Treatment of patients with advanced or bulky Hodgkin disease with a 12-week doxorubicin, bleomycin, vinblastine, and dacarbazine-like chemotherapy regimen followed by extended-field, full-dose radiotherapy

Long-term results of the Groupe Ouest et Est des Leucémies et Autres Maladies de Sang H90-a/b multicenter randomized trial

Authors


Abstract

BACKGROUND

This Phase II study was performed in patients with advanced or bulky Hodgkin disease (HD) to evaluate the results of a 7-drug chemotherapy (CT) regimen that was administered over 12 weeks according to 2 randomized modalities followed by high-dose lymph node irradiation.

METHODS

From 1990 to 1996, 162 patients with HD at clinical stages (CS) I–III with bulky disease (mediastinal mass ratio ≥ 0.45 and/or unilateral or bilateral pelvic plus lumboaortic disease; 86 patients) or CS IV (76 patients) were randomized to receive the same cumulated dose of a CT regimen consisting of epirubicin (240 mg/m2), bleomycin (60 mg/m2), vinblastine (20 mg/m2), vincristine (4 mg/m2), cyclophosphamide (4000 mg/m2), etoposide (900 mg/m2), and methotrexate (180 mg/m2) plus methylprednisolone (1500 mg/m2) over 12 weeks either every 4 weeks (Arm Y, 79 patients) or every 3 weeks (Arm Z, 83 patients). Patients with disease in complete remission (CR) or partial remission after CT received extended-field lymph node irradiation (involved areas, 40 grays [Gy]; noninvolved areas, 30 Gy).

RESULTS

Forty-two percent of patients achieved a post-CT CR, and 86% of patients achieved a CR after the completion of irradiation (there was no difference between Arm Y and Arm Z). Thirty-five patients developed recurrent disease; most of those patients were in post-CT partial remission. The 10-year freedom from first progression rate was 63.9% (there was no difference between Arm Y and Arm Z). Thirty-eight patients died: 24 patients from HD, 3 patients from CT-related early sepsis, 1 patient from radiation-induced pneumonitis, 6 patients from a second malignancy, and 4 patients from causes unrelated to treatment. The overall 10-year survival rate was 76.7%. Survival was slightly higher among patients in Arm Y (83.3%) compared with patients in Arm Z (70.2%; P = 0.12).

CONCLUSIONS

No differences were found when the same amount of CT was delivered in three courses or in four courses. In 1997, because most recurrences of the H90-A/B trial occurred in patients who achieved a post-CT partial remission, the authors decided to reinforce the intensity of the initial CT and designed a new randomized study comparing two modalities of more intensive CT plus consolidative radiotherapy (H97-LM trial). Cancer 2002;95:2169–79. © 2002 American Cancer Society.

DOI 10.1002/cncr.10932

The curative role of multiagent chemotherapy (CT) for the treatment of patients with advanced-stage Hodgkin disease was demonstrated decades ago, after the introduction by DeVita et al. in 1964 of the mechlorethamine, vincristine, procarbazine, and prednisone (MOPP) combination1 and after identification of the doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) regimen 10 years later (1974) by Bonadonna et al.2 It is now agreed that ≥ 6 monthly cycles of ABVD alone or hybridized with MOPP allow 60–70% of patients with advanced-stage Hodgkin disease to enjoy long-term remission.3, 4

Conversely, the use of lymph node irradiation given to patients with advanced disease who achieve a response to CT has a strong rationale in terms of disease sterilization.5–8 It relies on the observation that the privileged sites of recurrence after CT (including recurrences among patients with clinical Stage IV [CS IV] disease) are the lymph node areas that are affected initially by disease.9–13

Since 1981, our policy has been to combine a 12-week CT regimen with full-dose irradiation for the treatment of patients with all stages of Hodgkin disease, including patients with CS IV disease. In the Paris-Ouest-France 81 (POF81) prospective (but nonrandomized) trial (inclusion period, October, 1981 through September, 1988), all patients with CS I and CS II disease (POF81-12; 274 patients) and patients with CS III and CS IV disease (POF81-34; 133 patients) received 3 cycles of ABVD. All patients who achieved partial remission (PR) or complete remission (CR) then received extended-field irradiation (40 grays [Gy] for initially involved sites; 30 Gy for adjacent fields) plus 18–25 Gy, according to local tolerance, on CS IV-associated visceral areas.14–16

