of 6.9 years, rates of 5-, 10-, and 15-year OS were 91%, 83%, and 73%, respectively, for NLPHL patients and 81%, 74%, and 69%, respectively, for CHL patients (P<.01) (Fig. 1). DSS rates at 5, 10, and 15 years were 95%, 93%, and 88% for NLPHL
Volume 92 Number 1 2015
a
A SEER analysis of NLPHL
b
0.75
0.50
Nodular lymphocyte-predominant
0.25
1.00
Proportion Surviving
Proportion Surviving
1.00
0.75
0.50
Nodular lymphocyte-predominant
0.25
Classical
Classical
Log rank P=.0005
Log rank P=.0003
0.00 0
2
4
6
8
10
12
0.00
14
0
2
4
6
Years from Diagnosis No. at risk NLP 881 Classical 17814
c
715 14551
564 11926
423 9630
299 7527
214 5469
142 4034
99 3040
No. at risk NLP 281 Classical 11269
d
0.75
Proportion Surviving
Proportion Surviving
8
10
12
14
41 2409
24 1812
15 1425
10
12
14
41 2409
24 1812
15 1425
Years from Diagnosis
1.00
0.50
Nodular lymphocyte-predominant
0.25
79
193 7718
140 5908
98 4492
65 3368
6
8
1.00
0.75
0.50
0.25
Nodular lymphocyte-predominant
Classical
Classical
Log rank P<.0001
Log rank P<.0001
0.00
0.00
0 No. at risk NLP 881 Classical 17814
2 715 14551
4 564 11926
8 6 Years from Diagnosis 423 9630
299 7527
10 214 5469
12 142 4034
14 99 3040
0
No. at risk NLP 281 Classical 11269
2
4
Years from Diagnosis 193 7718
140 5908
98 4492
65 3368
Fig. 1. (a) Overall survival for early stage patients by subtype. (b) Overall survival for advanced stage patients by subtype. (c) Disease-specific survival for early stage patients by subtype. (d) Disease-specific survival for advanced stage patients by subtype. NLP Z nodular lymphocyte-predominant Hodgkin lymphoma. patients and 86%, 82% and 80% for CHL patients (P<.01). This survival advantage among NLPHL patients was present among both early stage patients (P<.01) and advanced stage patients (P<.01). Similarly, DSS was better in NLPHL patients with early (P<.01) or late stage (P<.01) disease. In multivariate analysis (MVA) for OS and DSS, NLPHL was associated with better OS (HR: 0.62, 95% CI 0.53-0.73, P<.01) and DSS (HR: 0.48, 95% CI 0.38-0.62, P<.01). The male predominance in NLPHL (vs CHL) was most pronounced among NLPHL patients <20 years old (78%) with the male predominance decreasing with increasing age. The more favorable baseline characteristics among NLPHL (vs CHL patients) such as lower incidence of B symptoms and extranodal disease were also most pronounced in the younger NLPHL patients (Table 2). As shown in Figure 2, radiation use differed by age among early stage patients, with younger patients least likely to receive radiation (38%) compared to patients 20 to 45 years of age (59%) and patients >45 years of age (66%). Use of radiation in NLPHL declined from 70% of NLPHL patients in 1988 to 1990 to 42% in 2006 to 2010 (Fig. 3),
representing a 40% decline over this time period. This trend was observed in both early-stage patients and advancedstage patients, with a 31% relative decrease among early stage patients and a 61% relative decrease among advanced stage patients over the study period. In NLPHL patients, younger age at diagnosis, early stage, and receipt of radiation were all associated with better OS and DSS, controlling for other disease and demographic characteristics (Table 3 and see Supplementary Table E1; available online at www.redjournal.com).
Discussion Much of the data for prognosis, outcomes, and treatment strategies in NLPHL were derived from series of HL, of which NLPHL patients comprise a small percentage of the total number of studied patients. When NLPHL is examined in comparison to CHL, conflicting data emerges. The GHSG analysis of greater than 8,000 patients demonstrated improved prognosis in NLPHL compared to CHL patients (2). Yet there was no difference in FFTF between NLPHL
80
International Journal of Radiation Oncology Biology Physics
Gerber et al.
Table 2
Nodular lymphocyte-predominant Hodgkin lymphoma patient characteristics by age at diagnosis Age at diagnosis <20 (nZ186)
Characteristic Sex Male Female Race White Black Other Geographic region Midwest Northeast South West Urban/rural Metropolitan Nonmetropolitan Income quartile Lowest quartile Second quartile Third quartile Fourth quartile Stage I/II III/IV Extranodal extension Yes No B symptoms Absent Present Unknown Radiation Received Did not receive Diagnosis year 1988-1994 1995-2010
>45 (nZ441)
20-45 (nZ535)
n
%
n
%
n
%
P
146 40
78% 22%
382 153
71% 29%
278 163
63% 37%
.001
147 28 11
79% 15% 6%
376 137 22
70% 26% 4%
330 93 18
75% 21% 4%
.04
23 29 40 94
12% 16% 22% 51%
88 67 98 282
16% 13% 18% 53%
79 75 75 212
18% 17% 17% 48%
NS
165 21
89% 11%
491 44
92% 8%
400 41
91% 9%
NS
38 41 49 58
20% 22% 26% 31%
116 149 118 152
22% 28% 22% 28%
91 122 86 142
21% 28% 20% 32%
NS
146 40
78% 22%
417 118
78% 22%
318 123
72% 28%
.0687
14 172
8% 92%
64 471
12% 88%
51 390
12% 88%
NS
146 8 32
78% 4% 17%
341 70 124
64% 13% 23%
278 53 110
63% 12% 25%
<.001
65 121
35% 65%
296 239
55% 45%
203 238
46% 54%
<.0001
12 174
6% 94%
57 478
11% 89%
44 397
10% 90%
NS
Abbreviation: NS Z not significant.
