Early-stage Hodgkin's disease: Long-term results with radiotherapy alone or combined radiotherapy and chemotherapy

Annals of Oncology 5 (Suppl. 2): S101-S106, 1994. C 1994 Kluwer Academic Publishers. Printed in the Netherlands.

Original article Early-stage Hodgkin's disease: Long-term results with radiotherapy alone or combined radiotherapy and chemotherapy E. Brusamolino,1'2 M. Lazzarino,1'2 E. Orlandi,1'2 A. Canevari,1'2 E. Morra, 12 G. Castelli,1'2 E. P. Alessandrino,1'2 G. Pagnucco,1-2 C. Astori,12 A. Livraghi,1'2 F. Corbella,3 P. Franchini3 & C. Bernasconi1'2 1

Summary

Background: Controversy still exists over the optimal management of early-stage Hodgkin's disease (HD); presentation features may have a different prognostic impact according to initial therapy, and long-term toxicity must be fully evaluated. Patients and methods: This study included 164 patients with stage IA-IIA HD treated with radiotherapy (RT) alone or combined radio- and chemotherapy (CT) according to presenting features and their attendant prognostic significance. The RT group included 88 patients with favorable prognostic features; the combined modality group included 76 patients with one or more unfavorable features. In the RT group, 85% of patients received extended-mantle or STNI; in the combined modality group, RT consisted of mantle(49%), extended mantle- (37%), and involved-field irradiation (14%); CT consisted of 6 cycles of MOPP before 1984; 3 cycles of ABVD were substituted for MOPP thereafter. Results: Complete remission was obtained in 94% and

Introduction

The diagnostic evaluation and the optimal management of patients with stage I and II HD remain controversial because different approaches have proved to be curative in a high proportion of patients. These include RT alone in surgically [1-3] or clinically [4,5] staged patients, combined modality therapy [6-10], and CT alone [11,12]. The disease-free survival in stage I and II HD treated with RT alone is up to 75% of patients: those with supradiaphragmatic disease in pathologic stages IA and HA have shown a 14-year freedom from progression and overall survival of 93% and 82%, respectively [3]. Patients with large mediastinal masses demonstrated a higher risk of relapse than patients with lesser or no mediastinal disease following treatment with RT alone [6, 13, 14]; in this setting, a combined modality approach has improved the RFS as compared to RT alone. However, overall survival (OS) advantages have not been obtained because of the efficacy of salvage

99% of patients of the RT and combined modality groups, respectively. The 10-year actuarial relapse-free survival (RFS) in the RT group was 62% and was influenced by stage (p — 0.04) and histology (p — 0.01); in the combined modality group, RFS was 88% and was influenced by the presence of bulky disease. Overall survival and tumor mortality between the therapy groups were comparable. RT-related toxicity consisted of mediastinal fibrosis (8 cases), myelitis (3), hypothyroidism (2); other long-term events included 2 cases of acute leukemia in the combined MOPP and RT group. Altogether, 8 of 20 patients who died were in their first complete remission. Conclusions: In stage IA-IIA HD, the combined modality therapy reduced the risk of relapse compared to radiation alone; long-term toxicity of RT was not negligible and relapses could occur late. Key words: combined modality, Hodgkin's disease, initial stages, radiotherapy

chemotherapy in patients relapsing after treatment with RT alone [6]. Several other characteristics have been suggested to indicate a poor prognosis in early-stage Hodgkin's disease treated with irradiation alone; these characteristics include histologic features of mixed cellularity and lymphocyte depletion, presence of B symptoms, of numerous sites of involvement, of contiguous extranodal (E) disease, of pulmonary hilum involvement, advanced age, and elevated ESR values [6, 1517]. Prospective studies with therapy tailored to specific prognostic factors [6, 10,18] and a large experience on clinically staged patients with stage I and II disease [4] confirmed the usefulness of radiotherapy alone in the most favorable subgroups (absence of systemic symptoms, pathologic stage I). All other disease stages require adding chemotherapy to reduce the risk of relapse. Toxicity of different diagnostic and therapeutic modalities must be weighed against their advantages. In this respect, we should take into account the morbidity

