- Email: [email protected]
Results of the 1988–1989 patterns of care study process survey for Hodgkin’s disease
Int. J. Radiation Oncology Biol. Phys., Vol. 43, No. 2, pp. 335–339, 1999 Copyright © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/99/$–see front matter
PII S0360-3016(98)00388-5
CLINICAL INVESTIGATION
Hodgkin’s Disease
RESULTS OF THE 1988 –1989 PATTERNS OF CARE STUDY PROCESS SURVEY FOR HODGKIN’S DISEASE MELANIE C. SMITT, M.D.,* NICOLE STOUFFER, M.S.,† JEAN B. OWEN, PH.D.,† RICHARD T. HOPPE, M.D.,* AND GERALD E. HANKS, M.D.‡ *Department of Radiation Oncology, Stanford University Medical Center, Stanford, CA; †American College of Radiology, Philadelphia, PA; and ‡Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA Purpose: To document national standards of care for patients receiving radiotherapy as part of curative treatment for Hodgkin’s disease. Materials and Methods: A national survey was conducted of 61 institutions treating 275 patients with Stages I–III Hodgkin’s disease and representing six facility type strata. Pretreatment evaluation, radiotherapy treatment parameters, and use of combined modality therapy were assessed. Results: Ann Arbor stage for the 275 patients was as follows: IA, 69 (25%); IB, 7 (3%); IIA, 123 (45%); IIB, 36 (13%); IIIA 23 (8%), IIIB, 14 (5%); unknown, 3 (1%). Pretreatment evaluation included complete blood count for 93%, sedimentation rate in 29%, chest CT in 88%, abdominal CT scan in 87%, and bone marrow biopsy in 81%. Lymphangiograms were obtained in 50% of cases; laparotomy was performed in 46%. The yield of positive findings in the spleen at laparotomy was 6.5% overall. Facility differences with respect to staging were seen only for the use of gallium scans, which were more commonly used in academic centers (44% vs. 15–23% elsewhere, p < 0.001). Radiotherapy was delivered with a linear accelerator in 94% of cases. Treatment simulation was performed for 94% and individualized blocks constructed for 95% overall; however, freestanding facilities had a lower rate of performance of these procedures (78% vs. 98 –99% for simulation and 88% vs. 96 –99% for customized blocking, p < 0.001). The mean supradiaphragmatic dose was 36.74 Gy and the mean subdiaphragmatic dose was 33.81 Gy. Planned combined modality therapy was given in 36% of patients. The use of combined modality therapy by stage was as follows: IA, 11%; IB, 43%; IIA, 30%; IIB, 68%; IIIA, 57%; IIIB, 100%. Chemotherapy was completed prior to radiation in 80% of cases and generally consisted of ABVD (32%), an alternating regimen (25%), or MOPP (22%). Among Stage I/II patients, use of chemotherapy was associated with reduced radiation doses (mean supradiaphragmatic dose 34.53 Gy vs. 38.43 Gy and mean subdiaphragmatic dose 31.27 Gy vs. 34.51 Gy), and reduced volumes of treatment (87% vs. 28% treated to one side of the diaphragm only). Laparotomy was not associated with decreased supra- or subdiaphragmatic radiation doses or decreased volumes of treatment. Conclusions: With the exception of gallium scans, pretreatment evaluation is relatively uniform across facility strata. Increased understanding of prognostic factors in Hodgkin’s disease and greater use of planned combined modality therapy for higher risk patients appears to have contributed to a decreased use of and low yield of positive findings for laparotomy. Laparotomy was not associated with reduced radiation volumes or doses. Freestanding radiation facilities had a lower rate than other facility types for the performance of treatment simulation and customized patient blocking. © 1999 Elsevier Science Inc. Hodgkin’s, Radiotherapy, Patterns of care.
INTRODUCTION
treatment parameters, and prevalence of combined modality treatment for Hodgkin’s disease at the national level (1). Outcome studies of the 1973 and 1983 patients demonstrated the importance of these processes on relapse rates for various stages of disease (2, 3). In addition to documenting national practice, the Patterns of Care studies have been important in developing and disseminating national guidelines for treatment. Consensus recommendations have been published in newsletter format following each survey. The PCS surveys continue to provide us with an understanding
The Patterns of Care Study (PCS) was designed to obtain data on national standards of care for several disease sites where radiotherapy plays an important role in management. The first nationwide PCS process survey for Hodgkin’s disease collected data for patients treated in 1973; subsequent surveys collected data on patients treated in 1983 and in 1989. The PCS has, through the process surveys, documented the extent of pretreatment evaluation, radiotherapy Reprint requests to: Melanie C. Smitt, M.D., Department of Radiation Oncology, Stanford Hospital, 300 Pasteur Dr, Stanford, CA 94305.
