Clinical features and outcomes of Hodgkin’s lymphoma in Korea: Consortium for Improving Survival of Lymphoma (CISL)
- First Online:
- Cite this article as:
- Won, Y., Kwon, J.H., Lee, S.I. et al. Ann Hematol (2012) 91: 223. doi:10.1007/s00277-011-1297-x
- 193 Views
Ethnic and regional differences in the epidemiology and pathological aspects of Hodgkin’s lymphoma (HL) between Western and Asian patients may be associated with differences in clinical features and prognosis. We retrospectively analyzed the clinical and histopathological characteristics, therapeutic outcomes, and prognostic factors of 539 HL patients treated at 16 centers in Korea. We found that the incidence of histological subtypes of HL in Korea was similar to that in Western and other Asian countries. However, the incidence peaked between 16 and 30 years of age, unlike the bimodal age distribution seen in Western countries. In patients with stage I–IIA non-bulky disease, the complete response (CR) rate was similar between combined modality therapy and chemotherapy alone (93% vs. 84%, P = 0.44), and there was no difference in relapse-free survival (RFS) and overall survival (OS). Patients with stage I–II disease plus unfavorable factors and those with advanced-stage disease treated with combination chemotherapy regimens had an overall CR rate of 77%, with no difference between doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) and non-ABVD regimens (77.2% vs. 76.8%, P = 0.95). Among those patients who achieved final CR, there was no significant difference in RFS or OS between those who achieved interim CR and PR. Only the presence of B symptoms was independently predictive of a shorter RFS. Age > 45 years, Eastern Cooperative Oncology Group 2–4, and B symptoms were independent risk factors for death. Although the incidence of HL was lower in Korea than in Western countries, the distribution of morphological subtypes, treatment outcomes, and patient prognosis were similar.
Hodgkin’s lymphoma (HL) is an uncommon malignancy involving the lymph nodes and the lymphatic system, with an incidence in the USA of 2.8 per 100,000 person-years . HL has become a curable disease due to the development of advanced combined treatment modalities, with combination chemotherapy regimens showing a complete response (CR) rate of 80–90%, a 5-year relapse-free survival (RFS) rate of 70–80% and a 5-year overall survival (OS) rate of 80–90% . Compared with other chemotherapy regimens, the anthracycline-based chemotherapy regimen of doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) has become the treatment of choice due to its similar efficacy and more favorable toxicity profile [3, 4]. However, many concerns remain about the long-term toxicities of anthracycline-based regimens, including pneumonitis, infertility, and secondary malignancies. As an alternative regimen, dose-escalated BEACOPP (cyclophosphamide, adriamycin, etoposide, vincristine, bleomycin, procarbazine and prednisone) is recommended for the patients with advanced-stage HL who have risk factors .
There are ethnic and regional differences in the distribution of HL, with the highest rates observed in North America and Australia, followed by Europe and Asia . In addition, the prognosis of patients with HL varies by morphology. Apart from the modestly better prognosis of nodular lymphocyte predominant HL, there is no reliable data as to major differences of different classical HL subtypes. In a retrospective review, the highest 5-year survival rates observed in patients with the lymphocyte predominant subtype (88%) and nodular sclerosis (87%), and the lowest 5-year survival rates observed in patients with lymphocyte depletion (50%), with intermediate survival rates in patients with mixed cellularity (75%) and NOS subtypes (not otherwise specified, 70%) . Ethnic and regional differences in the epidemiology and pathological aspects of HL may contribute to differences in clinical features and prognosis between Western and Asian patients. Few studies, however, have assessed the clinical features and treatment outcomes of HL patients in East Asia including Korea. We therefore retrospectively analyzed the clinical and histopathological characteristics, therapeutic outcomes, and prognostic factors of HL in Korea.