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

Cattedra di Ematologia, Universitd di Pavia, Pavia; 2Divisione di Ematologia, iServizio di Radiotcrapia, Policlinico San Matteo IRCCS, Pavia, Italy

102 and the infectious risk of staging laparotomy [19], the late effects of radiotherapy [20], the toxicity of the combined modality approach, including gonadal dysfunction [21,22], and the risk of acute leukemia [23-26]. Beginning in October 1977, we initiated a study to evaluate the role of radiotherapy alone or combined radiochemotherapy in patients with stages IA-IIA HD, and to assess the risk of unwanted long-term events. Diagnostic procedures and therapy program were based upon presenting features and their attendant prognostic significance. The study results are reported in this paper.

Patient allocation The study design included previously untreated patients, with a histologic diagnosis of Hodgkin's disease according to Lukes and Butler [27], in stage IA-IIA. Based upon their presenting features, the patients were divided into two groups (favorable and unfavorable) with different diagnostic and therapeutic programs. The favorable group included patients with stage I, and/or lymphocyte prevalence and nodular sclerosis histologies, without bulky or extranodal disease and pulmonary hilum involvement. The less favorable group included patients with stage n, mixed cellularity and lymphocyte depletion histologies, bulky or extranodal disease, and/or pulmonary hilum involvement The favorable group patients were pathologically staged with laparotomy and splenectomy and given RT alone; the majority of patients in the less favorable group had clinical stage only (61%) and entered a combined modality program.

Staging procedures Clinical staging procedures were performed according to the guidelines of the Ann Arbor Conference [28] and included a complete history, physical examination, routine laboratory tests including hemogram, erythrocyte sedimentation rate (ESR), liver and renal function tests, posteroanterior and lateral chest X-rays, bipedal lymphogram (before 1984), chest and abdominal computed tomography (after 1984). Pathological staging consisted of bilateral bone marrow biopsies in all cases and of laparotomy with splenectomy, nodal, and liver biopsies in 118 cases (72%).

Characteristics

Total evaluable Males Females Median age (yrs) Range (yrs) Histology LP NS MC LD Stage I Stage n Laparotomy Subdiaphragmatic Enlarged mediastinum Bulky mediastinum ESR < 25 mm Hr"1 ESR > 25 mm Hr"'

No. of cases

Radiotherapy alone

Combined therap)

No.

No.

164 93 71

88 50 38 33 16-64

30 93 39 2 61 103 98 14

22 46 20 0 48 40 88 7

77 14 104 60

30 0 65 23

(%) (57) (43)

(25) (52) (23) (55) (45) (100) (8) (34) (74) (26)

P value

(%)

76 43 (57) 33 (43) 27 14-72 8 47 19 2 13 63 30 7

(11) (62) (25) (2) (17) (83) (39) (9)

0.01" ns ns

47 14 39 37

(62) (18) (59) (41)

0.000 0.000 0.003

ns 0.000

ns

ESR - erythrocyte sedimentation rate.

group, therapy consisted of subtotal nodal irradiation sparing the pelvis (20% of cases) or mantle-field irradiation extended to paraaortic nodes (55%) at standard doses (40-44 Gy in 5-8 weeks by opposed fields). Patients with disease limited to the mediastinum received mantle-field only (15%), and those with high neck disease or axilla were given involved-field RT (6%). Patients with disease limited to the inguinal area (4%) received an inverted Y-field irradiation ending at the diaphragm. In the combined modality therapy group, RT consisted of mantle- (49% of cases), extended mantle(37%), and involved-field irradiation (14%) at doses of 30-35 Gy. The chemotherapy consisted before 1984 of 6 cycles of the MOPP, and after 1984, of 3 cycles of the ABVD combination. The MOPP regimen was delivered according to the NCI schedule [29] and either preceded the irradiation or was given through a sandwich technique (3 MOPP/RT/3 MOPP). The ABVD regimen was administered as originally described [30]. A dose-reduction schedule was adopted for patients showing myelosuppression at the time of CT administration.