When we started planning the H90 trials, we decided to design a specific protocol for treating high-risk patients. On the basis of the 5-year results of the POF81 protocol, we identified a group of 141 patients with either CS IV disease or CS I–III disease but with bulky mediastinal disease, defined as a mediastinal mass ratio ≥ 0.45, and/or bulky infradiaphragmatic disease, defined as the involvement of part or all the lumboaortic area plus the presence of disease in one or both of the pelvic or inguinal areas. The CR rates, 5-year freedom from progression (FFP) rates, and overall survival rates for the group of 141 patients were much lower (P < 0.001) compared with the rates for the 266 patients with early/intermediate Hodgkin disease that included patients with CS I–III nonbulky disease (as defined above) (Table 1).

Table 1. Five-Year and 10-Year Results of the Prospective Multicenter Paris-Ouest-France 81 Trial that Included 407 Patients with Clinical Stage IA–IVB Hodgkin Diseasea
Clinical stageNo. of patientsCR rate (%)FFP rate (%)OS rate (%)
After CTAfter RTFive yrsTen yrsFive yrsTen yrs
  • CR: complete remission; FFP: freedom from progression; OS: overall survival; CT: chemotherapy; RT: radiotherapy.

  • a

    Patients were reclassified with early/intermediate disease (clinical Stage I–III nonbulky disease) or with advanced/bulky disease (clinical Stage I–III bulky disease and clinical Stage IV disease).

  • b

    Chi-square test.

  • c

    Log-rank test.

IA–IVB40773.794.685.383.288.783.7
I–III nonbulky26683.899.295.192.795.992.1
I–III bulky and IV14154.685.866.464.975.167.4
P value<0.0001b<0.0001b< 0.0001c<0.0001c

This Phase II, randomized H90-A/B trial was designed for a group of high-risk patients with advanced and/or bulky disease. For this trial, we decided to maintain the same strategy that was used in the POF81 trial, i.e., a brief (12-week) CT regimen followed by radiotherapy (RT). However, instead of giving 3 monthly courses of ABVD, we decided to evaluate the efficacy and toxicity of a new combination of 7 drugs given over 12 weeks, according to two randomized modalities; 3 courses were given over 12 weeks in Arm Y, and 4 courses of the same CT were given over the same 12 weeks in Arm Z. It is important to note that the cumulative dose of each drug was the same for patients in both treatment arms (Table 2). Patients in both arms who achieved a CR or a PR after CT received the same high-dose, extended-field radiation therapy that was used in the POF81 trial (except that visceral areas in patients with CS IV disease were not irradiated). Here, we present the 10-year results of the H90-A/B prospective, randomized trial (inclusion period, January 1, 1990 through December 31, 1996), as evaluated in December, 2001.

Table 2. Chemotherapy Regimens for the Randomized MH90-A/B Trial
Chemotherapy agentArm Y (1 cycle every 4 weeks for 3 cycles): Dose (mg/m2 IV daily)Arm Z (1 cycle every 3 weeks for 4 cycles): Dose (mg/m2 IV daily)
DayTotal dose over 12 weeksDayTotal dose over 12 weeks
1234515
  • IV: intravenously.

  • a

    The maximum dose was limited to 2 mg.

Epirubicin40402403030240
Bleomycin1010607.57.560
Vinblastine6184.518
Vincristinea1.33.914
Cyclophosphamide65065039005005004000
Etoposide150150900110110880
Methotrexate303018022.522.5180
Methylprednisolone10010010010010015001851851480

MATERIALS AND METHODS

The diagnosis of Hodgkin disease and histologic classification were determined after patients underwent peripheral lymph node biopsy, thoracotomy, or mediastinoscopy (17 patients) or after patients underwent laparotomy or lapararoscopy (25 patients). All patients had a physical examination, chest X-ray, bipedal lymphangiography (or abdominal computed tomography scan), bone marrow biopsy, and routine biologic evaluation, including blood counts, sedimentation rate, and liver and kidney chemistries. A liver biopsy was performed in patients with liver enlargement and/or an increase > 2-fold in serum alkaline phosphatse level. Clinical and radiologic examination of the Waldeyer ring (with a biopsy when necessary) was performed in patients with upper cervical lymph nodes. Patients were classified with CS IA–IVB disease according to the Ann Arbor criteria.17 Each patient also was classified according to the number of involved lymph node areas (NINA), which were calculated according to the anatomic definition of lymph node regions adopted at the Rye symposium (i.e., right and left cervical lymph nodes, right and left axillary lymph nodes, right and left inguinal lymph nodes, right and left pelvic lymph nodes, mediastinal lymph nodes, lumboaortic lymph nodes, splenic lymph nodes, and Waldeyer ring).18 The mediastinal mass ratio (MMR), which was defined as the maximum mediastinal tumor width divided by the thoracic width at the sixth thoracic vertebra (T6), was identified in all patients with mediastinal disease. Bulky supradiaphragmatic disease was defined by an MMR ≥ 0.45.15, 19 Bulky infradiaphragmatic disease corresponded to involvement of the lumboaortic area and one or both pelvic areas.

