This study included 216 HL patients who were treated between 1991 and 2004 at the University Hospital of Salamanca (Salamanca, Spain) and the Puerta de Hierro University Hospital (Madrid, Spain). The local Ethics Committees of the participating centers approved the study protocol, taking into consideration the existing regulations at that time. The patient characteristics are summarized in Table I. The patients included 117 (54.2%) men and 99 (45.8%) women, with a median age of 37 years (range, 11–80 years). All the patients had histological evidence of HL confirmed by hematopathologists at the treating hospital; 148 patients (68.5%) had nodular sclerosis, 53 (24.5%) mixed-cellularity and 7 (3.3%) lymphocyte-predominant types. A total of 79 patients (36.6%) presented with B sympthoms at diagnosis and 144 (67.0%) were diagnosed with stage I or II disease. As regards the extension of the lymphoma, 69.0% of the patients did not exhibit bulky disease, whereas only 35 patients (16.0%) exhibited extranodal disease.

The treatment details are described in Table II. Considering the 144 patients with early-stage disease who completed the induction therapy, 27 (18.7%) received RT alone, 82 (57.0%) 3 cycles of doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) followed by RT and 35 (24.3%) received mechlorethamine, vincristine, procarbazine and prednisone (MOPP) followed by RT. Among the 72 advanced-stage patients, 58 (80.6%) received 3–8 cycles of ABVD and 13 (18.0%) 6–8 cycles of MOPP and components of the ABVD regimen without dacarbazine (MOPP/ABV regimen). One patient received methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisone and bleomycin (MACOP-B regimen). Finally, 55 patients (76.4%) received consolidation treatment with RT following chemotherapy.

Patients were considered eligible for the study when they had a ⁶⁷Ga-avid lymphoma established by a baseline gallium scan prior to treatment and a gallium scan for restaging performed midway through chemotherapy (114 patients, 52.7%) and/or at the end of the first-line treatment (204 patients, 94.4%). This cohort of patients were treated before PET was available at our institution; therefore, none of the patients underwent PET. Four patients (1.8%) underwent mid-chemotherapy gallium scan alone, 94 patients (43.5%) underwent end-first-line treatment gallium scan alone and 110 patients (51.0%) underwent both examinations. Apart from one, all patients with a positive restaging gallium scan at the end of treatment received additional chemotherapy or underwent high-dose therapy with stem-cell rescue.

Clinical response following treatment was determined based on the computed tomography (CT) scan or physical examination findings at the end of chemotherapy or RT, using the guidelines developed by the International Workshop on Response Criteria for non-HL (18). A complete response (CR) was defined as negative physical examination and residual disease of ≤1 cm; an unconfirmed CR (uCR) was defined as >75% reduction in tumor volume, with residual disease of ≤2 cm; a partial response (PR) was defined as >50% reduction in tumor volume; stable disease was defined as <50% reduction in tumor volume; and progressive disease was defined as an increase in the size of initial disease or development of new foci. Relapsed disease was defined as further disease progression or appearance of new disease foci after a response had been achieved with the initial course of treatment.

For gallium scan studies, the patients received 7–10 mCi (259–370 MBq) intravenous injection of ⁶⁷Ga citrate and imaging was performed at 72 h. Whole-body anterior and posterior planar images were acquired with a dual-head camera attached to a medium-energy collimator and 3 energy peaks of 93, 184 and 296 keV (20% window), using a 256×768 matrix and a scanning speed to achieve an information density of >1,500,000 counts for each view. In addition, single-photon emission CT (SPECT) of the chest, abdomen and pelvis was performed with the same camera at 3° intervals and 25 sec/stop using 64×64 matrix. Reconstruction of the SPECT data was performed using a low pass pre-filter (order 5.0, cut-off 0.31) and transverse reconstruction with a ramp filter. Attenuation correction was not applied and the files were reconstructed in transverse, sagittal and coronal planes. Additional delayed images were obtained in selected patients in whom the SPECT images failed to differentiate between normal and abnormal bowel activity. Nuclear medicine physicians did not have access to the CT images or results when evaluating the gallium scans. The gallium scan was considered positive for the presence of disease when abnormal focal or diffuse uptake of ⁶⁷Ga was observed in a localization incompatible with normal structures and unexplained by other causes. Diffuse lung uptake (19) and bilateral perihilar uptake (20) were not considered to indicate the presence of lymphoma.