Statistics Patient characteristics The series consisted of 164 consecutive patients admitted to our division from 1977 to 1988; the main characteristics are illustrated in Table 1 according to the treatment group. The RT-alone group consisted of 88 patients, and all underwent staging laparotomy with splenectomy: 55% were in stage I, 77% had favorable histologies,' and no case had bulky mediastinum, E lesions, or pulmonary hilum involvement The combined modality group included 76 patients, and laparotomy was done in 39% of them. Altogether, 18% were in stage I, 27% had *unfavorable histologies,' 18% had bulky disease, and 21% had pulmonary hilum involvement Significant differences in the presentation features between treatment groups were present as far as lymphocyte prevalence histology, stage, bulky disease, and ESR were concerned. Therapy Radiotherapy was delivered through m Co teletherapy or megavoltage linear accelerator with standard techniques. In the RT-alone

Differences in the patient characteristics between treatment groups were compared with the Pearson test for contingency tables. Timeto-event distributions for relapse-free and overall survival were estimated by the product-limit method of Kaplan-Meier [31], and the statistical significance of the differences observed was assessed by the log-rank test [32]. The Cox proportional hazard regression procedure was used for the analysis of covariates influencing RFS and OS [33).

Results Complete remission was achieved in 94% and 99% of patients in the RT and combined modality therapy groups, respectively. Table 2 summarizes the RFS and the OS according to initial therapy and pretreatment characteristics. Fig. 1 illustrates the actuarial curves of the two groups.

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

Patients and methods

Table I. Characteristics of patients according to therapy.

103 Table 2. Relapse-free and overall survival according to therapy. Characteristics

Combined modality

Radiotherapy alone

RFS

RFS

OS 10-yr

Total

91

88

Nodular sclerosis Others

87 94

87 90

ns

Nonbulky Bulky disease

91 88

91 74

Stage I Stage II

92 91

92 88

Mediastinal involvement

86

81

P value

P value

10-yr

5-yr

10-yr

10-yr

96

84

P value

81

69

62

95 100

75 83

58 80

49

°01

93 98

81 87

ns

0.008

98 90

88 70

ns

91 96

80 83

75 58

S °04

99 94

86 82

ns

98

78

65

48

92

87

73

0.01

c

'roportion

*—•

—>

8

84%

"

6

b

4

4

Total 88 Dead 10

*s62%

Total 83 Relapsed 30

2 0

O 0

20

40

80

60

100

120

0

140

10

20

30

40

Months

50

60

70

80

90

Months

B

D

Proportion 1

1

a

81%

Proportion

J-

.8

88%

.6

6

.4

.4 2

5-yr

1

a

2

P value

93

A

Proportion

5-yr

Total 76 . Dead 10

Total 75 Relapsed 8

.2 0

0

30

40

SO

60

70

80

90

1OO

Months

110

0

20

40

60

80

100 120

140

160 180

Months

Fig. 1. Survival curves of stages IA-D"A Hodgkin's disease. A and C — overall and relapse-free survival of RT patients; B and D — overall and relapse-free survival of combined modality therapy patients.