and CHL patients when controlling for stage at presentation. This lack of difference may be due to the smaller number of NLPHL patients for any given stage. In this analysis of more than 30,000 patients, NLPHL patients had better OS and DSS than CHL patients. In MVA accounting for other known prognostic features such as stage at presentation, NLPHL histology was associated with better OS and DSS. Our finding of improved prognosis in NLPHL patients after controlling for stage and other prognostic features is confirmed in the recently published series using the SEER data (10). Our findings support previously published reports that NLPHL patients have more favorable characteristics and earlier stage at presentation than CHL patients (2, 11). One novel contribution of the current series is that this
difference was most pronounced among NLPHL patients <20 years old who, as compared to older patients with NLPHL, had a lower incidence of B symptoms and extranodal extension. Even after controlling for the more favorable prognostic factors among pediatric patients, age <20 years old was associated with better OS and DSS in NLPHL patients. The association of age with outcome was linear with patients 20 to 45 years of age demonstrating improved prognosis compared to patients >45 years old. The more favorable presentation and outcomes of pediatric NLPHL patients is concordant with the data for all HL patients in which pediatric patients present with more favorable disease characteristics and demonstrate better survival (8). Based on data from the British Columbia Cancer Agency in which adolescents
Volume 92 Number 1 2015
A SEER analysis of NLPHL
% of Patients Receiving Radiation
1 0.8 69%
66%
0.6 0.4
58% 60%
60%
59%
<20 38%
37%
20-45
40%
>45
0.2 0 Stage I and II
Stage I
Stage II
Fig. 2. Use of radiation in early stage nodular lymphocyte-predominant Hodgkin lymphoma. (16-21 years old) and young adults (22-45 years of age) with HL demonstrated similar outcomes (12), it is likely that the improved prognosis in the patients <20 years old in this series was driven by the youngest patients, although the fact that young adults (20-45 years of age) demonstrated improved prognosis compared to the older adults in this series provides evidence that age may function as a continuous variable with older age associated with poorer outcomes in NLPHL. Radiation is instrumental in the cure of early stage NLPHL patients. Radiation alone is the recommended treatment for patients with stage IA or IIA disease (13). Large cooperative group trials and single-institution series demonstrate excellent outcomes from radiation alone in early stage NLPHL (14-16) or radiation in combination with chemotherapy (17-19). Recent data from the GHSG demonstrate excellent outcomes with involved-field RT (IFRT) alone in stage IA patients and show no benefit with the addition of chemotherapy in early stage patients (20). In this study, receipt of radiation was associated with better OS and DSS on MVA with a HR of 0.66 for OS and 0.45 for DSS. Yet, this study also demonstrated a decline in receipt of radiation over time, a trend which mirrors the declining use of radiation in CHL (9). The declining use of radiation may
% of Patients Receiving Radiation
1 Stage I/II NLPHL Only
0.8
0.6
0.4
81
stem in large part from concern for late effects as data concerning late effects of radiation in HL survivors have emerged over the past two and a half decades. Nonradiationbased strategies including surgical resection followed by observation and observation alone have been investigated, particularly in the pediatric population (21-23), but many of these patients develop relapse and observation alone is not routinely recommended, with the exception of certain pediatric patients (3). In the recent Children’s Oncology Group protocol for pediatric patients with early-stage HL, patients are treated with upfront chemotherapy with no further treatment if a complete response was achieved (Combination Chemotherapy Followed by Radiation Therapy in Treating Young Patients With Newly Diagnosed Hodgkin’s Lymphoma; clinicaltrials.gov/NCT00302003). These differing treatment paradigms in the pediatric population as well as the concern for late effects likely explains our finding of decreased radiation utilization among patients <20 years old. Given the expression of CD20 in NLPHL, prospective studies have also investigated rituximab alone in early stage disease but despite high response rates, the relapse rate with rituximab alone is high and thus rituximab is not routinely used in early-stage disease (24-26). Although no definitive conclusions can be drawn regarding the causal association of better OS and DSS with receipt of radiation, the finding is noteworthy and the downward trend in radiation utilization is significant in light of this finding. Despite this downward trend in utilization, radiation techniques have improved and modern techniques such as involved site RT should be utilized in all patients with LPHL receiving RT to minimize treatment fields and treatment morbidity (27). This study has several limitations owing in large part to relevant prognostic and treatment data not available in the SEER database such as ESR, bulky disease, the involvement of three or more nodal regions, the use of chemotherapy, and information on relapse. Given these limitations, this study is unable to answer important questions with regard to NLPHL in comparison to CHL such as time-course and rates of relapse, and the use of chemotherapy and/or rituximab and its association with OS and DSS. It also introduces potential confounders associated with histology that cannot be accounted for in the analysis. An additional weakness of this study is that we were unable to confirm the cases classified as NLPHL with central pathology review which, given prior studies showing high rates of misclassification of NLPHL (1, 28), likely led to erroneous classification of some of our patients.
0.2
0 1988-1990
Conclusions 1991-1995
1996-2000
2001-2005
2006-2010
Year of Diagnosis
Fig. 3. Use of radiation by year in all nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) patients and those with early stage disease.
Taken together, this large population-based series demonstrates that better survival of NLPHL patients as compared to CHL patients persists even after controlling for more favorable presenting characteristics. It demonstrates an
82
International Journal of Radiation Oncology Biology Physics
Gerber et al.
Table 3 Multivariate analysis of factors associated with overall and disease-specific survival in nodular lymphocyte-predominant Hodgkin lymphoma patients Overall survival Factor Age >45 20-45 <20 Sex Female Male Race White Black and other Geographic region Northeast Midwest South West Urban/rural Metropolitan Nonmetropolitan Income quartile Lowest quartile Second quartile Third quartile Fourth quartile Stage III and IV I and II Extranodal extension No Yes B symptoms Absent Present Unknown Radiation Did not receive Received Diagnosis year 1995-2010 1988-1994
Disease-specific survival
HR, 95% CI
P
HR, 95% CI
P
Ref 0.26 (0.18-0.37) 0.09 (0.03-0.25)
<.0001
Ref 0.31 (0.18-0.52) 0.06 (0.01-0.41)
<.0001
Ref 0.88 (0.62-1.24)
NS
Ref 0.89 (0.53-1.51)
NS
Ref 1.28 (0.87-1.88)
NS
Ref 1.28 (0.73-2.24)
NS
Ref 0.91 (0.48-1.75) 0.84 (0.42-1.70) 1.34 (0.77-2.33) Ref 1.03 (0.60-1.78) Ref 0.85 (0.56-1.30) 0.67 (0.39-1.14) 0.92 (0.54-1.57)
NS
NS
NS
Ref 1.03 (0.43-2.45) 0.60 (0.21-1.71) 1.10 (0.52-2.34) Ref 1.03 (0.43-2.48) Ref 1.21 (0.64-2.31) 0.62 (0.27-1.45) 1.11 (0.49-2.47)
NS
NS
NS
Ref 0.61 (0.41-0.91)
.02
Ref 0.47 (0.27-0.82)
.01
Ref 1.43 (0.90-2.27)
NS
Ref 1.28 (0.66-2.47)
NS
Ref 1.23 (0.75-2.02) 1.15 (0.79-1.68)
NS
Ref 1.66 (0.84-3.29) 1.62 (0.93-2.83)
NS
Ref 0.64 (0.44-0.93)
.03
Ref 0.43 (0.24-0.78)
.01
Ref 1.69 (1.08-2.63)
.02
Ref 1.75 (0.87-3.53)
NS
Abbreviations: CI Z confidence interval; HR Z hazard ratio; NS Z not significant; Ref Z reference value.
association between radiation receipt and better survival in NLPHL patients while confirming a decline in radiation utilization for patients with this disease. Further research refining unique treatment strategies in both early and advanced NLPHL patients is needed.