After verification of the inclusion criteria (diagnosis of Hodgkin disease, age ≥ 18 years and ≤ 65 years, CS I–III with supradiaphragmatic and/or infradiaphragmatic bulky disease or CS IV, absence of serious simultaneous disease, absence of pregnancy, and informed consent), patients were included in the H90-A/B trial by the central secretary, and they were randomized to Arm Y or Arm Z.

In Arm Y, initial CT was administered over 12 weeks in 3 courses, whereas the same total dose of CT was administered in 4 courses over the same 12 weeks in Arm Z (Table 2). CT generally was administered on an inpatient basis in Arm Y and on an outpatient basis in Arm Z. The dosage of each drug was reduced to two-thirds of the normal dose for patients with leukocyte counts between 2.0 g/L and 2.5 g/L or platelet counts between 60 g/L and 80 g/L; CT infusion was postponed for 1 week when leukocyte and/or platelet counts were < 2.0 g/L and/or 20.0 g/L, respectively.

Two weeks after the last CT infusion, patient status was assessed: CR was defined as the complete disappearance of clinical and radiologic disease. PR was defined as a decrease > 50% in lymph node tumor burden with the disappearance of visceral disease in patients with CS IV tumors; and failure was defined as a decrease < 50% in lymph node tumor burden or the persistence, progression, or reappearance of lymph node disease and/or the persistence of visceral disease.

Patients who achieved a CR or a PR after CT received RT, which was started 4–5 weeks after the last CT infusion. Supradiaphragmatic irradiation included initially involved lymph nodes (40 Gy; 10 Gy per week) as well as adjacent, noninvolved lymph nodes sites (30 Gy). When upper cervical areas initially were not involved clinically, the upper limit of the irradiation field was lowered to the third cervical vertebra. All patients with initial lymph node disease limited to the upper hemitorso received adjuvant infradiaphragmatic irradiation (30 Gy) that encompassed the lumboaortic area (from T11–T12 to the third lumbar vertebra) and the spleen.

Patient status was reassessed 1 month after the completion of RT. Patients who were without clinical or radiologic disease were considered in CR. Patients with a residual mediastinal mass after irradiation (whether this mass was identical in size or had decreased compared with its size after CT) were considered in CR provided the mass had not regrown by 6 months after completion of RT. Treatment was considered a failure in patients with mediastinal masses that regrew under irradiation or within 6 months after the completion of treatment. Patients with disease that failed to respond to CT or RT were treated by other combination CT regimens (mainly with vinorelbine, cytarabine, and cisplatin) or with intensive CT regimens with or without autologous stem cell transplantation.20, 21

Remission status was checked every 3 months during the first year, then twice each year until the fifth year, and once each year thereafter. A recurrence was defined by the reappearance of disease; pathologic material obtained at the time of recurrence (biopsy or fine-needle aspiration) was examined in parallel with the initial material. Patients who developed recurrent disease were treated with intensive salvage CT or with various regular CT regimens and irradiation whenever possible.

Survival probabilities were calculated from the starting date of CT; all deaths were taken into account. Probabilities of FFP also were calculated from the beginning of CT; the events included were failure (after CT or RT) or recurrence and iatrogenic deaths observed during treatment; patients who died in CR were censored at the time of death. Probabilities of survival and FFP were calculated by the Kaplan–Meier method, and the differences between curves were assessed by log-rank tests.22 Chi-square tests or, when appropriate, Fisher exact tests (two-sided) were used for comparing qualitative data. Multivariate analysis based on a Cox proportional hazards regression model was performed to select disease characteristics that contributed significantly to prognosis. Binary disease characteristics were entered into the model: gender (male vs. female), age (< 45 years vs. ≥ 45 years), histology (nodular sclerosis type vs. other subtypes), disease extent (bulky CS I–III vs. CS IV and NINA < 5 vs. NINA ≥ 5), and contiguous extralymph node involvement (present vs. absent). The result of initial CT (CR vs. absence of CR) also was taken into account as a time dependent covariate.23 All statistical analyses were performed with SPSS 10.0 software (SPSS Inc., Chicago, IL).