Multiple regression analysis indicated mediastinal involvement as the single most important factor influIn the RT-alone series, 30 of 83 patients who achieved encing the probability of complete response to radiaCR have relapsed (36%); the 5- and 10-year actuarial tion alone (p = 0.04; R2 - 0.47), while the Cox's RFS were 69% and 62%, respectively. In univariate proportional hazard model indicated that clinical stage analysis, the probability of continuous remission in this was the only significant factor in terms of duration of group was adversely influenced by stage n and nodular remission (p - 0.007). sclerosis histology. A 10-year follow-up indicated a In the combined modality series, 8 of 75 patients 51% freedom from relapse for stage II patients vs. 68% who achieved CR have relapsed (11%); the 5- and 10of stage I (p - 0.04), and a 49% freedom from relapse year actuarial RFS were 91% and 88%, respectively. In for patients with nodular sclerosis histology vs. 73% for univariate analysis, the probability of continuous rethose with other histologies (p = 0.01). Relapses oc- mission was adversely influenced by the presence of curred in nodal sites in 19 cases (63% of relapses), in bulky disease; patients with bulky mediastinum showed extranodal sites in 5 cases (17%), and in both nodal a 10-year freedom from relapse of 74% vs. 91% of and extranodal in 6 cases (20%). Altogether, relapses those without bulky masses (p = 0.008); no differences in irradiated areas occurred in 10 patients (33%). were found between patients given the MOPP or Relapse-free survival

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

5-yr

OS

104 ABVD regimen in terms of CR, acute toxicity, and RFS. All relapses occurred in nodal sites (4 of 8 in previously irradiated areas). Multivariate analysis with the Cox's model indicated the ERS («S25 mm Hr"1) as the only significant variable for duration of CR in the combined modality therapy group (p — 0.05). Overall survival, long-term events, and causes of death

Late relapses and therapy of relapses The majority of relapses occurred within 2 years after CR; the median duration of CR in patients who relapsed was 20 months. However, late relapses were observed. 13 patients (34%) relapsed in the third year after completion of therapy, and 6 beyond the fourth year, the occurrence of late relapses was significantly related to the type of primary therapy (10 of 13 had been treated with RT only). Salvage therapy consisted of ABVD or MOPP alternating to ABVD in 26 cases (68%); 24 of them (92%) entered a second remission. Altogether, 32 patients (88%) have been salvaged by the second-line therapy, and 28 are alive with no evidence of disease.

Discussion Patients with stage IA-HA HD who had been given a combination of radio- and chemotherapy had significantly less risk of relapse compared to those given radiotherapy alone. The combined modality therapy counterbalanced the adverse prognostic impact of stage II and nodular sclerosis histology and allowed a 10-year RFS of 88%, compared to 62% of the RTalone group. No difference in the 10-year OS was observed between the two groups of treatment. This finding confirms consohdated observations on the efficacy

Long-term events

Total

No. of cases Radiation alone

Mediastinal fibrosis Solid tumors Arterial thrombosis Myelitis Hypothyroidism Acute leukemia Non-Hodgkin's lymphoma Bone marrow aplasia Pyloric stenosis Femozal necrosis Osseous tuberculosis

9 9 3 3 2 2 2 1 1 1 1

Causes of death

Total

%

6 2 2 5 3 2

30 10 10 25 15 10

Progression of HD Infections Acute leukemia Solid tumors Myocardial infarction Nonrelated Total1

20

Combined modality

5 4 2 2 2 0 1 1 1 1 0

4 5 1 1 0 2 1 0 0 0 1

3 2 0 3 2

3 0 2 2 3

10

10

* 8 of 20 died in their first complete remission.

of salvage therapy in patients relapsing after RT in early-stage Hodgkin's disease [6-8,10]. Nevertheless, can we rely on salvage therapy to cure a fraction of patients with IA-HA HD (from 20% to 50% of total, according to presenting features), or should we attempt to improve the results of first-line therapy with a combined approach? Alternative important approaches are being tested and consist of using chemotherapy alone [11,12] or a combination of limited-field radiation with chemotherapy. To limit the long-term complications of combined radio- and chemotherapy, the CT regimen should include nonmutagenic drugs like vinblastine, bleomycin, and methotrexate (VBM) [34] or a combination like ABVD that proved to be devoid of leukemogenic risk [24]. In this study, we combined RT with six cycles of MOPP before 1984 and 3 cycles of ABVD thereafter. The dropping of MOPP was prompted by the accumulating evidence of a leukemogenic risk inherent in the combined modality therapy using this regimen [23,37]. The choice of ABVD was supported by the excellent Milan results with this combination in advanced HD [35], and by our own subsequent experience [36]. The length of the ABVD program was limited in view of the fact that almost all patients who achieved CR with MOPP did so with the first three cycles. In our experience, the relative efficacy and acute toxicity of MOPP vs. ABVD were superimposable. However, both cases of metacronous acute leukemia belonged to the RT and MOPP group. On the other hand, long-term sequelae of RT were not negligible,