2.
3. 4.
References 1. Diehl V, Sextro M, Franklin J, et al. Clinical presentation, course, and prognostic factors in lymphocyte-predominant Hodgkin’s disease and lymphocyte-rich classical Hodgkin’s disease: Report from the
5.
6.
European Task Force on Lymphoma Project on LymphocytePredominant Hodgkin’s Disease. J Clin Oncol 1999;17:776-783. Nogova L, Reineke T, Brillant C, et al. Lymphocyte-predominant and classical Hodgkin’s lymphoma: A comprehensive analysis from the German Hodgkin Study Group. J Clin Oncol 2008;26:434-439. Advani RH, Hoppe RT. How I treat nodular lymphocyte predominant Hodgkin lymphoma. Blood 2013;122:4182-4188. Fanale MA, Younes A. Nodular lymphocyte predominant Hodgkin’s lymphoma. Cancer Treat Res 2008;142:367-381. Allemani C, Sant M, De Angelis R, et al. Hodgkin disease survival in Europe and the U.S.: Prognostic significance of morphologic groups. Cancer 2006;107:352-360. Feugier P, Labouyrie E, Djeridane M, et al. Comparison of initial characteristics and long-term outcome of patients with lymphocytepredominant Hodgkin lymphoma and classical Hodgkin lymphoma
Volume 92 Number 1 2015
7.
8.
9.
10.
11.
12. 13.
14.
15.
16. 17.
at clinical stages IA and IIA prospectively treated by brief anthracycline-based chemotherapies plus extended high-dose irradiation. Blood 2004;104:2675-2681. Xing KH, Conners JM, Al Monsour M, et al. Advanced stage nodular lymphocyte predominant Hodgkin’s lymphoma compared to classical Hodgkin lymphoma: A matched pair outcome analysis. Blood 2014; 123:3567-3573. Bazzeh F, Rihani R, Howard S, et al. Comparing adult and pediatric Hodgkin lymphoma in the Surveillance, Epidemiology and End Results Program, 1988-2005: An analysis of 21,734 cases. Leuk Lymphoma 2010;51:2198-2207. Koshy M, Rich SE, Mahmood U, et al. The declining utilization of radiation therapy in stage I and II Hodgkin’s disease and its impact on survival and secondary malignancies. Int J Radiat Oncol Biol Phys 2012;82:619-625. Ali S, Olszewski AJ. Disparate survival and risk of secondary nonHodgkin lymphoma in histologic subtypes of Hodgkin lymphoma: A population-based study. Leuk Lymphoma 2014;55:1570-1577. Bodis S, Kraus MD, Pinkus G, et al. Clinical presentation and outcome in lymphocyte-predominant Hodgkin’s disease. J Clin Oncol 1997;15:3060-3066. Foltz LM, Song KW, Connors JM. Hodgkin’s lymphoma in adolescents. J Clin Oncol 2006;24:2520-2526. Eichenauer DA, Engert A, Dreyling M, et al. Hodgkin’s lymphoma: ESMO clinical practice guidelines for diagnosis, treatment and followup. Ann Oncol 2011;22:55-58. Chen RC, Chin MS, Ng AK, et al. Early-stage, lymphocytepredominant Hodgkin’s lymphoma: Patient outcomes from a large, single-institution series with long follow-up. J Clin Oncol 2010;28: 136-141. Wirth A, Yuen K, Barton M, et al. Long-term outcome after radiotherapy alone for lymphocyte-predominant Hodgkin lymphoma: A retrospective multicenter study of the Australasian Radiation Oncology Lymphoma Group. Cancer 2005;104:1221-1229. Schlembach P, Wilder R, Jones D, et al. Radiotherapy alone for lymphocyte-predominant Hodgkin’s disease. Cancer J 2002;8:377-383. Jackson C, Sirohi B, Cunningham D, et al. Lymphocyte-predominant Hodgkin lymphomadClinical features and treatment outcomes from a 30-year experience. Ann Oncol 2010;21:2061-2068.
A SEER analysis of NLPHL
83
18. Savage KJ, Skinnider B, Al-Mansour M, et al. Treating limited-stage nodular lymphocyte predominant Hodgkin lymphoma similarly to classical Hodgkin lymphoma with ABVD may improve outcome. Blood 2011;118:4585-4590. 19. Lazarovici J, Brice P, Oberic L. Management and outcomes of nodular lymphocyte predominant hodgkin lymphoma: A retrospective study by the lymphoma study association. Blood 2014;124:3054. 20. Eichenauer DA, Pluetschow A, Fuchs M, et al. Treatment of stage IA nodular lymphocyte-predominant hodgkin lymphoma (NLPHL): An analysis from the German Hodgkin Study Group (GHSG). Blood 2014;124:3058. 21. Murphy SB, Morgan ER, Katzenstein HM, et al. Results of little or no treatment for lymphocyte-predominant Hodgkin disease in children and adolescents. J Pediatr Hematol Oncol 2003;25:684-687. 22. Pellegrino B, Terrier-Lacombe MJ, Oberlin O, et al. Lymphocytepredominant Hodgkin’s lymphoma in children: Therapeutic abstention after initial lymph node resectiondA study of the French Society of Pediatric Oncology. J Clin Oncol 2003;21:2948-2952. 23. Mauz-Korholz C, Gorde-Grosjean S, Hasenclever D, et al. Resection alone in 58 children with limited stage, lymphocyte-predominant Hodgkin lymphoma-experience from the European network group on pediatric Hodgkin lymphoma. Cancer 2007;110:179-185. 24. Advani RH, Horning SJ, Hoppe RT, et al. Mature results of a phase II study of rituximab therapy for nodular lymphocyte-predominant Hodgkin lymphoma. J Clin Oncol 2014;32:912-918. 25. Ekstrand BC, Lucas JB, Horwitz SM, et al. Rituximab in lymphocytepredominant Hodgkin disease: results of a phase 2 trial. Blood 2003; 101:4285-4289. 26. Eichenauer DA, Fuchs M, Pluetschow A, et al. Phase 2 study of rituximab in newly diagnosed stage IA nodular lymphocyte-predominant Hodgkin lymphoma: A report from the German Hodgkin Study Group. Blood 2011;118:4363-4365. 27. Specht L, Yahalom J, Illidge T, et al. Modern radiation therapy for Hodgkin lymphoma: Field and dose guidelines from the International Lymphoma Radiation Oncology Group (ILROG). Int J Radiat Oncol Biol Phys 2014;89:854-862. 28. Pappa VI, Norton AJ, Gupta RK, et al. Nodular type of lymphocyte predominant Hodgkin’s disease. A clinical study of 50 cases. Ann Oncol 1995;6:559-565.