RESULTS

Characteristics of the 162 patients who participated in the trial are listed in Table 3. The median age was 33 years (32 years in Arm Y and 35 years in Arm Z). Eighty-six patients had bulky CS I–III disease: 23 patients had bulky CS IA–IIA disease (14 patients in Arm Y and 9 patients in Arm Z), 17 patients had bulky CS IB–IIB disease (7 patients in Arm Y and 10 patients in Arm Z), 17 patients had bulky CS IIIA disease (12 patients in Arm Y and 5 patients in Arm Z), and 29 patients had bulky CS IIIB disease (11 patients in Arm Y and 18 patients in Arm Z). Seventy-six patients had CS IV disease (35 patients in Arm Y and 41 patients in Arm Z). Overall, the number of patients in each arm was similar with regard to essential characteristics.

Table 3. Initial Characteristics of the 162 Adult Patients with Bulky Clinical Stage I–III or Clinical Stage IV Hodgkin Disease who Were Included in the Randomized Multicenter MH90-A/B Trial
CharacteristicArm Y (n = 79 patients)Arm Z (n = 83 patients)Total (n = 162 patients)
  • CS: clinical stage.

  • a

    Mediastinal mass ratio ≥0.45.

  • b

    Pelvic and lumboaortic involvement.

Gender   
 Male5252104
 Female273158
Age (yrs)   
 < 456460124
 ≥ 45152338
Histology   
 Nodular sclerosis484997
 Other subtypes313465
CS I–III disease   
 With bulky supradiaphragmatic disease onlya201939
 With bulky infradiaphragmatic disease onlyb242145
 With bulky disease on both sides of the diaphragm-22
CS IV disease   
 With one extranodal site7069139
 With two or more extranodal sites91423
Contiguous extranodal involvement   
 No6469133
 Yes151429
No. of involved lymph node areas   
 < 5494897
 ≥ 5303565

Full doses of CT were administered to all patients, except for three patients (two patients in Arm Y and one patient in Arm Z) who received ≥ 75% of the planned dose. Twenty-three patients had febrile neutropenia that lasted ≥ 5 days (9 patients in Arm Y and 14 patients in Arm Z), among whom 3 patients in Arm Z died. Thirty-five patients received packed red cell transfusions (20 patients in Arm Y and 15 patients in Arm Z), and 6 patients received platelet transfusions (4 patients in Arm Y and 2 patients in Arm Z). The planned doses of RT (40/30 Gy) were administered to all patients who achieved a CR or a PR after CT, except 4 patients for whom doses were reduced to 30/25 Gy because of persistent cytopenia. The median duration of treatment was 27.9 weeks. All patients in the H90-A/B trial were followed until their date of death (38 patients) or until December, 2001 (no patient was lost to follow-up). The median follow-up for the 124 living patients was 90 months (range, 59–132 months).

The outcome of the 162 patients is summarized in Table 4. After the completion of CT, 68 patients (41.9%) achieved a CR, whereas 78 patients (48%) achieved a PR, and 13 patients failed to respond to CT (no differences between Arm Y and Arm Z). A total of 146 patients (90%) who achieved a CR or a PR after CT received RT; 140 of those patients achieved a CR, whereas 6 patients failed to respond to RT.

Table 4. Treatment Results and Events Observed in 162 Patients with Bulky Clinical Stage I–III and Clinical Stage IV Hodgkin Disease who Were Included in the Randomized Multicenter MH90-A/B Trial
VariableArm YArm ZTotal
  1. CT: chemotherapy; CR: complete remission; PR: partial remission.