Accepted for publication 28 August 1998.
335
336
I. J. Radiation Oncology
●
Biology
●
Physics
of the evolution of national practice for Hodgkin’s disease and of areas where further improvement may be appropriate. MATERIALS AND METHODS Methodology for the Patterns of Care Surveys and calculation of national average data has been previously published (4). Briefly, a consensus panel designs the process survey questionnaire for a particular disease site. Facility and patient selection occurs via a stratified, two-stage cluster sampling with simple random sampling at each stage for each stratum. Six facility strata were utilized for the 1989 study: A1, RTOG or Comprehensive Cancer Center; A2, teaching hospital with .2 residents, not in A1; HE, hospital-based facility with electron capability; HN, hospitalbased facility without electron capability; FE, freestanding facility with electron capability; FN, freestanding facility without electron capability. Two hundred and seventy five eligible patients treated at 61 facilities were included in the 1989 Hodgkin’s disease process survey. Data were collected from patient charts by a three-person surveyor team, consisting of a data manager, radiation oncologist, and radiation physicist. Patients were excluded from the Hodgkin’s disease survey if they were not considered suitable for definitive radiation treatment or had disseminated extralymphatic involvement (Stage IV disease). RESULTS Relevant characteristics of the 275 treated patients are given in Table 1. The final Ann Arbor stage as assessed by the PCS surveyor, was as follows: IA, 69 (25%); IB, 7 (3%); IIA, 123 (45%); IIB, 36 (13%); IIIA 23 (8%), IIIB, 14 (5%); unknown, 3 (1%). The histologic type was characterized as lymphocyte predominance for 10.5%, nodular sclerosis for 68.4%, mixed cellularity for 19.3%, lymphocyte depletion for 0.7%, and other/unknown for 1.1%. The mediastinum was involved in 60.6%; an E lesion was identified in 4%. Four percent of patients were treated on institutional or cooperative group protocols. National average data for various staging procedures are given in Table 2. Complete blood counts were obtained in 93% of patients; a sedimentation rate was determined for 29%. Plain films of the chest (CXR) were performed for 92%, chest CT for 88%, and CXR or CT for 100%. Abdominal CT was performed in 87% of patients, lymphangiogram in 50%, and 94% had an abdominal CT or lymphangiogram (LAG). The primary reason for not obtaining a LAG was that it was felt to be unnecessary (14%). However, 6% of surveyed practices felt that the technology was not available. Gallium scans were obtained in 26%. Sampling of bone marrow occurred for 81% overall and for 80% of stage I/IIA patients. Forty-six percent of patients underwent a laparotomy. Pathologic findings from splenectomy in these patients were negative 93.5% of the time, and only 0.8% of removed spleens had 5 or more nodules of
Volume 43, Number 2, 1999
Table 1. Patient characteristics Age
%
,20 20–29 30–39 40–49 .50
11.6 36 22.2 14.5 15.7
Race
%
White Black Hispanic Asian Sex Male Female Tissue diagnosis LPHD NSHD MCHD LDHD Other/Unk
88 6.6 3.6 0.4 % 51.1 48.9 % 10.5 68.4 19.3 0.7 1.1
Stage
%
IA IB IIA IIB IIIA IIIB Unknown
25 3 45 13 8 5 1
Hodgkin’s disease. These findings partly reflect selection of high risk patients for combined modality treatment prior to laparotomy or, possibly, management of patients with PS IIIA disease with chemotherapy alone. Patients with B symptoms rarely underwent laparotomy; none of the Stage IB patients and 23% of Stage IIB patients were pathologically staged with laparotomy. Use and findings of laparotomy by final stage are shown in Table 3. A complete analysis of the relationship between clinical findings and pathologic staging is not, unfortunately, available within the context of the survey. Gallium scans were obtained by 44% of academic facilities but by only 15–23% of other facility types (p , 0.001). Relevant facility characteristics and RT treatment processes for patients treated in 1989 are given in Table 4. Although overall standards of care are excellent, significant differences are seen by facility type. Freestanding facilities were less likely to perform simulation than academic or hospital-based facilities (78% vs. 98 and 99%, p , 0.001); they also had a lower rate of taking port films at each field change (48% vs. 79 and 83%, p 5 0.003) or using individually shaped blocks (88% vs. 96 and 99%, p , 0.001) and
Hodgkin’s PCS 1989 Process Survey
Table 2. Use of staging procedures1 (percentage, national averages)
● M. C. SMITT et al.