Radiation Oncology biology
physics
www.redjournal.org
Clinical Investigation
Characteristics and Outcomes of Patients With Nodular Lymphocyte-Predominant Hodgkin Lymphoma Versus Those With Classical Hodgkin Lymphoma: A Population-Based Analysis Naamit K. Gerber, MD,* Coral L. Atoria, MPH,y Elena B. Elkin, PhD,y and Joachim Yahalom, MD* *Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, New York; and yDepartment of Epidemiology and Biostatistics, Health Outcomes Research Group, New York, New York Received Dec 15, 2014, and in revised form Jan 27, 2015. Accepted for publication Feb 5, 2015.
Summary This population-based analysis compared patients with nodular lymphocytepredominant Hodgkin lymphoma (NLPHL) to those with classical Hodgkin lymphoma (CHL). We found that controlling for earlier stage and more favorable presenting characteristics resulted in a better prognosis for patients with NLPHL than for CHL patients. We also found an association between radiation therapy and improved survival in NLPHL patients, whereas we showed a decline in the use of radiation therapy over time among NLPHL patients.
Purpose: Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is rare, comprising approximately 5% of all Hodgkin lymphoma (HL) cases. Patients with NLPHL tend to have better prognoses than those with classical HL (CHL). Our goal was to assess differences in survival between NLPHL and CHL patients, controlling for differences in patient and disease characteristics. Methods and Materials: Using data from the population-based Surveillance, Epidemiology and End Results (SEER) cancer registry program, we identified patients diagnosed with pathologically confirmed HL between 1988 and 2010. Results: We identified 1,162 patients with NLPHL and 29,083 patients with CHL. With a median follow-up of 7 years, 5- and 10-year overall survival (OS) rates were 91% and 83% for NLPHL, respectively, and 81% and 74% for CHL, respectively. After adjusting for all available characteristics, NLPHL (vs CHL) was associated with higher OS (hazard ratio [HR]: 0.62, P<.01) and disease-specific survival (DSS; HR: 0.48, P<.01). The male predominance of NLPHL, compared to CHL, as well as the more favorable prognostic features in NLPHL patients are most pronounced in NLPHL patients <20 years old. Among all NLPHL patients, younger patients were less likely to receive radiation, and radiation use has declined by 40% for all patients from 1988 to 2010. Receipt of radiation was associated with better OS (HR: 0.64, PZ.03) and DSS (HR: 0.45, PZ.01) in NLPHL patients after controlling for available baseline characteristics. Other factors associated with OS and DSS in NLPHL patients are younger age and early stage. Conclusions: Our results in a large population dataset demonstrated that NLPHL patients have improved prognosis compared to CHL patients, even after accounting for
Reprint requests to: Joachim Yahalom, MD, Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 10065. Tel: (212) 639-5999; E-mail: [email protected]
Conflict of interest: none. Supplementary material www.redjournal.org.
for
this
article
can
be
found
at
Int J Radiation Oncol Biol Phys, Vol. 92, No. 1, pp. 76e83, 2015 0360-3016/Ó 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-ncnd/4.0/). http://dx.doi.org/10.1016/j.ijrobp.2015.02.012
Volume 92 Number 1 2015
A SEER analysis of NLPHL
77
stage and baseline characteristics. Use of radiation is declining among NLPHL patients despite an association in this series between radiation and better DSS and OS. Unique treatment strategies for NLPHL are warranted in both early and advanced stage disease. Ó 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is a rare histological subtype, comprising approximately 5% of all HL cases (1). The clinical characteristics, treatment approach, and outcomes of patients with NLPHL differ from those of classic HL (CHL) (1-4). NLPHL patients tend to present at an earlier stage and have an improved prognosis compared to patients with CHL (2, 5). A study from the German Hodgkin Study Group (GHSG) comparing 394 patients with NLPHL to 7,904 patients with CHL found that patients with NLPHL had lower rates of adverse prognostic features such as advanced stage at presentation, B-symptoms, involvement of 3 nodal areas, elevated erythrocyte sedimentation rate (ESR), elevated lactate dehydrogenase level, mediastinal bulky tumor or extranodal involvement. This study found improved overall survival (OS) and freedom from treatment failure (FFTF) in NLPHL patients compared with CHL patients (2). Whether this survival advantage of NLPHL patients persists after controlling for the more favorable presentation and earlier stage of NLPHL patients is not known. In the GHSG data, there were no differences in FFTF after controlling for early favorable, early unfavorable, and advanced stage at presentation (2). Similarly, a comparison of patients with stage IA or IIA NLPHL and those with CHL without mediastinal involvement found no differences in OS between the 2 groups and, in contrast to the GHSG data, similar presenting characteristics between the 2 groups (6). A matched-pairs analysis from the British Columbia Cancer Agency of advanced stage patients with NLPHL and CHL patients treated with chemotherapy revealed shorter time to progression among NLPHL patients compared to CHL patients, although OS and FFTF were similar between the 2 groups (7). Among patients with NLPHL alone, there is a paucity of data regarding prognostic factors as much of the data are extrapolated from studies that include all HL patients, most of whom have CHL. Furthermore, although differences between adult and pediatric patients with HL have been described (8), it is unclear which of these differences applies specifically to NLPHL patients. Finally, there is evidence that the use of radiation is declining among patients with CHL and it is unclear if this trend applies to NLPHL patients as well (9). Our objective was to examine differences between patients with NLPHL and CHL in a large population-based cohort. Specifically, we examined differences between the 2
groups in their demographic and disease characteristics and survival. We also studied NLPHL patients alone to evaluate trends in the use of radiation, and associations between receipt of radiation and demographic and disease characteristics.