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

No difference was found in the overall survival between the two groups of therapy; the 10-year OS of the RT and combined modality groups were 84% and 81%, respectively. Multivariate analysis indicated the stage and ESR as borderline factors predicting for the overall survival. The 10-year freedom from tumor mortality for the whole series of patients was 94%; in fact, only 6 patients died of HD progression. Table 3 illustrates the long-term events and the causes of death according to therapy. Radiation-related complications occurred in 16 cases and consisted of mediastinal fibrosis (9 cases), myelitis (2 cases), hypothyroidism (2 cases), and piloric stenosis and femural head necrosis (1 case each). Solid tumors occurred in 9 cases (4 in irradiated areas), malignant lymphoma in 2 cases, and acute leukemia in 2 cases treated with combination RT and MOPP chemotherapy. Secondary neoplasia represented the cause of death in 35% of cases; non-HD-related causes of death amounted to 25% of the total.

Table 3. Long-term events and causes of death.

105

21.

References 1. Hoppe RT. Radiation therapy in the management of Hodgkin's disease. Sem Oncol 1990; 17: 704-15. 2. Cornbleet MA, Vitolo U, Ultmann JE et al. Pathologic stages IA and IIA Hodgkin's disease: Results of treatment with radiotherapy alone (1968-1980). J Clin Oncol 1985; 3: 758-68. 3. Mauch P, Tarbell N, Weinstein H et al. Stage IA and IIA supradiaphragmatic Hodgkin's disease: Prognostic factors in surgically staged patients treated with mantle and para-aortic irradiation. J Clin Oncol 1988; 6:1576-83. 4. Gospodarowicz MK, Sutcliffe SB> BergsagelDE etal. Radiation therapy in clinical stage I and II Hodgkin's disease. Eur J Cancer 1992; 28A: 1841-6. 5. Carde P, Burgers JMV, Henry-Amar M et al. Clinical stages I and II Hodgkin's disease: A specifically tailored therapy according to prognostic factors. J Clin Oncol 1988; 6: 239-52. 6. Hoppe RT, Coleman CN, Cox RS et al. The management of stage I-II Hodgkin's disease with irradiation alone or combined modality therapy: The Stanford experience. Blood 1982; 59:455-65. 7. Hagemeister FB, Fuller LM, Velasquez WS et al. Stage I and II

22. 23. 24. 25. 26. 27. 28.