In order to identify the factors that affected outcome, the following clinical and biological characteristics were evaluated in univariate and multivariate analyses: Gender, age, Eastern Cooperative Oncology Group performance status, Ann Arbor stage, size of tumor, bulky disease, presence of B symptoms, bone marrow involvement, number of extranodal sites, erythrocyte sedimentation rate (ESR), lactate dehydrogenase, β2-microglobulin, copper and ceruloplasmin levels, number of leukocytes and lymphocytes and hemoglobin levels. Finally, CT and gallium scan results after 3 cycles of treatment and at the end of induction therapy were included.

Comparison of response rates was performed using the χ² test. The patients were analyzed with respect to event-free survival (EFS) and overall survival (OS). OS was calculated from the day of diagnosis to the day of death or last follow-up. EFS was calculated from the day of diagnosis to the day of treatment failure (relapse, progression or no response to induction therapy). Probabilities were estimated using the Kaplan-Meier method. Differences between groups according to the different covariates were analyzed by the log-rank test. A forward stepwise Cox proportional hazards regression model was used for multivariate analysis. All the P-values reported were two-sided and P<0.05 was considered to indicate statistically significant differences.

The median follow-up time for the 216 patients was 107 months. The 5-year Kaplan-Meier OS and EFS estimates were 91 and 82%, respectively. The corresponding OS and EFS estimates for the 144 patients with stage I–II disease at diagnosis were 92 and 87%, respectively, whereas the corresponding OS and EFS for the 72 patients with stage III–IV disease were 86 and 77%, respectively.

The results of restaging gallium scanning are presented in Table III. Among the 114 patients who underwent a mid-chemotherapy gallium scan, 12 (10.5%) had a positive result. For the 204 patients who underwent a post-chemotherapy gallium scan, the results were positive in 10 (4.9%). Among the 110 patients who underwent both a mid- and a post-chemotherapy gallium scan, 5 patients (4.6%) had a positive scan midway through treatment and became negative after treatment; 2 patients (1.8%) had a negative gallium scan at mid-chemotherapy and became positive at the end of chemotherapy; and in 5 patients (4.6%), the results were positive in both mid- and post-chemotherapy gallium scans. The remaining 98 patients (89%), had negative results at both time points.

Based on the CT scan or physical examination findings, 206 patients (94%) achieved complete or uCR after first-line treatment, 4 patients (2%) had PR and 6 patients (4%) had progressive disease, without any patient achieving stable disease. All 10 patients with a positive gallium scan at the end of first-line treatment had progressive disease, 80% of whom succumbed due to disease progression and 20% were rescued with treatment. Of the 194 patients with negative gallium scan at the end of treatment, 23 (11.8%) suffered an event, whereas 141 of them exhibited CR (72.6%) and 53 uCR (27.3%).

Of the 216 patients included in the series, 16 presented a positive gallium scan result and treatment was changed in 4 (25%) patients after 2 cycles of chemotherapy. The 4 patients were rescued with chemotherapy and ASCT. Another 3 patients (18.7%) continued with the first-line treatment despite the gallium scan findings at mid-treatment, with repeat positive results at the end of first-line treatment, and received rescue chemotherapy followed by TASPE. Of these 7 patients, 3 remained alive and disease-free at the last follow-up (2 of whom had positive gallium scan results at the end of first-line treatment). The remaining 4 patients succumbed to progressive lymphoma. Of the 9 patients with positive gallium scans and without change of treatment, 6 patients succumbed, 2 of whom due to secondary neoplasms. Of the remaining 3 cases, 1 should be considered a false-positive result secondary to RT, the second patient suffered a late relapse and remained lymphoma-free following conventional rescue treatment plus consolidation with RT, whereas the third patient became negative (after a mid-treatment positive result) following 4 more chemotherapy cycles and consolidation RT of bulky mediastinal disease, and remained disease-free at the last follow-up.