Methods and Materials Data We analyzed data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) cancer registry program, a consortium of population-based cancer registries. We identified cases of HL diagnosed between 1988 and 2000 from 13 registries and cases diagnosed between 2000 and 2010 from 18 registries covering 14% and 28% of the US population in each time period, respectively, reflecting an expansion of the SEER program after 2000. SEER registries collect information regarding site and extent of disease, first course of cancer-directed therapy, and sociodemographic characteristics, with active follow-up for date and cause of death for all incident cancers. The SEER data are publicly available, and the study was considered exempt from informed consent requirements.
Study cohort We identified patients with a pathologically confirmed diagnosis of HL. We excluded patients with a diagnosis of Hodgkin sarcoma or Hodgkin granuloma. We also excluded patients with a previous cancer diagnosis in SEER and those whose condition was diagnosed only at the time of death. We excluded patients who were missing information about stage or radiation.
Outcomes and covariates The outcomes were OS and disease-specific survival (DSS). HL-specific deaths were obtained from the SEER CauseSpecific Death Classification, where deaths attributed to the cancer of interest are treated as events, and deaths from other causes are censored. Death records were complete through December 2011. The predictor of interest was type of HL, classified as CHL and NLPHL. Demographic covariates included patient age at diagnosis, sex, race and geographic region. Age at diagnosis was classified as under age 20, 20-45, and over 45. Disease characteristics included histologic subtype, stage,
78
Gerber et al.
extra-nodal involvement, B-symptoms, receipt of radiation and diagnosis year. SEER staging is based on the Ann Arbor Staging system and was classified as early (stages I and II) and late (stages III and IV) stage. Diagnosis years were grouped as 1988 to 1994 and 1995 to 2010.
International Journal of Radiation Oncology Biology Physics Table 1 Characteristics of nodular lymphocyte-predominant versus those of classical Hodgkin lymphoma patients
Characteristic
Statistical analysis Unadjusted associations between Hodgkin type (CHL vs NLPHL) and demographic and disease characteristics were assessed using c2 statistics. Kaplan-Meier methods were used to estimate survival functions, and the log-rank test was used to compare the resulting unadjusted survival curves. Cox proportional-hazards regression was used to estimate the impact of type of HL (CHL vs NLPHL) on overall and HL-specific survival, controlling for demographic and disease characteristics. All variables were entered into the multivariate model. Hazard ratios, 95% confidence intervals, and 2-sided P values were estimated for each covariate. Competing-risks regression models were also used to assess the impact of characteristics on outcomes, with observations censored at the end of follow-up and death from other causes treated as competing events. We performed stratified analyses to assess whether the relationship between types of HL and death varied by stage at diagnosis. Similar analyses were also performed in the subset of patients with a diagnosis of NLPHL to assess the impact of demographic and disease characteristics on the risk of overall and HL-specific death. In addition, we assessed unadjusted associations of age at diagnosis with demographic and disease characteristics in this subset. We estimated trends in radiation use over time and use of radiation by age and stage. Analysis was performed using SAS version 9.2 software (SAS Institute, Cary, NC) and R software (R Foundation for Statistical Computing, Vienna, Austria).
Results We identified 1,162 patients with NLPHL and 29,083 patients with CHL. Median age at diagnosis of NLPHL and CHL was 38 and 35 years of age, respectively. NLPHL patients were more likely to be male (69% vs 54%, respectively, P<.01) and black (22% vs 11%, respectively, P<.01) (Table 1). NLPHL patients were more likely to present with stage I disease (50% vs 20%, respectively, P<.01) and less likely to have B symptoms (11% vs 37%, respectively, P<.01) or extranodal extension (11% vs 24%, respectively, P<.01). NLPHL patients were more likely to have received radiation therapy (RT; 49% vs 40%, P<.01). Among early-stage patients, 59% of NLPHL versus 52% of CHL patients received radiation. NLPHL patients were more prevalent among the highest income quartile whereas CHL patients were more prevalent in the lowest income quartile.
NLPHL
Classic HL
nZ1162
nZ29,083
n
%
n
%
P
Median age at 38 (25-53) 35 (24-51) .0562* diagnosis (IQR) Age at diagnosis <20 186 16% 3936 14% <.0001 20-45 441 38% 9308 32% >45 535 46% 15,839 54% Sex Male 806 69% 15,723 54% <.0001 Female 356 31% 13,360 46% Race White 853 73% 24,682 85% <.0001 Black 258 22% 3119 11% Other 51 4% 1282 4% Geographic region Midwest 190 16% 4132 14% .02 Northeast 171 15% 4984 17% South 213 18% 4832 17% West 588 51% 15,135 52% Urban/rural Metropolitan 1056 91% 25,931 89% NS Nonmetropolitan 106 9% 3152 11% Income quartile Lowest quartile 245 21% 7370 25% <.0001 Second quartile 312 27% 8113 28% Third quartile 253 22% 6458 22% Fourth quartile 352 30% 7142 25% Stage I 578 50% 5735 20% <.0001 II 303 26% 12,079 42% III 204 18% 5921 20% IV 77 7% 5348 18% Extranodal extension Yes 129 11% 7011 24% <.0001 No 1033 89% 22,072 76% B symptoms Absent 765 66% 12,128 42% <.0001 Present 131 11% 10,815 37% Unknown 266 23% 6140 21% Radiation Received 564 49% 11,585 40% <.0001 Did not receive 598 51% 17,498 60% Diagnosis year 1988-1994 113 10% 5064 17% <.0001 1995-2010 1049 90% 24,019 83% Abbreviations: IQR Z interquartile range; NLPHL Z nodular lymphocyte-predominant Hodgkin lymphoma; NS Z not significant. * Kruskal-Wallis test was used for differences in medians.