Hodgkin's disease: Involved-field radiotherapy vs. extendedfield radiotherapy vs. involved-field radiotherapy followed by six cycles of MOPP. Cancer Treat Rep 1982; 66: 789-98. Nissen NI, Nordentoft AM. Radiotherapy versus combined modality treatment of stage I and II Hodgkin's disease. Cancer Treat Rep 1982; 66: 799-803. Straus DJ, Yahalom J, Gaynor J et al. Four cycles of chemotherapy and regional radiation therapy for clinical early-stage and intermediate-stage Hodgkin's disease. Cancer 1992; 69: 1052-60. Bonfante V, Santoro A, Viviani S et al. Early-stage Hodgkin's disease: Ten-year results of a non-randomised study with radiotherapy alone or combined with MOPP. Eur J Cancer 1993; 29A: 24-9. Longo DL, Glatstein E, Duffey PL et al. Radiation therapy versus combination chemotherapy in the treatment of earlystage Hodgkin's disease: Seven-year results of a prospective randomized trial. J Clin Oncol 1991; 6: 906-17. Bid GP, Cimino G, Cartoni C et al. Extended-field radiotherapy is superior to MOPP chemotherapy for the treatment of pathologic stage I-IIA Hodgkin's disease: Eight-year update of an Italian prospective randomized study. J Clin Oncol 1992; 10: 378-82. Leopold KA, Canellos GP, Rosenthal D et al. Stage IA-UB Hodgkin's disease: Staging and treatment of patients with large mediastinal adenopathy. J Clin Oncol 1989; 7:1059-65. Prosnitz LR, Curtis AM, Knowlton AH et al. Supradiaphragmatic Hodgkin's disease: Significance of large mediastinal masses. Int J Radiat Oncol Biol Phys 1980; 6: 809-15. Cmkowich MJ, Hoppe RT, Rosenberg SA. Stage IIB Hodgkin's disease: The Stanford experience. J Clin Oncol 1986; 4: 472-9. Tubiana M, Henry-Amar M, van der Werf-Messing B et al. A multivariate analysis of prognostic factors in early stage Hodgkin's disease. Int J Radiat Oncol Biol Phys 1985; 11: 23-30. Zagard G, Rubin P. Laparotomy-staged IA versus HA Hodgkin's disease. A comparative study with evaluation of prognostic factors for stage IIA. Cancer 1985; 56: 864-73. Tubiana M, Henry-Amar P, Carde P et al. Toward comprehensive management tailored to prognostic factors of patients with clinical stages I and II in Hodgkin's disease: The EORTC Lymphoma Group controlled clinical trial, 1964-1987. Blood 1989; 73: 47-56. Coker DD, Morris DM, Coleman JJ et al. Infection among 210 patients with surgically staged Hodgkin's disease. Am J Med 1983; 75: 97-109. Carmel RJ, Kaplan HS. Mantle irradiation in Hodgkin's disease. Cancer 1976; 54:1928-42. Chapman RM, Sutcliffe SB, Malpas JS. Male gonadal dysfunction in Hodgkin's disease: A prospective study. J Am Med Assoc 1981; 245:1323-8. Homing SJ, Hoppe RT, Kaplan HS et al. Female reproductive potential after treatment for Hodgkin's disease. N Engl J Med 1981; 304: 1377-82. Brusamolino E, Lazzarino M, Salvaneschi L et al. Risk of leukemia in patients treated for Hodgkin's disease. Eur J Cancer Clin Oncol 1982; 18: 237-42. Valagussa P, Santoro A, Fossati-Bellani F et al. Second acute leukemia and other malignancies following treatment for Hodgkin's disease. J Clin Oncol 1986; 4: 830-7. Tucker MA, ColemanCN, Cox RS et al. Risk of second cancer after treatment for Hodgkin's disease. N Engl J Med 1988; 318: 76-81. Young RC, Bookman MA, Longo DL. Late complications of Hodgkin's disease management Natl Cancer Inst Monogr 1990; 10: 55H5O. Lukes RJ, Butler JJ. The pathology and nomenclature of Hodgkin's disease. Cancer Res 1966; 26:1063-83. Carbone PP, Kaplan HS, Musshof K et al. Report of the committee on Hodgkin's disease classification. Cancer Res 1971; 31:1860-1.