Patients who received CT7983162
 No. of deaths during CT (iatrogenic)33
 CR after CT (%)35 (44.3)33 (39.8)68 (41.9)
 PR after CT403878
 Failure after CT4913
Patients who received lymph node irradiation7571146
 CR after irradiation (%)70 (88.6)70 (84.3)140 (86.4)
 Failure after irradiation516
Patients who received to salvage therapy after failure91019
 Alive with sustained CR5611
 Deceased in CR11
 Deceased with active disease437
Patients who were followed after completing irradiation7070140
 Alive with sustained first CR514495
 Deceased in first CR4610
 Recurrence after first CR152035
 Alive with second or third CR after recurrence10818
 Deceased with active disease after recurrence51217

Among the 19 patients (12% of the whole group) who failed to respond to treatment, 6 patients (3 in each arm) died with Hodgkin disease, and 13 patients (6 patients in Arm Y and 7 patients in Arm Z) achieved a CR after salvage treatment. Nine of those 13 patients remained alive in sustained remission (5 patients in Arm Y and 4 patients in Arm Z); 1 patient died from a secondary acute myeloid leukemia (sAML) without Hodgkin disease after 26 months in CR (Arm Z); and 3 patients developed recurrent disease, 2 of whom enjoyed a second, permanent CR (Arm Z), and 1 patient died with Hodgkin disease (Arm Y). Overall, 11 patients were alive in CR, and 8 patients died, 7 patients with Hodgkin disease and 1 patient in CR from an sAML (Table 4).

Among 140 patients who achieved a CR after treatment, 35 patients developed recurrent disease after 3–58 months in CR (median, 13 months; 12.5 months in Arm Y and 13.5 months in Arm Z). Twenty-five patients (71%) developed recurrent disease in the initially involved and irradiated lymph node areas (with or without other sites), 4 patients developed recurrent disease in lymph nodes that were irradiated prophylactically, 3 patients developed recurrent disease in initially nonirradiated (and noninvolved) lymph node areas, and the final 3 patients developed recurrent disease in visceral areas only. Nine patients failed to respond to salvage therapy (3 patients in Arm Y and 6 patients in Arm Z) and died, whereas the 26 other patients (12 patients in Arm Y and 14 patients in Arm Z) obtained a second CR. Of these, 17 patients (8 patients in Arm Y and 9 patients in Arm Z) were alive and free of Hodgkin disease in December, 2001 (1 patient had been treated for malignant melanoma).The nine other patients (three patients in Arm Y and six patients in Arm Z) developed second recurrences that were retreated successfully in one patient, whereas the other eight patients eventually died with Hodgkin disease (two patients in Arm Y and six patients in Arm Z). Overall, of 35 patients who developed recurrent disease, 18 patients were alive in CR, and 17 patients died with Hodgkin disease (Table 4).

Fatal and life-threatening events that occurred in patients who achieved a first CR were recorded carefully. One patient died from radiation-induced pneumonitis 2 months after completing mediastinal irradiation (Arm Z). Nine patients acquired a second tumor after a median of 52 months after the completion of therapy. Four of those patients developed a non-Hodgkin lymphoma (two patients with Arm Y and two patients in Arm Z), three of whom died (one patient in Arm Y and two patients in Arm Z). Another patient suffered from a fatal sAML (Arm Y). Moreover, four patients developed a solid tumor that was fatal in one patient (lung carcinoma, Arm Y) and was treated successfully in three patients (squamous cell carcinoma of the skin, Arm Y; breast carcinoma, Arm Y; urethral tumor, Arm Z). Moreover, four patients died in first CR from conditions that probably were not related to CT or RT: One patient died from a sigmoiditis-induced peritonitis (Arm Z), two patients died from cerebral strokes (one patient in each arm), and one patient committed suicide (Arm Z). Conversely, two patients suffered from a myocardial infarction (Arm Y) after 2 years and 5 years in CR, but neither patient died. Both of these patients were heavy smokers, both had infradiaphragmatic disease, and neither had received mediastinal irradiation.

Altogether, a total of 38 patients died: 3 patients from hematologic toxicity during CT, 24 patients with Hodgkin disease (7 patients after failure and 17 patients after developing recurrent disease), and 11 patients in CR (10 patients in CR after regular therapy and 1 patient in CR after salvage treatment) (Table 4).

Milder complications also were recorded. Chronic hepatitis virus C infection was diagnosed after 2 years of CR in a patient who was transfused during CT (Arm Y). Moreover, 10 patients acquired herpes zoster (5 patients in each arm). Finally, hyperthyroidism or hypothyroidism occurred in seven patients (three patients in Arm Y and four patients in Arm Z).

Overall, the 10-year FFP rate for the whole group of 162 patients was 63.9% (Arm Y, 69.2%; Arm Z, 58.8% P = 0.16), and the 10-year overall survival rate was 76.7% (Arm Y, 83.3%; Arm Z, 70.2%; P = 0.12) (Table 5). It is noteworthy that all living patients were free of Hodgkin disease in December, 2001.