With a median follow-up of 6.9 years, rates of 5-, 10-, and 15-year OS were 91%, 83%, and 73%, respectively, for NLPHL patients and 81%, 74%, and 69%, respectively, for CHL patients (P<.01) (Fig. 1). DSS rates at 5, 10, and 15 years were 95%, 93%, and 88% for NLPHL
Volume 92 Number 1 2015
a
A SEER analysis of NLPHL
b
0.75
0.50
Nodular lymphocyte-predominant
0.25
1.00
Proportion Surviving
Proportion Surviving
1.00
0.75
0.50
Nodular lymphocyte-predominant
0.25
Classical
Classical
Log rank P=.0005
Log rank P=.0003
0.00 0
2
4
6
8
10
12
0.00
14
0
2
4
6
Years from Diagnosis No. at risk NLP 881 Classical 17814
c
715 14551
564 11926
423 9630
299 7527
214 5469
142 4034
99 3040
No. at risk NLP 281 Classical 11269
d
0.75
Proportion Surviving
Proportion Surviving
8
10
12
14
41 2409
24 1812
15 1425
10
12
14
41 2409
24 1812
15 1425
Years from Diagnosis
1.00
0.50
Nodular lymphocyte-predominant
0.25
79
193 7718
140 5908
98 4492
65 3368
6
8
1.00
0.75
0.50
0.25
Nodular lymphocyte-predominant
Classical
Classical
Log rank P<.0001
Log rank P<.0001
0.00
0.00
0 No. at risk NLP 881 Classical 17814
2 715 14551
4 564 11926
8 6 Years from Diagnosis 423 9630
299 7527
10 214 5469
12 142 4034
14 99 3040
0
No. at risk NLP 281 Classical 11269
2
4
Years from Diagnosis 193 7718
140 5908
98 4492
65 3368
Fig. 1. (a) Overall survival for early stage patients by subtype. (b) Overall survival for advanced stage patients by subtype. (c) Disease-specific survival for early stage patients by subtype. (d) Disease-specific survival for advanced stage patients by subtype. NLP Z nodular lymphocyte-predominant Hodgkin lymphoma. patients and 86%, 82% and 80% for CHL patients (P<.01). This survival advantage among NLPHL patients was present among both early stage patients (P<.01) and advanced stage patients (P<.01). Similarly, DSS was better in NLPHL patients with early (P<.01) or late stage (P<.01) disease. In multivariate analysis (MVA) for OS and DSS, NLPHL was associated with better OS (HR: 0.62, 95% CI 0.53-0.73, P<.01) and DSS (HR: 0.48, 95% CI 0.38-0.62, P<.01). The male predominance in NLPHL (vs CHL) was most pronounced among NLPHL patients <20 years old (78%) with the male predominance decreasing with increasing age. The more favorable baseline characteristics among NLPHL (vs CHL patients) such as lower incidence of B symptoms and extranodal disease were also most pronounced in the younger NLPHL patients (Table 2). As shown in Figure 2, radiation use differed by age among early stage patients, with younger patients least likely to receive radiation (38%) compared to patients 20 to 45 years of age (59%) and patients >45 years of age (66%). Use of radiation in NLPHL declined from 70% of NLPHL patients in 1988 to 1990 to 42% in 2006 to 2010 (Fig. 3),
representing a 40% decline over this time period. This trend was observed in both early-stage patients and advancedstage patients, with a 31% relative decrease among early stage patients and a 61% relative decrease among advanced stage patients over the study period. In NLPHL patients, younger age at diagnosis, early stage, and receipt of radiation were all associated with better OS and DSS, controlling for other disease and demographic characteristics (Table 3 and see Supplementary Table E1; available online at www.redjournal.com).
Discussion Much of the data for prognosis, outcomes, and treatment strategies in NLPHL were derived from series of HL, of which NLPHL patients comprise a small percentage of the total number of studied patients. When NLPHL is examined in comparison to CHL, conflicting data emerges. The GHSG analysis of greater than 8,000 patients demonstrated improved prognosis in NLPHL compared to CHL patients (2). Yet there was no difference in FFTF between NLPHL
80
International Journal of Radiation Oncology Biology Physics
Gerber et al.
Table 2
Nodular lymphocyte-predominant Hodgkin lymphoma patient characteristics by age at diagnosis Age at diagnosis <20 (nZ186)
Characteristic Sex Male Female Race White Black Other Geographic region Midwest Northeast South West Urban/rural Metropolitan Nonmetropolitan Income quartile Lowest quartile Second quartile Third quartile Fourth quartile Stage I/II III/IV Extranodal extension Yes No B symptoms Absent Present Unknown Radiation Received Did not receive Diagnosis year 1988-1994 1995-2010
>45 (nZ441)
20-45 (nZ535)
n
%
n
%
n
%
P
146 40
78% 22%
382 153
71% 29%
278 163
63% 37%
.001
147 28 11
79% 15% 6%
376 137 22
70% 26% 4%
330 93 18
75% 21% 4%
.04
23 29 40 94
12% 16% 22% 51%
88 67 98 282
16% 13% 18% 53%
79 75 75 212
18% 17% 17% 48%
NS
165 21
89% 11%
491 44
92% 8%
400 41
91% 9%
NS
38 41 49 58
20% 22% 26% 31%
116 149 118 152
22% 28% 22% 28%
91 122 86 142
21% 28% 20% 32%
NS
146 40
78% 22%
417 118
78% 22%
318 123
72% 28%
.0687
14 172
8% 92%
64 471
12% 88%
51 390
12% 88%
NS
146 8 32
78% 4% 17%
341 70 124
64% 13% 23%
278 53 110
63% 12% 25%
<.001
65 121
35% 65%
296 239
55% 45%
203 238
46% 54%
<.0001
12 174
6% 94%
57 478
11% 89%
44 397
10% 90%
NS
Abbreviation: NS Z not significant.