Patients and methods
Study patients and clinical data collection
We retrospectively screened patients diagnosed with histologically confirmed HL at 16 institutions throughout South Korea. Histological subtypes were classified according to the World Health Organization (WHO) classification, with the Revised European-American Lymphoma classification used in patients diagnosed before the WHO classification was published in 2001. The patients on the regular follow-up survey in each institution between October 2009 and October 2010 were enrolled. Patients without available medical records and those aged <16 years old were excluded. Clinicopathological data retrieved from medical records included patient gender, age at the time of diagnosis, Eastern Cooperative Oncology Group (ECOG) performance status (PS), histology, Ann Arbor stage, presenting symptoms, B symptoms, and number of involved sites. The course of treatment was also reviewed, including treatment modalities, chemotherapy regimens, outcomes, complications of treatments, and dates of outcome measures. The final CR was assessed by computed tomography (CT) scan with/without positron emission tomography (PET) scan. If a residual lesion or mass was observed after the completion of treatment, the final CR was confirmed by tissue diagnosis.
The patients with stage I–II HL were classified into two groups: stage I–IIA non-bulky and stage I–II with unfavorable factors. The risk factors, such as age, B symptoms, a bulky mass, number of disease sites, and number of extranodal sites, were used for the distinction of two groups.
Treatment outcomes included response rate (RR), RFS, and OS. RFS was measured from the date of documented CR until the date that recurrent disease was objectively documented. OS was calculated from the date of diagnosis to the date of death or censored at last follow-up.
Descriptive statistics were summarized as frequencies and percentages for categorical variables and as median and range for continuous variables. Categorical variables were compared using the chi-square test or Fisher’s exact test. RFS and OS were estimated using the Kaplan–Meier method, with between-group comparisons made using log-rank tests. Multivariate analyses with a Cox proportional multiple regression model were performed to assess the impact of clinical determinants on RFS and OS. Two-sided P < 0.05 were considered significant for all end points, and 95% confidence intervals (CIs) were calculated.
Patient demographic and clinical characteristics
Patient demographic and clinical characteristics
Treatment and outcome of patients with stage I–IIA non-bulky CHL
Of the 252 patients with stage I–II CHL, 206 (81.7%) had stage I–IIA non-bulky disease. Of these, 19 (9.2%) were treated with involved-field radiation therapy (IFRT) alone, with all achieving CR. However, during follow-up, four patients (21.1%) had disease relapse. All four responded to following salvage chemotherapy and remain alive without disease or serious treatment-related complications for more than 5 years. Of the remaining 187 patients with stage I–IIA non-bulky disease, 134 were treated with chemotherapy alone. The ABVD regimen were the most commonly used in 108 patients (80.6%), followed by alternating ABVD/COPP (cyclophosphamide, vincristine, procarbazine and prednisone) in 12 patients (9.0%), modified BACOP (bleomycin, adriamycin, cyclophosphamide, vincristine and prednisone) in six patients (4.5%), C-MOPP (cyclophosphamide, mechlorethamine, vincristine, procarbazine and prednisone) in two patients (1.5%), and one other regimen in each of six patients. Following a median 6 cycles of chemotherapy, 113 (84.3%) of these patients achieved CR, 11 (8.2%) achieved partial response (PR), and five (3.7%) had progressive disease (PD). The CR rate with the ABVD regimen was 82.4% (89 patients). Although the number of patients was small, the alternating ABVD/COPP regimen had the highest CR rate, of 100.0% (12 patients).
Combined modality therapy (CMT), consisting of standard induction chemotherapy followed by radiation therapy, was used in 53 patients. After a median 4 cycles (range, 2–6 cycles) of induction chemotherapy, 37 patients (69.8%) achieved CR and 12 (22.6%) achieved PR; the latter 12 achieved CR after completing radiation therapy. Thus, the CR rate at completion of CMT was 92.5%.
Treatment and outcome of patients with stage I–II CHL with unfavorable factors and stage III–IV CHL
Among the 46 patients with stage I–II CHL plus unfavorable factors and the 254 patients with stage III–IV CHL, 12 (4.0%) were excluded from analyses of treatment outcomes due to insufficient data. Of the remaining 288 patients, all were initially treated with combination chemotherapy according to the protocol at each center at the time of treatment. Most patients (222 patients. 75.5%) were treated with the ABVD regimen, 21 (7.1%) were treated with alternating ABVD/COPP, 15 (5.1%) with C-MOPP, 14 (4.8%) with alternating C-MOPP/ABVD, six (2.0%) with modified BACOP, 4 (1.4%) with CHOP (cyclophosphamide, doxorubicin, vincristine and prednisone), four (1.4%) with COPP, and two (0.7%) with BEACOPP.