Table 5. Rates of Complete Remission, Freedom from Disease Progression, and Overall Survival for 162 Patients with Bulky Clinical Stage I–III and Clinical Stage IV Hodgkin Disease who Were Included in the Randomized Multicenter H90-A/B Trial
Treatment armNo. of patientsCR rates (%)FFP rates (%)OS rates (%)
After CTAfter RTFive yrsTen yrsFive yrsTen yrs
  • CR: complete remission; FFP: freedom from progression; OS: overall survival.

  • a

    Log-rank test (P = 0.12).

Arm Y7944.388.670.569.286.183.3
Arm Z8339.884.360.258.878.370.2a
Total16242.086.465.363.980.976.7

Posttreatment CR rates were higher in patients who had bulky CS I–III disease compared with patients who had CS IV disease (96.0% vs. 75.0%; P < 0.01), in patients who had NINA < 5 compared with patients who had NINA ≥ 5 (93.9% vs. 75.0%; P < 0.01), and in patients who achieved a CR after CT compared with patients who did not achieve a CR after CT (98.5% vs. 77.7%; P < 0.01). The 10-year FFP rates were higher in patients who had bulky CS I–III disease compared with patients who had CS IV disease (74.9% vs. 51.4%; P = 0.0003; log-rank test), in patients who had NINA < 5 compared with patients who had NINA ≥ 5 (72.3% vs. 51.4%; P = 0.004; log-rank test), in patients without contiguous extralymph node involvement compared with patients who had extralymph node involvement (68.8% vs. 45.5%; P = 0.02; log-rank test), and in patients who achieved a CR after CT compared with patients who did not achieve a CR after CT (73.9% vs. 56.7%; P = 0.01) (Fig. 1A–C). Conversely, the 10-year survival rate was affected marginally by the initial extent of disease (bulky CS I–III disease vs. CS IV disease; P = 0.051) but not by the NINA (P = 0.09) (Fig. 1D,E). Moreover, the post-CT status was not a predictor of survival (Fig. 1F).

Figure 1.

Ten-year freedom from progression (FFP) and overall survival (OS) rates for patients with Hodgkin disease according to the extent their disease (bulky clinical Stage [CS] I–III vs. CS IV) (A,D), the number of involved lymph node areas (NINA: < 5 areas vs. ≥ 5 areas) (B,E), and the postchemotherapy (Post-CT) status (complete remission [CR] vs. no CR) (C,F).

The 10-year multivariate analysis that was performed first on initial disease characteristics and then by taking into account post-CT status confirmed the results of the univariate analysis: Among the parameters entered in the Cox model, the extent of disease (bulky CS I–III disease vs. CS IV disease; NINA < 5 vs. NINA ≥ 5) and the status at completion of CT (CR vs. no CR) were the only parameters that had independent influence on FFP rates (odds ratio: CS, 2.51 [range, 1.45–4.33]; P = 0.001; NINA, 1.78 [range, 1.05–3.02]; P = 0.032; post-CT status, 2.05 [range, 1.15–3.64]; P = 0.014). Conversely, the survival rate was influenced independently by age (< 45 years vs. ≥ 45 years) and the extent of disease (bulky CS I–III disease vs. CS IV disease odds ratio: age, 3.75 [range, 1.95–7.20]; P = 0.0001; CS, 2.02 [range, 1.05–3.88]; P = 0.034).

DISCUSSION

For decades, there has been general agreement on the worse prognosis of patients with CS IV Hodgkin disease; more recently, it was found that patients with CS I–III disease with large amounts of lymph node tumor, either because they had a high numbers of involved lymph node areas16, 24, 25 or because they had one site (generally the mediastinum) affected by an important volume of disease,15, 26–29 also were at high risk of failure or disease recurrence. The H90-A/B trial studied patients who had CS IV Hodgkin disease or CS I–III Hodgkin disease with bulky supradiaphragmatic and/or infradiaphragmatic disease. This group of patients was identified on the basis of the results of our previous POF81 protocol, in which all patients with CS IA–IVB disease received the same short CT regimen (three cycles of ABVD) followed by extended, high-dose RT.14–16

The two arms of this Phase II randomized study differed only in the dose density of the initial 12-week CT regimen. The same cumulative dose was delivered in three courses to patients in Arm Y and in four less intense courses to patients in Arm Z. The efficacy of the two randomized CT regimens was similar in terms of post-CT CR, PR, or failure rates. Overall, the toxicity for both arms was not significantly different. Five-year and 10-year FFP and survival probabilities also were similar for patients in both arms.