Univariate analysis of stage I–II patients. The significant variables in the univariate analysis are listed in Table IV. For stages I–II, the analysis was restricted to 144 patients. The variables included in the analysis were gender, age at diagnosis, bulky mediastinal involvement, B symptoms, ESR, post-chemotherapy status, and gallium scan at mid-treatment and at the end of first-line treatment.

The factors found to significantly affect the EFS were bulky mediastinal involvement (yes vs. no, 98 vs. 79%, respectively; P=0.01), ESR [good vs. adverse (>50 without or >30 with B symptoms), 98 vs. 88%, respectively; P=0.03], presence of B symptoms (no vs. yes, 94 vs. 78%, respectively; P=0.006), post-chemotherapy disease status (CR vs. uCR vs. PR vs. PGR, 95 vs. 90 vs. 87 vs. 0%, respectively; P<0.01) and gallium scan results at mid-treatment and at the end of treatment (negative vs. positive, 88 vs. 20%, P<0.001; and 85 vs. 10%, P<0.001, respectively). As regards OS, the factors with statistical significance included age (<50 vs. ≥50 years, 96 vs. 78%, respectively; P=0.01), presence of B symptoms (no vs. yes, 97 vs. 87%, respectively; P=0.03), post-chemotherapy disease status (CR vs. uCR vs. PR vs. PGR, 95 vs. 90 vs. 90 vs. 0%, respectively; P<0.01) and gallium scan results at mid-treatment and at the end of treatment (negative vs. positive, 87 vs. 60%, P<0.001; and 95 vs. 0%, P<0.001, respectively).

For stage III–IV disease, the analysis was restricted to 72 patients and the results are listed in Table IV. The variables included in the analysis were gender, age at diagnosis, Hassenclever index classified into two groups according to risk (1–3 vs. 4–6), and gallium scan at mid-treatment and at the end of first-line treatment.

The factors found to significantly affect the EFS were Hassenclever index (1–3 vs. 4–6, 80 vs. 57%, respectively; P=0.05) and gallium scan results at mid-treatment and at the end of treatment (negative vs. positive, 84 vs. 18%, P<0.001; and 84 vs. 0%, P<0.001, respectively). Age at diagnosis (<50 vs. ≥50 years, 92 vs. 78%, respectively; P=0.04), Hassenclever index (1–3 vs. 4–6, 86 vs. 61%, respectively; P=0.04) and gallium scan results at mid-treatment and at the end of treatment (negative vs. positive, 98 vs. 40%, P<0.001; and 97 vs. 23%, P<0.001, respectively) were the factors significantly affecting OS.

Considering patients with stage I–II disease, only gallium scan results at the end of first-line treatment retained statistical significance in the Cox regression model [P<0.001, hazard ratio (HR)=2.1 and 95% confidence interval (CI): 1.91–7.9 for EFS; and P=0.03, HR=1.9 and 95% CI: 1.08–5.8 for OS]. Similarly, when considering stage III–IV patients, gallium scan at the end of treatment retained statistical significance (P=0.028, HR=3 and 95% CI: 1.07–3.4 for EFS; and P=0.003, HR=4 and 95% CI: 1.45–6.07 for OS) (data not shown).

A total of 22 patients had succumbed to the disease at the time of the analysis. Of these, 9 patients succumbed to progressive HL, 7 secondary to infectious complications (3 of whom were under treatment with high-dose chemotherapy and stem cell transplantation), and 1 patient secondary to the bleomycin pulmonary toxicity. The remaining 5 patients developed secondary neoplasms (4 solid tumors and 1 myelodysplastic syndrome with a high-grade International Prognostic Scoring System score).