and CHL patients when controlling for stage at presentation. This lack of difference may be due to the smaller number of NLPHL patients for any given stage. In this analysis of more than 30,000 patients, NLPHL patients had better OS and DSS than CHL patients. In MVA accounting for other known prognostic features such as stage at presentation, NLPHL histology was associated with better OS and DSS. Our finding of improved prognosis in NLPHL patients after controlling for stage and other prognostic features is confirmed in the recently published series using the SEER data (10). Our findings support previously published reports that NLPHL patients have more favorable characteristics and earlier stage at presentation than CHL patients (2, 11). One novel contribution of the current series is that this
difference was most pronounced among NLPHL patients <20 years old who, as compared to older patients with NLPHL, had a lower incidence of B symptoms and extranodal extension. Even after controlling for the more favorable prognostic factors among pediatric patients, age <20 years old was associated with better OS and DSS in NLPHL patients. The association of age with outcome was linear with patients 20 to 45 years of age demonstrating improved prognosis compared to patients >45 years old. The more favorable presentation and outcomes of pediatric NLPHL patients is concordant with the data for all HL patients in which pediatric patients present with more favorable disease characteristics and demonstrate better survival (8). Based on data from the British Columbia Cancer Agency in which adolescents
Volume 92 Number 1 2015
A SEER analysis of NLPHL
% of Patients Receiving Radiation
1 0.8 69%
66%
0.6 0.4
58% 60%
60%
59%
<20 38%
37%
20-45
40%
>45
0.2 0 Stage I and II
Stage I
Stage II
Fig. 2. Use of radiation in early stage nodular lymphocyte-predominant Hodgkin lymphoma. (16-21 years old) and young adults (22-45 years of age) with HL demonstrated similar outcomes (12), it is likely that the improved prognosis in the patients <20 years old in this series was driven by the youngest patients, although the fact that young adults (20-45 years of age) demonstrated improved prognosis compared to the older adults in this series provides evidence that age may function as a continuous variable with older age associated with poorer outcomes in NLPHL. Radiation is instrumental in the cure of early stage NLPHL patients. Radiation alone is the recommended treatment for patients with stage IA or IIA disease (13). Large cooperative group trials and single-institution series demonstrate excellent outcomes from radiation alone in early stage NLPHL (14-16) or radiation in combination with chemotherapy (17-19). Recent data from the GHSG demonstrate excellent outcomes with involved-field RT (IFRT) alone in stage IA patients and show no benefit with the addition of chemotherapy in early stage patients (20). In this study, receipt of radiation was associated with better OS and DSS on MVA with a HR of 0.66 for OS and 0.45 for DSS. Yet, this study also demonstrated a decline in receipt of radiation over time, a trend which mirrors the declining use of radiation in CHL (9). The declining use of radiation may
% of Patients Receiving Radiation
1 Stage I/II NLPHL Only
0.8
0.6
0.4
81
stem in large part from concern for late effects as data concerning late effects of radiation in HL survivors have emerged over the past two and a half decades. Nonradiationbased strategies including surgical resection followed by observation and observation alone have been investigated, particularly in the pediatric population (21-23), but many of these patients develop relapse and observation alone is not routinely recommended, with the exception of certain pediatric patients (3). In the recent Children’s Oncology Group protocol for pediatric patients with early-stage HL, patients are treated with upfront chemotherapy with no further treatment if a complete response was achieved (Combination Chemotherapy Followed by Radiation Therapy in Treating Young Patients With Newly Diagnosed Hodgkin’s Lymphoma; clinicaltrials.gov/NCT00302003). These differing treatment paradigms in the pediatric population as well as the concern for late effects likely explains our finding of decreased radiation utilization among patients <20 years old. Given the expression of CD20 in NLPHL, prospective studies have also investigated rituximab alone in early stage disease but despite high response rates, the relapse rate with rituximab alone is high and thus rituximab is not routinely used in early-stage disease (24-26). Although no definitive conclusions can be drawn regarding the causal association of better OS and DSS with receipt of radiation, the finding is noteworthy and the downward trend in radiation utilization is significant in light of this finding. Despite this downward trend in utilization, radiation techniques have improved and modern techniques such as involved site RT should be utilized in all patients with LPHL receiving RT to minimize treatment fields and treatment morbidity (27). This study has several limitations owing in large part to relevant prognostic and treatment data not available in the SEER database such as ESR, bulky disease, the involvement of three or more nodal regions, the use of chemotherapy, and information on relapse. Given these limitations, this study is unable to answer important questions with regard to NLPHL in comparison to CHL such as time-course and rates of relapse, and the use of chemotherapy and/or rituximab and its association with OS and DSS. It also introduces potential confounders associated with histology that cannot be accounted for in the analysis. An additional weakness of this study is that we were unable to confirm the cases classified as NLPHL with central pathology review which, given prior studies showing high rates of misclassification of NLPHL (1, 28), likely led to erroneous classification of some of our patients.
0.2
0 1988-1990
Conclusions 1991-1995
1996-2000
2001-2005
2006-2010
Year of Diagnosis
Fig. 3. Use of radiation by year in all nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) patients and those with early stage disease.
Taken together, this large population-based series demonstrates that better survival of NLPHL patients as compared to CHL patients persists even after controlling for more favorable presenting characteristics. It demonstrates an
82
International Journal of Radiation Oncology Biology Physics
Gerber et al.
Table 3 Multivariate analysis of factors associated with overall and disease-specific survival in nodular lymphocyte-predominant Hodgkin lymphoma patients Overall survival Factor Age >45 20-45 <20 Sex Female Male Race White Black and other Geographic region Northeast Midwest South West Urban/rural Metropolitan Nonmetropolitan Income quartile Lowest quartile Second quartile Third quartile Fourth quartile Stage III and IV I and II Extranodal extension No Yes B symptoms Absent Present Unknown Radiation Did not receive Received Diagnosis year 1995-2010 1988-1994
Disease-specific survival
HR, 95% CI
P
HR, 95% CI
P
Ref 0.26 (0.18-0.37) 0.09 (0.03-0.25)
<.0001
Ref 0.31 (0.18-0.52) 0.06 (0.01-0.41)
<.0001
Ref 0.88 (0.62-1.24)
NS
Ref 0.89 (0.53-1.51)
NS
Ref 1.28 (0.87-1.88)
NS
Ref 1.28 (0.73-2.24)
NS
Ref 0.91 (0.48-1.75) 0.84 (0.42-1.70) 1.34 (0.77-2.33) Ref 1.03 (0.60-1.78) Ref 0.85 (0.56-1.30) 0.67 (0.39-1.14) 0.92 (0.54-1.57)
NS
NS
NS
Ref 1.03 (0.43-2.45) 0.60 (0.21-1.71) 1.10 (0.52-2.34) Ref 1.03 (0.43-2.48) Ref 1.21 (0.64-2.31) 0.62 (0.27-1.45) 1.11 (0.49-2.47)
NS
NS
NS
Ref 0.61 (0.41-0.91)
.02
Ref 0.47 (0.27-0.82)
.01
Ref 1.43 (0.90-2.27)
NS
Ref 1.28 (0.66-2.47)
NS
Ref 1.23 (0.75-2.02) 1.15 (0.79-1.68)
NS
Ref 1.66 (0.84-3.29) 1.62 (0.93-2.83)
NS
Ref 0.64 (0.44-0.93)
.03
Ref 0.43 (0.24-0.78)
.01
Ref 1.69 (1.08-2.63)
.02
Ref 1.75 (0.87-3.53)
NS
Abbreviations: CI Z confidence interval; HR Z hazard ratio; NS Z not significant; Ref Z reference value.
association between radiation receipt and better survival in NLPHL patients while confirming a decline in radiation utilization for patients with this disease. Further research refining unique treatment strategies in both early and advanced NLPHL patients is needed.
2.
3. 4.
References 1. Diehl V, Sextro M, Franklin J, et al. Clinical presentation, course, and prognostic factors in lymphocyte-predominant Hodgkin’s disease and lymphocyte-rich classical Hodgkin’s disease: Report from the
5.