337
Table 4. RT treatment processes (percentage, national averages) All Academic Hospital Freestanding
Procedure
All
Academic
Hospital
Freestanding
ESR CXR LAG CXR or chest CT chest CT abdomen CT LAG or abdCT marrow bx marrow bx I/IIA gallium scan laparotomy*
29 92 50 100 88 87 94 81 80 26 46
21 97 57 100 83 86 97 87 85 44 42
37 88 45 100 91 84 92 79 78 23 50
15 98 54 100 86 93 97 81 80 15 50
Linac# Simulation Ports Ports each change Indiv blocks* Field sep calc Special blocks
other special blocks (57% vs. 94 and 94%, p , 0.001). Freestanding facilities without electron capability generally had the lowest rates of performance in these areas. Only 43% of these facilities performed simulations; 73% used individually shaped blocks. The mean supradiaphragmatic dose was 36.74 Gy and the mean subdiaphragmatic dose was 33.81 Gy. For Stage I/II patients treated with RT alone, the mean supradiaphragmatic dose was 38.43 Gy (median 39.16, range 24.90 – 48.00) and the mean subdiaphragmatic dose was 34.51 Gy (median 35.83, range 25.48 – 42.00). Seventy-two percent of these patients received treatment to both sides of the diaphragm. Laparotomy did not appear to reduce the dose or volume of RT administered among these Stage I/II patients. The mean subdiaphragmatic dose was 34.64 Gy and 34.24 Gy with or without laparotomy respectively, and the mean supradiaphragmatic dose was higher among patients undergoing laparotomy 39.42 Gy vs. 36.91 Gy. Twenty-one percent of Stage I/II patients treated with RT alone who underwent laparotomy had fields restricted to one side of the diaphragm vs. 40% of those who did not undergo laparotomy. For Stage I/IIA patients who had undergone laparotTable 3. Percentage of patients undergoing laparotomy %
IA IB IIA IIB IIIA IIIB
38 0 60 24 62 23
Findings of laparotomy
%
Negative Spleen .5 nodules
75 95 63 87
92 99 100
100 98 99
91 78 100
79 99 39 94
83 96 86 94
48 88 35 57
,0.001 0.003 ,0.001 ,0.001
* Individualized blocks constructed. # Use of linear accelerator for treatment, absolute values. 1 Freestanding facilities vs academic and hospital-based facilities.
* indicates use of absolute value where national average data not available. 1 Tables 2–5 were modified and reprinted from Smitt MC, Buzydlowski J, Hoppe RT. Over 20 years of progress in radiation oncology: Hodgkin’s disease. Seminars in Radiation Oncology 7(2):127–34, 1997.
By Final Stage
94 94 99.3
p1
93.5 0.8
omy, 63% were treated with a standard mantle and subdiaphragmatic field, 21% were treated to one side of the diaphragm alone, and the remainder had some modification of the mantle field along with subdiaphragmatic treatment. Thirty-six percent of patients were treated with planned chemotherapy. The use of chemotherapy by stage is shown in Table 5. Combined modality therapy was used in the majority of patients with B symptoms, bulky mediastinal disease, or stage III disease. ABVD was used in 32% of CMT cases, MOPP was used in 22%, and an alternating regimen was employed in 25%. Patients treated at freestanding radiation facilities appeared more likely to receive MOPP chemotherapy (47%) than those treated at other facilities (15–20%) although this difference was not statistically significant. Eighty percent of patients undergoing CMT completed chemotherapy prior to radiation. The mean supradiaphragmatic radiation dose among Stage I/II patients receiving combined modality therapy was 34.53 Gy (median 35.24, range 20.40 – 45.00) and the mean subdiaphragmatic dose was 31.27 Gy (median 30.60, range 20.00 – 36.60). Of these patients, 87% had treatment limited to one side of the diaphragm.