Overall, the combination of this 12-week, 7-drug CT regimen plus full-dose extended RT resulted in 124 of 62 patients (≈ 75%) with high-risk Hodgkin disease who were alive and free of disease after a median follow-up of 90 months. Very similar rates were observed for the same group of high-risk patients who were included in the POF81 trial, in which the initial CT regimen consisted of three courses of ABVD, a combination that had the same efficacy but lower toxicity15, 16 (Fig. 2).

Figure 2.

Ten-year freedom from progression (FFP) and overall survival rates for patients with advanced/bulky Hodgkin disease (clinical Stage [CS] I–III bulky and CS IV). Results of the prospective, nonrandomized Paris-Ouest-France 81 (POF81) trial (solid line) are compared with results of the prospective, randomized H90-A/B trial (hatched line).

To date, there has been no definitive option for treating patients with high-risk Hodgkin disease. Over the last 2 decades, two main strategies—CT alone and combined-modality therapy—have been compared in several randomized trials. The CT-alone strategy has evolved from 6–12 monthly cycles of MOPP or similar CT regimens to the more effective and less toxic 6–12 cycles of ABVD or ABVD hybridized with MOPP or MOPP-like CT regimens. With the combined treatment strategy, the same types of CT regimens have been applied generally over the same duration, but lymph node irradiation (with doses ranging from 15–20 Gy to 35–45 Gy) has been added after the completion of CT or in a sandwich manner between two CT sequences. A recent meta-analysis dealing with 16 randomized trials clarified the benefits of each strategy. Overall, the bimodal therapy generally provided a significant advantage in terms of disease control; conversely, concerning overall survival, CT alone sometimes was better, although both types of therapy frequently provided similar results.7, 8

In contrast to these conventional approaches of treatment, our group began development 20 years ago of an original program for treating patients with advanced Hodgkin disease. It combined an initial CT regimen of 12 weeks (3 cycles of ABVD from 1981 to 1988, 3 or 4 cycles of a 7-drug combination CT regimen from 1990 to 1996) followed by full-dose lymph node irradiation. It is interesting to note that the survival rates for this original bimodal therapy were equivalent to the rates obtained by conventional methods (Table 6).

Table 6. Survival Rates in Nine Randomized Trials Comparing Chemotherapy Alone or Combined with Irradiation for Patients with Advanced-Stage Hodgkin Disease
ReferenceNo. of patientsClinical stageCT type (duration in months)RT type and dose (Gy)% Survival (yrs)
  • CT: chemotherapy; RT: radiotherapy; Gy: grays; MOPP: mechlorethamine, vincristine, procarbazine, and prednisone; BleoMOPP: bleomycin, mechlorethamine, vincristine, procarbazine, and prednisone; ABVD: doxorubicin, bleomycin, vinblastine, and dacarbazine; CVPP: lomustine, vinblastine, procarbazine, and prednisone; EF: extended field; IF: involved field; (S)TNI: (sub)total lymph node irradiation; COPP: cyclophosphamide, vincristine, procarbazine, and prednisone; BEACOPP: bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone; EBVVCEM: epirubicin, bleomycin, vinblastine, vincristine, cyclophosphamide, etoposide, and methotrexate.

  • a

    Randomization was restricted to patients who achieved a complete or good partial response to initial CT.

  • b

    The survival rate was calculating using only patients who responded to initial CT. Overall survival rates are not shown.

  • c

    Irradiation of initial bulky disease (30 Gy) or residual disease (40 Gy).

  • d

    The survival rates shown are for 5 years and 10 years (5/10), respectively.

