A survey of the medical records at Asan Medical Center (Seoul, Korea) identified 65 patients who met the WHO classification between February 1995 and June 2006.2 We excluded (1) 6 patients under 15 years of age, (2) 16 with primary cutaneous ALCL, (3) 1 with secondary systemic ALCL from Hodgkin's disease, (4) 2 who died before receiving any induction therapy including chemotherapy, radiation, and resection, (5) 4 whose medical records were insufficient to determine their clinical features and outcomes. Thus, our patient cohort consisted of 36 patients with adult primary systemic ALCL.

The medical records of these 36 adult patients with primary systemic ALCL were retrospectively reviewed. Diagnoses were based on a combination of routine histology, immunohistochemistry, cytogenetics, molecular genetics, and/or flow cytometry. The morphology was reviewed in all 36 cases. The available cases were tested for CD3, CD4, CD8, CD20, CD30, CD79a, ALK-1, UCHL-1, and PAX-5 to exclude B-cell lineage. Immunohistochemical analysis of ALK expression was performed on performed on 4 µm-thick paraffin sections taken from paraffin blocks of tissue samples from 30 patients. The biopsies of the remaining 6 patients were processed at outside institutions or obtained too long ago, making their paraffin blocks unavailable. Using the standard avidin-biotin-peroxidase technique, immunostaining with mouse monoclonal antibody to ALK (1 : 25; DAKO Diagnostics, Glousrup, Denmark) was performed in the automated immunostainer after heat-induced antigen retrieval protocols (Ventana ES automated immunohistochemistry stainer, Tucson, AZ, USA). Hyperplastic tonsils and a known case of ALK+ ALCL were used as negative and positive controls, respectively.

The clinical features evaluated for potential prognostic importance were gender, age, Ann Arbor stage, B symptoms, performance status, type and number of extranodal sites, and serum lactate dehydrogenase (LDH) level. Performance status was assessed according to the Eastern Cooperative Oncology Group (ECOG) scale.

Thirty-five patients (97%) received induction chemotherapy. One patient did not receive any chemotherapy and radiotherapy after complete surgical excision of localized cervical lymphadenopathy and had been in continuous CR for 17 months. Of 35 patients treated with chemotherapy, 31 (89%) received CHOP regimen; 1 received CHOP regimen and cytarabine together; 1 received regimens that included cyclophosphamide, etoposide, vincristine, bleomycin, methotrexate, and prednisolone plus or minus cytarabine (Vanderbilt);11 1 patient received vincristine, methotrexate, dexamethasone, ifosfamide, etoposide, cytarabine; vincristine, methotrexate, dexamethasone, cyclophosphamide, and doxorubicin in addition to intrathecal injection of methotrexate, cytarabine, and hydrocortisone (BNHL); and 1 received other regimens. Response was assessed according to IWC criteria.12

Overall survival (OS) was measured from diagnosis to death from any cause, with surviving patient follow up censored at the last contact date. EFS was defined as the time from the first therapy to the first occurrence of relapse, progression, or death from any cause. Follow up of patients not experiencing any of these events was censored at their date of last contact. To compare the clinical characteristics between ALK+ and ALK ALCL, independent t test was used for age and Chi-square test was used for the other factors. OS and EFS distributions were calculated using Kaplan and Meier methods. Survival curves were compared by log-rank or Breslow test according to the presence of cross line on the survival curve. Tests for comparison were regarded as significant if the p value was less than 0.05.

The demographic and clinical characteristics of the 36 patients are shown in Table 1. Of the 36 patients, 29 (81%) were male. The median age was 39 years (range, 17 - 67 years). Five patients (14%) were classified as Ann Arbor stage I, 5 (14%) as stage II, 11 (30%) as stage III, and 15 (42%) as stage IV. Lymphadenopathy was present in 30 patients (83%) at diagnosis: intraabdominal and retroperitoneal lymphadenopathy in 19 (63%), and mediastinal lymphadenopathy in 11 (37%). Splenomegaly was observed in 8 patients (22%). Twenty-six patients (72%) presented with extranodal involvement and 15 (42%) showed involvement at 2 or more extranodal sites. Six patients (17%) were assessed as having primary extranodal lymphoma without any nodal involvement. Bone (n = 7, 19%) and soft tissue (n = 6, 17%) were the most frequently involved sites, followed by the lung, pleura, muscle, bone marrow, and liver (n = 5, each). Soft tissue sites included mesenteric, peritoneal, and retroperitoneal soft tissue masses. Involvement of the gastrointestinal system including the stomach (n = 2) and small bowel (n = 3) was observed in 5 patients (14%). Thirty-two patients (89%) were scored as ECOG scale 1 or 2 and 4 (11%) as ECOG scale 3 or 4. B symptoms occurred in 21 patients (58%), and serum LDH was elevated in 25 (70%). Twenty-two patients (61%) were classified as low and low-intermediate IPI score and 14 (39%) as high-intermediate and high IPI score. However, 24 patients (66%) were scored as high-intermediate and high age-adjusted IPI.