Of these 288 patients, 222 (77.1%) achieved CR and 44 (15.3%) achieved PR. Of the 219 patients who received the ABVD regimen, 169 (77.2%) achieved CR and 34 (15.5%) achieved PR. Of the 69 patients treated with non-ABVD regimens, 53 (76.8%) achieved CR and 10 achieved PR (14.5%). The CR (77.2% vs. 76.8%, P = 0.951) and objective response (92.7% vs. 91.3%, P = 0.705) rates were similar in patients who received ABVD and non-ABVD regimens. When we analyzed associations between clinical factors and CR rate, we found that a small number of extranodal sites (0 or 1) was associated with a significantly better response to first-line chemotherapy than ≥2 sites (83.6% vs. 70.7%, P = 0.011). However, age, gender, histology, ECOG PS, Ann Arbor stage, B symptom, and bulky disease were not associated with CR.
Interim response evaluation in patients with stage I–II CHL with unfavorable factors and stage III–IV CHL
Relapse-free survival and overall survival of patients with CHL
To date, 74 (14.6%) of the 506 patients with CHL have died, with median OS not yet reached. The estimated 3- and 5-year OS rates were 85.7% and 80.2%, respectively (Fig. 3d). Univariate analyses showed that male gender (P = 0.049), age (>45 vs. ≤45 years, P < 0.0001), stage (III/IV vs. I/II, P = 0.007), PS (ECOG 2–4 vs. 0/1, P < 0.0001), and B symptoms (presence vs. absence, P = 0.001) were significantly associated with a shorter OS. Multivariate analyses showed that age >45 years (HR, 3.57; 95% CI, 2.14–5.94; P < 0.0001), ECOG PS of 2–4 (HR, 3.04; 95% CI, 1.65–5.59; P < 0.0001), and B symptoms (HR, 1.71; 95% CI, 1.05–2.80; P = 0.032) were independent risk factors for death.
Clinical aspects and treatment outcome of nodular lymphocyte predominant HL and comparison with CHL
The comparison between patients with nodular lymphocyte predominant (NLP) HL and CHL is summarized in Table 1. Except for the incidence of palpable mass among presenting symptoms, there were no significant differences in the characteristics of patients with NLP HL or CHL. The median age of patients with NLP HL and CHL were 41 years (range, 20–81 years) and 40 years (range, 16–88 years), respectively, their distributions of age groups were similar, and 33.3% and 31.0% of patients with NLP HL and CHL, respectively, were aged 16–30 years.
Of the 33 patients with NLP HL, 17 had stage I-IIA non-bulky disease. Of these, six patients were treated with IFRT alone, with all achieving CR. These patients have remained alive without disease relapse or serious complications for more than 5 years. Of the remaining patients, six were treated with CMT and five with chemotherapy alone. On the other hand, three patients had stage I–II disease with unfavorable risk factors, with two having bulky mediastinal masses. Thirteen patients had advanced-stage NLP HL at diagnosis. These 16 patients were treated with various first-line combination chemotherapy regimens.
Of the 27 patients treated with either chemotherapy or CMT, ABVD (23 patients, 85.2%) was the most common regimen. We found that 26 patients (96.3%) achieved CR, with the remaining patient having PD after completion of 4 cycles of BEACOPP. Of the 32 NLP HL patients who achieved CR, nine patients (28.1%) experienced disease recurrence at a median follow-up of 39.6 months. We could not find any prognostic factor for RFS or OS, and there were no differences in RFS (HR, 1.28; 95% CI, 0.64–2.56; P = 0.48) and OS (HR, 1.16; 95% CI, 0.54–2.53; P = 0.70) between patients with NLP HL and those with CHL (Fig. 3c, d).