6.
European Task Force on Lymphoma Project on LymphocytePredominant Hodgkin’s Disease. J Clin Oncol 1999;17:776-783. Nogova L, Reineke T, Brillant C, et al. Lymphocyte-predominant and classical Hodgkin’s lymphoma: A comprehensive analysis from the German Hodgkin Study Group. J Clin Oncol 2008;26:434-439. Advani RH, Hoppe RT. How I treat nodular lymphocyte predominant Hodgkin lymphoma. Blood 2013;122:4182-4188. Fanale MA, Younes A. Nodular lymphocyte predominant Hodgkin’s lymphoma. Cancer Treat Res 2008;142:367-381. Allemani C, Sant M, De Angelis R, et al. Hodgkin disease survival in Europe and the U.S.: Prognostic significance of morphologic groups. Cancer 2006;107:352-360. Feugier P, Labouyrie E, Djeridane M, et al. Comparison of initial characteristics and long-term outcome of patients with lymphocytepredominant Hodgkin lymphoma and classical Hodgkin lymphoma
Volume 92 Number 1 2015
7.
8.
9.
10.
11.
12. 13.
14.
15.
16. 17.
at clinical stages IA and IIA prospectively treated by brief anthracycline-based chemotherapies plus extended high-dose irradiation. Blood 2004;104:2675-2681. Xing KH, Conners JM, Al Monsour M, et al. Advanced stage nodular lymphocyte predominant Hodgkin’s lymphoma compared to classical Hodgkin lymphoma: A matched pair outcome analysis. Blood 2014; 123:3567-3573. Bazzeh F, Rihani R, Howard S, et al. Comparing adult and pediatric Hodgkin lymphoma in the Surveillance, Epidemiology and End Results Program, 1988-2005: An analysis of 21,734 cases. Leuk Lymphoma 2010;51:2198-2207. Koshy M, Rich SE, Mahmood U, et al. The declining utilization of radiation therapy in stage I and II Hodgkin’s disease and its impact on survival and secondary malignancies. Int J Radiat Oncol Biol Phys 2012;82:619-625. Ali S, Olszewski AJ. Disparate survival and risk of secondary nonHodgkin lymphoma in histologic subtypes of Hodgkin lymphoma: A population-based study. Leuk Lymphoma 2014;55:1570-1577. Bodis S, Kraus MD, Pinkus G, et al. Clinical presentation and outcome in lymphocyte-predominant Hodgkin’s disease. J Clin Oncol 1997;15:3060-3066. Foltz LM, Song KW, Connors JM. Hodgkin’s lymphoma in adolescents. J Clin Oncol 2006;24:2520-2526. Eichenauer DA, Engert A, Dreyling M, et al. Hodgkin’s lymphoma: ESMO clinical practice guidelines for diagnosis, treatment and followup. Ann Oncol 2011;22:55-58. Chen RC, Chin MS, Ng AK, et al. Early-stage, lymphocytepredominant Hodgkin’s lymphoma: Patient outcomes from a large, single-institution series with long follow-up. J Clin Oncol 2010;28: 136-141. Wirth A, Yuen K, Barton M, et al. Long-term outcome after radiotherapy alone for lymphocyte-predominant Hodgkin lymphoma: A retrospective multicenter study of the Australasian Radiation Oncology Lymphoma Group. Cancer 2005;104:1221-1229. Schlembach P, Wilder R, Jones D, et al. Radiotherapy alone for lymphocyte-predominant Hodgkin’s disease. Cancer J 2002;8:377-383. Jackson C, Sirohi B, Cunningham D, et al. Lymphocyte-predominant Hodgkin lymphomadClinical features and treatment outcomes from a 30-year experience. Ann Oncol 2010;21:2061-2068.
A SEER analysis of NLPHL
83
18. Savage KJ, Skinnider B, Al-Mansour M, et al. Treating limited-stage nodular lymphocyte predominant Hodgkin lymphoma similarly to classical Hodgkin lymphoma with ABVD may improve outcome. Blood 2011;118:4585-4590. 19. Lazarovici J, Brice P, Oberic L. Management and outcomes of nodular lymphocyte predominant hodgkin lymphoma: A retrospective study by the lymphoma study association. Blood 2014;124:3054. 20. Eichenauer DA, Pluetschow A, Fuchs M, et al. Treatment of stage IA nodular lymphocyte-predominant hodgkin lymphoma (NLPHL): An analysis from the German Hodgkin Study Group (GHSG). Blood 2014;124:3058. 21. Murphy SB, Morgan ER, Katzenstein HM, et al. Results of little or no treatment for lymphocyte-predominant Hodgkin disease in children and adolescents. J Pediatr Hematol Oncol 2003;25:684-687. 22. Pellegrino B, Terrier-Lacombe MJ, Oberlin O, et al. Lymphocytepredominant Hodgkin’s lymphoma in children: Therapeutic abstention after initial lymph node resectiondA study of the French Society of Pediatric Oncology. J Clin Oncol 2003;21:2948-2952. 23. Mauz-Korholz C, Gorde-Grosjean S, Hasenclever D, et al. Resection alone in 58 children with limited stage, lymphocyte-predominant Hodgkin lymphoma-experience from the European network group on pediatric Hodgkin lymphoma. Cancer 2007;110:179-185. 24. Advani RH, Horning SJ, Hoppe RT, et al. Mature results of a phase II study of rituximab therapy for nodular lymphocyte-predominant Hodgkin lymphoma. J Clin Oncol 2014;32:912-918. 25. Ekstrand BC, Lucas JB, Horwitz SM, et al. Rituximab in lymphocytepredominant Hodgkin disease: results of a phase 2 trial. Blood 2003; 101:4285-4289. 26. Eichenauer DA, Fuchs M, Pluetschow A, et al. Phase 2 study of rituximab in newly diagnosed stage IA nodular lymphocyte-predominant Hodgkin lymphoma: A report from the German Hodgkin Study Group. Blood 2011;118:4363-4365. 27. Specht L, Yahalom J, Illidge T, et al. Modern radiation therapy for Hodgkin lymphoma: Field and dose guidelines from the International Lymphoma Radiation Oncology Group (ILROG). Int J Radiat Oncol Biol Phys 2014;89:854-862. 28. Pappa VI, Norton AJ, Gupta RK, et al. Nodular type of lymphocyte predominant Hodgkin’s disease. A clinical study of 50 cases. Ann Oncol 1995;6:559-565.