DISCUSSION With the exception of gallium scans, patterns of assessment and staging for Hodgkin’s disease were fairly uniform among the facility strata in the 1989 survey. Further improvement in workup in all strata would be achieved by routine incorporation of the ESR, a well-documented prognostic factor, into staging evaluation and in reduction of use of bone marrow biopsy for low risk stage I/IIA patients. In comparison with the 1973 and 1983 results, this survey documents an increased use of computerized tomography in staging with a concomitant decline in the use of lymphangiograms and laparotomy (14). In 1983, 56% of patients underwent chest CT vs. 88% in 1989. During that same period, use of LAG dropped from 70% to 50% and
338
I. J. Radiation Oncology
●
Biology
●
Physics
Table 5. Use of combined modality therapy Stage All IA IB IIA IIB IIIA IIIB BulkyMed*
% 36 11 43 30 68 57 100 90
Type
%
MOPP ABVD Alternating Other/Unknown
22 32 25 21
Sequence CT Before RT CT After RT Sandwich Concurrent
80 9 2 2
* Bulky mediastinal disease defined as mediastinal mass . 8 cm where mediastinum was bulkiest site of disease.
performance of laparotomy from 77% to 46%. A relatively low yield of positive laparotomy findings (6.5%) was observed because planned combined modality therapy without laparotomy was more often employed in high risk patients in the 1989 survey and it is likely that some laparotomy staged patients with PSIIIA disease were not referred for radiation treatment but were managed instead with chemotherapy alone. The use of laparotomy was not associated with a reduction in supra- or subdiaphragmatic radiation doses or radiation treatment volumes. In part this reflects selection factors in the use of laparotomy and in the design of radiotherapy fields. However, along with the lack of proven survival benefit for use of laparotomy (5), these findings document and support continued reduction in the use of staging laparotomy for Hodgkin’s disease. Radiotherapy treatment processes have clearly improved at the national level since the 1973 and 1983 surveys. For example, only 45% of patients in 1973 and 84% of patients in 1983 were treated on a linear accelerator as compared to 94% in 1989. The importance of radiotherapy processes on outcome was demonstrated in a multivariate analysis of Stage I/II patients treated with radiotherapy alone (2). Equipment type and the use of individually shaped blocks were found to be independent prognostic factors for relapse along with stage and extent of treatment field. Outcome data from 1973 showed patients treated at freestanding facilities to have significantly poorer relapse-free survival (6). Of concern in the 1989 survey is the persistence of differences
Volume 43, Number 2, 1999
in radiotherapy processes among facility types. Freestanding facilities as a group, and particularly those without electron capability, have a substandard level of performance in several areas. Whether further substantial facility upgrading can be encouraged in the current cost-conscious medical climate is unclear. It is conceivable that insurers could direct patients, by contracting arrangements, to facilities with high standards of care, such as those that utilize dedicated simulation and individualized blocking. However, if cost is the only consideration in contracting, facility upgrading may be discouraged. Future PCS surveys may examine the use of staging procedures, treatment facility types, and outcomes based on payor information. The mean supradiaphragmatic and subdiaphragmatic radiation doses have declined in the PCS surveys since 1973. The mean supradiaphragmatic dose was 39.90 cGy in 1973, 39.12 cGy in 1983, and 36.74 Gy in 1989. The mean subdiaphragmatic dose was 38.62 cGy in 1973, 35.18 cGy in 1983, and 33.81 Gy in 1989. Decreases from 1973 to 1983 most likely reflect the publication of the PCS doseresponse data; changes from 1983 also reflect the impact of increasing use of combined modality treatments. The risk of major bowel complications has been related to use of radiation doses greater than 3500 cGy, especially if a laparotomy is performed (7). Decreases in dose should diminish radiation-related treatment toxicity, provided that the radiation dose-response curve for complications is not substantially affected by chemotherapy. It has been previously reported that almost 20% of cases treated at either academic or freestanding facilities did not undergo daily treatment of both anterior and posterior fields (8). More universal treatment of both anterior and posterior fields daily could diminish late effects of treatment. Reductions in radiation volumes are also important, particularly in regards to secondary carcinogenesis (9) and cardiovascular mortality (10). Further surveys may address the use of more limited fields of radiation in the context of CMT or based on risk profiles. The 1973–1989 surveys document an increasing use of combined modality treatment. Thirteen percent of patients in 1973 received planned CMT, twenty-two percent in 1983, and thirty-six percent in 1989. This trend was primarily evident for poorer prognosis patients, specifically those with B symptoms or Stage III disease. Traditional MOPP chemotherapy use dropped from 59% in 1983 to 22% in 1989 while use of ABVD or alternating regimens went from 3% to 57%. However, treatment at a freestanding radiation facility appeared to be associated with increased likelihood of receiving MOPP chemotherapy. Recently, new chemotherapy programs have been described which are designed to limit acute and long-term toxicities of treatment (11, 12). In addition, abbreviated versions of traditional chemotherapy may be effective in low risk patients (13, 14). Future surveys will be important in evaluating the application of these toxicity-reducing strategies.