Toxicities of treatment and causes of death
Of the 472 patients treated with bleomycin-containing regimens, 16 (3.4%) experienced non-infectious pneumonitis associated with bleomycin, with 10 (62.5%) showing this complication within 12 months (range, 1–114 months) and four (25.0%) dying due to pneumonitis-related causes. Three patients in CR developed non-Hodgkin’s lymphoma during follow-up. To date, 81 of the 539 patients have died. The most common cause of death was disease progression (32 patients), followed by infections (12 patients), treatment-related complications (five patients), stroke (one patient), and gastrointestinal bleeding (one patient). We could not identify the cause of death for the remaining 30 patients because they were lost to follow-up.
In Korea, the age-adjusted rate of HL has been reported to be 0.4 per 100,000 person-years, with HL constituting approximately 6% of all lymphomas . In contrast, the rate of HL in the USA is 2.8 per 100,000 person-years, with HL constituting 11.7% of all lymphomas [1, 9]. The relatively low incidence of HL in Korea, however, is in good agreement with findings in other Asian countries [10, 11]. For example, in Taiwan, the age-adjusted rate of HL from 1979 to 2002 was 0.21–0.60 per 100,000 person-years , and in Japan, HL was reported to constitute 7% of all malignant lymphomas .
Incidence pattern of histological subtypes in Korea, Taiwan, Japan, EU, and USA
We found that young adults (aged 16–30 years) were more affected than other age groups, with NS CHL especially occurring more frequently in this group. However, we did not observe a bimodal age distribution in Korean HL patients, in contrast to HL patients in the USA, where CHL peaked between the ages of 15 and 30 years, and again during the sixth decade of life. The early peak is caused primarily by the high incidence of the NS subtype of HL in younger patients . We also found that the proportion of female was higher in patients with NS histology than in other subtypes, a finding similar to results showing that NS CHL was the only subtype in which there was a slight female predominance .
Previously, radiotherapy alone was the standard therapy for patients with early-stage HL with favorable characteristics . Many subsequent studies have shown, however, that CMT, consisting of a short course of chemotherapy followed by radiation therapy, was associated with similar or superior efficacy outcomes [19, 20]. Recently, CMT using 2 cycles of ABVD followed by 20 Gy IFRT is being regarded the standard of care in early favorable HL patients based on the German Hodgkin Study Group (GHSG) HD10 study . Chemotherapy alone has also been investigated in patients with early-stage HL with favorable characteristics. For example, 6 cycles of ABVD were reported to be effective in patients with limited stage, non-bulky disease . At present, chemotherapy regimens for the advanced-stage HL patients are used to treat patients with early-stage disease . In our study, the majority (65%) of patients with stage I–IIA non-bulky disease were treated with chemotherapy alone, whereas 26% received CMT. Analysis showed that the CR rates of the chemotherapy alone and CMT groups were similar (84.3% vs. 92.5%, P = 0.444). Moreover, CMT did not show superior RFS or OS compared with chemotherapy alone. These findings are in agreement with the results of a randomized phase III trial that showed no significant differences in CR duration or OS between patients treated with CMT (ABVD followed by radiotherapy) and those treated with ABVD alone . However, as shown in the Fig. 1, the RFS curve of chemotherapy alone is under that of CMT. In addition, a randomized trial demonstrated a slight advantage of the CMT approach versus chemotherapy alone in terms of RFS . A recent review demonstrated that CMT improves tumor control and OS in patients with early-stage HL .