Canellos et al.3361III–IV (with recurrent disease after RT)MOPP (6–8)66 (5)
   ABVD (6–8)73
   MOPP/ABVD (12)73
Viviani et al.4427IB–IV (with recurrent disease after RT)MOPP/ABV (12)72 (10)
   MOPP/aBVD (12)74
Glick et al.33691III–IV (with recurrent disease after RT)MOPP (6–8)/ABVD (3)71 (8)
   MOPP/ABV (8–12)79
Pavlovsky et al.34151III, IVCVPP (6)58 (7)
   CVPP (3)/RT and CVPP (3)IF (30)71
Fabian et al.13278aIII, IVMOPP/BAP (6)86 (8)b
   MOPP/BAP (6), RTIF (20)79
Glick et al.35172aIIIB, IVBleoMOPP/ABVD (9)78 (8)b
   BleoMOPP (6), RTIF (15–20)65
Glick et al.36319aIII, IVBCVPP (8–12)68 (6)b
   MOPP/ABVD (9–12)75
   BCVPP (8–12), RTIF (15–20)63
Ferme et al.37418aIIIB, IVMOPP/ABV (8)88 (5)b
   MOPP/ABV (6), RT(S)TNI (30–40)94
Diehl et al.311195IIB–IIIA (with risk factors) and IIIB–IVCOPP/ABVD (6), RTIF (30–40)c83 (5)
   BEACOPP baseline (6), RTIF (30–40)c88
   BEACOPP escalated (6), RTIF (30–40)c91
H90-A/B(current study)162I–III bulky and IVEBVVCEM 3 courses (3), RTEF (40–30)86/83 (5/10)d
   EBVVCEM 4 courses (3), RTEF (40–30)78/70 (5/10)d

From a general point of view, to cure patients with high-risk Hodgkin disease necessitates several successful interventions: 1)sterilization of visceral lesions, whether they are occult or identifiable (CS IV disease), a goal that can be reached only with CT; 2) sterilization of lymph node lesions, which can be realized with CT alone or combined with RT; and 3)whatever the type of treatment, it should induce the lowest level of fatal events. In this setting, it is interesting to analyze the extent to which the results of our H90-A/B trial fulfilled the above-mentioned items. 1) After our 12-week CT regimen, we observed failures in 13 of 162 patients (8%), 4 of which were purely visceral; moreover, visceral recurrences developed in 3 of 140 patients who achieved a CR. These observations demonstrate that visceral lesions are sensitive even to a brief and nonintensive CT regimen. Thus, a more intensive regimen probably would lower this already low rate of failure. 2) Concerning lymph node disease, with our initial CT regimen, 41.9% of patients achieved a CR, and 48% of patients achieved a PR; the CR rate increased to 86.4% after irradiation. It remains true that most recurrences developed in lymph node areas that had been irradiated at full dose or almost full dose (40/30 Gy). It is noteworthy that a high proportion of these incidents of recurrence occurred in patients who did not achieved a CR after CT (Fig. 1C). Thus, to increase the sterilization rate of lymph node disease, an interesting possibility may be to increase the intensity of initial CT before delivering RT; such a strategy would increase the post-CT CR rate and probably would decrease the recurrence rate, as suggested by the low rate of recurrence observed in a Phase II study with the 12-week Stanford V CT regimen and 36-Gy consolidation RT to initial sites of bulky disease.30 3) The proportion of fatal events that resulted from our regimen should be evaluated in keeping with the number of patients who still die of Hodgkin disease. In the H90-A/B trial, three patients died from CT-related sepsis, one patient died from radiation-induced pneumonitis, and five patients died in first CR from a second tumor. However, the risk of dying from Hodgkin disease was far greater (23 patients). Thus, when designing new protocols for patients with high-risk Hodgkin disease, the priority remains to increase the sterilization rate of Hodgkin disease (without increasing the level of iatrogenic deaths).

In conclusion, results of this Phase II randomized trial strongly confirm the results obtained previously in our prospective, nonrandomized POF81 trial that used three courses of ABVD plus RT (Fig. 2). It is interesting to note that the results from both of these trials using a brief, nonintensive course of CT plus full-dose RT compete favorably with the results reported using conventional treatment approaches (Table 6). Detailed analysis of failures and recurrences suggests that increasing the intensity of the initial CT should increase the rate of post-CT CR and decrease the rate of recurrences, as suggested by the Stanford Phase II trial30 and by early results from a German trial.31 In this setting, in January, 1997, our group initiated a new randomized trial (Groupe Ouest et Est des Leucémies et Autres Maladies de Sang [GOELAMS] H97-LM) comparing two programs of brief but intensive CT followed by consolidative irradiation. The first arm of this trial included three monthly cycles of an intensive CT (which already had been tested successfully as salvage treatment21) plus consolidative RT (36 Gy for initial masses measuring > 5 cm, 20 Gy for other initially involved lymph nodes); the alternative arm received four monthly cycles of the ABVD combination plus one cycle of CT with carmustine, etoposide, aracytine, and melphalan (a myeloablative CT) followed by autologous peripheral stem cell transplantation and consolidative RT (36 Gy on initial masses measuring > 5 cm).32

Ancillary