Hodgkin’s PCS 1989 Process Survey
● M. C. SMITT et al.
339
REFERENCES 1. Smitt MC, Buzydlowski J, Hoppe RT. Over 20 years of progress in radiation oncology: Hodgkin’s disease. Semin Radiat Oncol 1997;7:127–134. 2. Hoppe RT, Hanlon AL, Hanks GE, et al. Progress in the treatment of Hodgkin’s disease in the United States, 1973 versus 1983. The Patterns of Care Study. Cancer 1994;74: 3198 –3203. 3. Kinzie JJ, Hanks GE, Maclean CJ, et al. Patterns of Care Study: Hodgkin’s disease relapse rates and adequacy of portals. Cancer 1983;52:2223–2226. 4. Owen JB, Sedransk J, Papak TF. National averages for process and outcome in radiation oncology: methodology of the Patterns of Care Study. Sem Radiat Oncol 1997;7:101–107. 5. Carde P, Hagenbeek A, Hayat M, et al. Clinical staging versus laparotomy and combined modality with MOPP versus ABVD in early-stage Hodgkin’s disease: the H6 twin randomized studies from the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Group. J Clin Oncol 1993;11:2258 –2272. 6. Hanks GE. Hodgkin’s disease. Successful dissemination of high technology treatment throughout the United States. Cancer Surveys 1985;4:477– 485. 7. Coia LR, Hanks GE. Complications from large field intermediate dose infradiaphragmatic radiation: an analysis of the
8. 9. 10. 11. 12.
13. 14.
Patterns of Care outcome studies for Hodgkin’s disease and seminoma. Int J Radiat Oncol Biol Phys 1988;15:29 –35. Hughes DB, Smith AR, Hoppe RT, et al. Treatment planning for Hodgkin’s disease: a Patterns of Care Study. Int J Radiat Oncol Biol Phys 1995;33:519 –524. Hancock SL, Tucker MA, Hoppe RT. Breast cancer after treatment of Hodgkin’s disease JNCI 1993;85:25–31. Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin’s disease. JAMA 1993;270:1949 –1955. Bartlett NL, Rosenberg SA, Mason J, et al. Brief chemotherapy and adjuvant radiotherapy for bulky or advanced stage Hodgkin’s disease. J Clin Oncol 1995;13:1080 –1088. Horning SJ, Hoppe RT, Mason J, et al. Stanford-Kaiser Permanente G1 study for clinical stage I to IIA Hodgkin’s disease: subtotal lymphoid irradiation versus vinblastine, methotrexate, and bleomycin chemotherapy and regional irradiation. J Clin Oncol 1997;15:1736 –1744. Bonfante V, Santoro A, Viviani S, et al. ABVD plus radiotherapy (subtotal nodal vs. involved field) in early-stage Hodgkin’s disease. (Abstr.) Proc ASCO 1994;13:373. Klasa RJ, Connors J, Hoskins P, et al. Early stage Hodgkin’s disease: impact of brief chemotherapy together with radiotherapy without staging laparotomy. (Abstr.) Proc ASCO 1994; 13:372.