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

with particular reference to mediastinal fibrosis and myelitis (3 cases). The choice of RT alone in HD entails the issue of laparotomy and its potential complications; 8. the morbidity rate from this procedure is 10%-15%, with a lifelong increased risk of serious infections [19]. Moreover, a significant increased risk of secondary 9. leukemia after splenectomy has been observed in our [38] and others' experience [39]. Late relapses can occur in HD [40]. In this study, 10. a correlation was observed between the use of RT alone and risk of relapse beyond the third year of continuous remission. This fact indicates that a nonsystemic approach, even in early stages, may allow the 11. persistence of clinically undetectable residual disease and that prolonged follow-up is particularly needed in this setting. 12. No significant differences between treatment programs were seen for intercurrent diseases or Hodgkin's disease mortality, except for acute leukemia (2 cases), which developed only in the RT plus MOPP subgroup. 13. Altogether, 8 of 20 patients who died had no evidence of disease, and in agreement with previous reports [10,41], the occurrence of a secondary neoplasia was 14. the most frequent cause of death. In conclusion, patients with early-stage HD have an 15. overall excellent prognosis with state-of-the-art therapies. However, as indicated in a recent editorial [42], 16. further refinements in management programs are needed to minimize long-term sequelae of therapy and to optimize the cure rate. The use of a short-term (three 17. to four cycles), nonleukemogenic chemotherapy (ABVD or VBM) with limited-field radiotherapy (minimum therapy) can spare laparotomy with splenec- 18. tomy, reduce early and late relapses compared to RT alone, and prove equally curative and less toxic than RT with adjuvant MOPP. The confirmation of the efficacy of CT alone in early-stage HD will prompt a 19. prospective comparison of CT alone with such a limited combined modality approach. 20.

106

38. 39. 40. 41. 42.

phocytic leukemia, preleukemia, and acute myeloproliferative syndrome up to 10 years after treatment of Hodgkin's disease. N Engl J Med 1982; 307: 965-71. Tura S, Fiacchini M, Zinzani PL et al. Splenectomy and the increasing risk of secondary acute leukemia in Hodgkin's disease. J Clin Oncol 1993; 11: 925-30. Kaldor JM, Day NE, Clarke EA et al. Leukemia following Hodgkin's disease. N Engl J Med 1990; 322: 7-13. Herman TS, Hoppe RT, Donaldson SS et al. Late relapse among patients treated for Hodgkin's disease. Ann Intern Med 1985; 102: 292-7. Hancock SL, Hoppe RT, Horning SJ et al. Intercurrent death after Hodgkin's disease therapy in radiotherapy and adjuvant MOPP trials. Ann Intern Med 1988; 109:183-9. Hoppe RT. Early-stage Hodgkin's disease: A choice of treatments or a treatment of choice? J Clin Oncol 1991; 9: 897901.

Correspondence to: Dr. Ercole Brusamolino Divisione & Cattedra di Ematologia Policlinico San Matteo IRCCS 27100 Pavia Italy

Downloaded from http://annonc.oxfordjournals.org/ at University of Tennessee ? Knoxville Libraries on September 20, 2015

29. DeVita VT, Serpick AA, Carbone PP et al. Combination chemotherapy in the treatment of advanced Hodgkin's disease. Ann Intern Med 1970; 73: 881-95. 30. Bonadonna G, Zucali R, Monfardini S et al. Combination chemotherapy of Hodgkin's disease with Adriamycin, bleomycin, vinblastine, and imidazole carboxamide versus MOPP. Cancer 1975; 36: 252-9. 31. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53:457-81. 32. Peto R, Pike MC, Armitage P et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II, Analysis and examples. Br J Cancer 1977; 35: 1-39. 33. Cox DR. Regression models and life tables. J R Stat Soc B 1972; 34:187-220. 34. Homing SJ, Hoppe RT, Hancock SL et al. Vinblastine, bleomycin, and methotrexate: An effective adjuvant in favorable Hodgkin's disease. J Clin Oncol 1988; 6:1822-31. 35. Bonadonna G. Chemotherapy strategies to improve the control of Hodgkin's disease. Cancer Res 1982; 42: 4309-20. 36. Brusamolino E, Lazzarino M, Morra E et al. Combination chemotherapy with alternating MOPP-ABVD in advanced Hodgkin's disease. Haematologica 1989; 74:173-9. 37. Pedersen-Bjergaard J, Larsen SO. Incidence of acute nonlym-