ALK staining was performed again in 30 patients (83%) by a pathologist because the biopsy specimens from 6 patients were unavailable. We defined the cases, showing any nuclear or cytoplasmic positivity, as ALK+ ALCL. Of these 30 patients, 13 (43%) were ALK+ and 17 (57%) were ALK (Table 2). The median age of ALK+ patients (33 years; range, 19 - 49 years) was significantly lower than that of ALK patients (43 years; range, 17 - 67 years) (p = 0.047) but male to female ratio (5.5 vs. 3.3, p = 0.469), percent of patients with advanced stage (III, IV) disease (85 vs. 59%, p = 0.130), and percent of patients with extranodal involvement (77 vs. 76%) did not differ significantly. Eight ALK+ patients (62%) and 11 ALK- (65%) were low and low-intermediate IPI (p =0.861), and 4 and 7, respectively, were low and low-intermediate age-adjusted IPI (p = 0.421). As a whole, 8 survived and 5 died out of 13 ALK+ patients and 8 survived and 9 died out of 17 ALK patients.

Thirty-five patients (97%) were treated with combination chemotherapy regimens. The overall response rate was 91% with CR in 51% (n = 18) and PR in 40% (n = 14). Thirty-two patients (89%) received CHOP-based chemotherapy regimen. The overall response rate of CHOP-based chemotherapy regimen was 91% with CR in 50% (n = 16) and PR in 41% (n = 13). Two progressive disease (PD) and 1 unknown response were noted. Eighteen of 35 patients (50%) who received chemotherapy relapsed; 15 patients within 12 months of diagnosis, with the others showing relapse after 32, 40, and 52 months, respectively. In 7 patients, the relapse sites included previous disease sites whereas in 5, recurrences were at both previous and new sites and 4 relapsed only at new sites. Relapse sites of the remaining 2 patients were not reported.

Seventeen of 18 relapsed patients received second-line chemotherapy: 4 received cytarabine, cisplatin, and dexamethasone (DHAP); 3 received Vanderbilt;11 3 received etoposide, methylprednisolone, cytarabine, and cisplatin (ESHAP); and 7 received other regimens including CDDP plus taxol and T-LBL. It showed 2 CRs (12%); 3, PRs (18%); 1, stable disease (SD) (6%); and 8, PDs (47%), respectively, and the medical records of 3 patients were insufficient to assess the clinical responses of second-line chemotherapy.

One patient who relapsed after first-line chemotherapy and 2 after second-line chemotherapy were treated with high-dose chemotherapy, followed by autologous peripheral blood stem cell transplantation. One patient has remained in continuous CR for 8 years and 1 had a 23-month response whereas the third had a transient response with 2-month duration from relapse to death.

The median follow-up period was 47 months. OS rate was 57% at 2 years and 45% at 5 years, while EFS rate was 45% at 2 years and 15% at 5 years (Fig. 1). Median OS and EFS were 49 and 17 months, respectively.

Univariate analysis showed that performance (p = 0.035), IPI (p = 0.025), and age-adjusted IPI (p = 0.034) were significantly related to OS (Table 3 and Fig. 2). The survival rates were 61.5 and 20.0% in ALK+ and ALK ALCL groups, respectively. However, OS did not differ significantly between ALK+ and ALK groups (p = 0.483) (Fig. 3).

Sixteen patients (44%) died during the study period, 8 from progression of ALCL including superior vena cava syndrome, venoocclusive disease proven by autopsy and acute respiratory distress syndrome (ARDS) due to lung metastasis. Six patients died of infection and 2 died of unknown causes. Among the 6 patients who died of infection, 4 were diagnosed as sepsis or septic shock with neutropenia, 1, biliary sepsis due to hepatic infiltration of lymphoma, and the other, ARDS resulting from pneumonia. The parents or responsible individuals of some patients agreed to DNR (Do Not Resuscitate), therefore, some patients were discharged to the primary hospital with no hope of recovery.