PII S0360-3016(98)00388-5
CLINICAL INVESTIGATION
Hodgkin’s Disease
RESULTS OF THE 1988 –1989 PATTERNS OF CARE STUDY PROCESS SURVEY FOR HODGKIN’S DISEASE MELANIE C. SMITT, M.D.,* NICOLE STOUFFER, M.S.,† JEAN B. OWEN, PH.D.,† RICHARD T. HOPPE, M.D.,* AND GERALD E. HANKS, M.D.‡ *Department of Radiation Oncology, Stanford University Medical Center, Stanford, CA; †American College of Radiology, Philadelphia, PA; and ‡Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA Purpose: To document national standards of care for patients receiving radiotherapy as part of curative treatment for Hodgkin’s disease. Materials and Methods: A national survey was conducted of 61 institutions treating 275 patients with Stages I–III Hodgkin’s disease and representing six facility type strata. Pretreatment evaluation, radiotherapy treatment parameters, and use of combined modality therapy were assessed. Results: Ann Arbor stage for the 275 patients was as follows: IA, 69 (25%); IB, 7 (3%); IIA, 123 (45%); IIB, 36 (13%); IIIA 23 (8%), IIIB, 14 (5%); unknown, 3 (1%). Pretreatment evaluation included complete blood count for 93%, sedimentation rate in 29%, chest CT in 88%, abdominal CT scan in 87%, and bone marrow biopsy in 81%. Lymphangiograms were obtained in 50% of cases; laparotomy was performed in 46%. The yield of positive findings in the spleen at laparotomy was 6.5% overall. Facility differences with respect to staging were seen only for the use of gallium scans, which were more commonly used in academic centers (44% vs. 15–23% elsewhere, p < 0.001). Radiotherapy was delivered with a linear accelerator in 94% of cases. Treatment simulation was performed for 94% and individualized blocks constructed for 95% overall; however, freestanding facilities had a lower rate of performance of these procedures (78% vs. 98 –99% for simulation and 88% vs. 96 –99% for customized blocking, p < 0.001). The mean supradiaphragmatic dose was 36.74 Gy and the mean subdiaphragmatic dose was 33.81 Gy. Planned combined modality therapy was given in 36% of patients. The use of combined modality therapy by stage was as follows: IA, 11%; IB, 43%; IIA, 30%; IIB, 68%; IIIA, 57%; IIIB, 100%. Chemotherapy was completed prior to radiation in 80% of cases and generally consisted of ABVD (32%), an alternating regimen (25%), or MOPP (22%). Among Stage I/II patients, use of chemotherapy was associated with reduced radiation doses (mean supradiaphragmatic dose 34.53 Gy vs. 38.43 Gy and mean subdiaphragmatic dose 31.27 Gy vs. 34.51 Gy), and reduced volumes of treatment (87% vs. 28% treated to one side of the diaphragm only). Laparotomy was not associated with decreased supra- or subdiaphragmatic radiation doses or decreased volumes of treatment. Conclusions: With the exception of gallium scans, pretreatment evaluation is relatively uniform across facility strata. Increased understanding of prognostic factors in Hodgkin’s disease and greater use of planned combined modality therapy for higher risk patients appears to have contributed to a decreased use of and low yield of positive findings for laparotomy. Laparotomy was not associated with reduced radiation volumes or doses. Freestanding radiation facilities had a lower rate than other facility types for the performance of treatment simulation and customized patient blocking. © 1999 Elsevier Science Inc. Hodgkin’s, Radiotherapy, Patterns of care.
INTRODUCTION
treatment parameters, and prevalence of combined modality treatment for Hodgkin’s disease at the national level (1). Outcome studies of the 1973 and 1983 patients demonstrated the importance of these processes on relapse rates for various stages of disease (2, 3). In addition to documenting national practice, the Patterns of Care studies have been important in developing and disseminating national guidelines for treatment. Consensus recommendations have been published in newsletter format following each survey. The PCS surveys continue to provide us with an understanding
The Patterns of Care Study (PCS) was designed to obtain data on national standards of care for several disease sites where radiotherapy plays an important role in management. The first nationwide PCS process survey for Hodgkin’s disease collected data for patients treated in 1973; subsequent surveys collected data on patients treated in 1983 and in 1989. The PCS has, through the process surveys, documented the extent of pretreatment evaluation, radiotherapy Reprint requests to: Melanie C. Smitt, M.D., Department of Radiation Oncology, Stanford Hospital, 300 Pasteur Dr, Stanford, CA 94305.
Accepted for publication 28 August 1998.