In our study, all patients with early-stage CHL plus unfavorable factors and those with advanced-stage disease were initially treated with various combination chemotherapy regimens, although ABVD was the most commonly used. Among these patients, the overall CR rate was 77%, and the objective RR was 92%, with no difference in the latter between ABVD and non-ABVD regimens (P = 0.71). These results were consistent with those of previous studies from Western countries. For example, a large intergroup study comparing the ABVD and MOPP/ABV regimens in patients with advanced HL found that the CR rates were similar (76% vs. 80%) . Among the advanced-stage HL patients treated with ABVD or ABVD-based alternating regimen, the tumor control at 5 years ranges between 61% and 66% in multicenter studies with an OS of 73% to 83% [26–28]. In other words, between 30% and 40% of advanced-stage HL patients treated with these regimens experience treatment failure after first-line treatment. In order to overcome these limitations, the GHSG developed the BEACOPP regimen. BEACOPP regimen has been introduced in three different prospective clinical trials of the GHSG . In the HD9 trial of the GHSG with 1,196 advanced-stage HL patients, median follow-up was 111 months. At 10 years, freedom from treatment failure was 64%, 70%, and 82% with OS rates of 75%, 80%, and 86% for patients treated with COPP/ABVD, BEACOPP baseline, and escalated BEACOPP, respectively. As an alternative regimen, dose-escalated BEACOPP is recommended for the patients with advanced-stage HL who have risk factors . Unfortunately, only three patients received BEACOPP regimen as a first-line treatment in our study. Therefore, reliable analyses for these patients could not be done.
Previous studies have reported that a mediastinal bulky mass, B symptoms, numerous sites of disease, and significantly elevated ESR (erythrocyte sedimentation rate) were unfavorable prognostic factors in patients with early-stage HL [30–33]. In patients with advanced HL, prognostic factors include advanced age, gender, stage, B symptoms, number of nodal sites, albumin, hemoglobin (Hb), white cell count, and lymphocyte count [34, 35]. In the present study, however, we could not collect complete data having prognostic significance, such as white blood cell count, ESR, serum albumin, and the presence of a bulky mediastinal mass. In particular, the presence of a bulky disease in some patients was unknown. As a result, there was a chance that the number of bulky disease was lower than expected. Although the previously identified prognostic factors were not evaluated completely in our study, we found that B symptoms were associated with a shorter RFS and that advanced age (≥45 years), poor ECOG PS (2–4), and B symptoms were associated with a shorter OS.
Because NLP HL and CHL differ in natural history and response to therapy, their treatment should differ. Patients with NLP HL should receive less intensive therapy than patients with CHL, thus avoiding the toxic complications that contribute to overall mortality. In adults with early-stage, favorable NLP HL, radiotherapy may be curative, with recent clinical trials suggesting that IFRT is effective [36, 37]. On the other hand, one way to possibly reduce the amount of chemotherapy would be the addition of rituximab. The use of the monoclonal anti-CD20 antibody rituximab has been studied in several recent trials. Collectively, the data of these studies suggested that rituximab is an active agent in NLP HL both in the upfront and relapsed setting. In addition, rituximab was associated with fewer long-term adverse events. Therefore, rituximab is an effective alternative therapy for NLP HL .
In contrast, patients with advanced NLP HL or unfavorable prognostic factors should receive treatment similar to that of patients with CHL . Although the number of NLP HL patients in our analysis was small, we found that patients with stage I–IIA non-bulky disease showed good outcomes with IFRT alone, while those with early-stage disease and unfavorable factors and those with advanced-stage disease showed higher responses to CMT or standard chemotherapy. Although the clinical characteristics of patients with CHL and NLP HL did not differ significantly, unfavorable prognostic factors (ECOG PS of 2–4, advanced stage, and B symptoms) tended to be less common in patients with NLP HL. Nevertheless, RFS (HR, 1.28; 95% CI, 0.64–2.56; P = 0.48) and OS rates (HR, 1.16; 95% CI, 0.54–2.53; P = 0.70) were similar for patients with NLP HL and those with CHL. In contrast, several studies reported that NLP HL patients have a better prognosis with OS of 80–90%, compared with those with CHL [39–42].
This is the first multicenter analysis of the clinical characteristics, treatment practices, and outcomes of 539 HL patients treated in Korea, and this large-scale data can be used as the representative of Asian HL. Although we found that the incidence of HL in Korea was lower than in Western countries, the characteristics and treatment outcomes were similar.
This study was supported by a grant of the Korea Healthcare technology R&D Project, Ministry of Health and Welfare, Republic of Korea (A062260).
Conflict of interest
The authors declare that there was no conflict of interest.