335
336
I. J. Radiation Oncology
●
Biology
●
Physics
of the evolution of national practice for Hodgkin’s disease and of areas where further improvement may be appropriate. MATERIALS AND METHODS Methodology for the Patterns of Care Surveys and calculation of national average data has been previously published (4). Briefly, a consensus panel designs the process survey questionnaire for a particular disease site. Facility and patient selection occurs via a stratified, two-stage cluster sampling with simple random sampling at each stage for each stratum. Six facility strata were utilized for the 1989 study: A1, RTOG or Comprehensive Cancer Center; A2, teaching hospital with .2 residents, not in A1; HE, hospital-based facility with electron capability; HN, hospitalbased facility without electron capability; FE, freestanding facility with electron capability; FN, freestanding facility without electron capability. Two hundred and seventy five eligible patients treated at 61 facilities were included in the 1989 Hodgkin’s disease process survey. Data were collected from patient charts by a three-person surveyor team, consisting of a data manager, radiation oncologist, and radiation physicist. Patients were excluded from the Hodgkin’s disease survey if they were not considered suitable for definitive radiation treatment or had disseminated extralymphatic involvement (Stage IV disease). RESULTS Relevant characteristics of the 275 treated patients are given in Table 1. The final Ann Arbor stage as assessed by the PCS surveyor, was as follows: IA, 69 (25%); IB, 7 (3%); IIA, 123 (45%); IIB, 36 (13%); IIIA 23 (8%), IIIB, 14 (5%); unknown, 3 (1%). The histologic type was characterized as lymphocyte predominance for 10.5%, nodular sclerosis for 68.4%, mixed cellularity for 19.3%, lymphocyte depletion for 0.7%, and other/unknown for 1.1%. The mediastinum was involved in 60.6%; an E lesion was identified in 4%. Four percent of patients were treated on institutional or cooperative group protocols. National average data for various staging procedures are given in Table 2. Complete blood counts were obtained in 93% of patients; a sedimentation rate was determined for 29%. Plain films of the chest (CXR) were performed for 92%, chest CT for 88%, and CXR or CT for 100%. Abdominal CT was performed in 87% of patients, lymphangiogram in 50%, and 94% had an abdominal CT or lymphangiogram (LAG). The primary reason for not obtaining a LAG was that it was felt to be unnecessary (14%). However, 6% of surveyed practices felt that the technology was not available. Gallium scans were obtained in 26%. Sampling of bone marrow occurred for 81% overall and for 80% of stage I/IIA patients. Forty-six percent of patients underwent a laparotomy. Pathologic findings from splenectomy in these patients were negative 93.5% of the time, and only 0.8% of removed spleens had 5 or more nodules of
Volume 43, Number 2, 1999
Table 1. Patient characteristics Age
%
,20 20–29 30–39 40–49 .50
11.6 36 22.2 14.5 15.7
Race
%
White Black Hispanic Asian Sex Male Female Tissue diagnosis LPHD NSHD MCHD LDHD Other/Unk
88 6.6 3.6 0.4 % 51.1 48.9 % 10.5 68.4 19.3 0.7 1.1
Stage
%
IA IB IIA IIB IIIA IIIB Unknown
25 3 45 13 8 5 1
Hodgkin’s disease. These findings partly reflect selection of high risk patients for combined modality treatment prior to laparotomy or, possibly, management of patients with PS IIIA disease with chemotherapy alone. Patients with B symptoms rarely underwent laparotomy; none of the Stage IB patients and 23% of Stage IIB patients were pathologically staged with laparotomy. Use and findings of laparotomy by final stage are shown in Table 3. A complete analysis of the relationship between clinical findings and pathologic staging is not, unfortunately, available within the context of the survey. Gallium scans were obtained by 44% of academic facilities but by only 15–23% of other facility types (p , 0.001). Relevant facility characteristics and RT treatment processes for patients treated in 1989 are given in Table 4. Although overall standards of care are excellent, significant differences are seen by facility type. Freestanding facilities were less likely to perform simulation than academic or hospital-based facilities (78% vs. 98 and 99%, p , 0.001); they also had a lower rate of taking port films at each field change (48% vs. 79 and 83%, p 5 0.003) or using individually shaped blocks (88% vs. 96 and 99%, p , 0.001) and
Hodgkin’s PCS 1989 Process Survey
Table 2. Use of staging procedures1 (percentage, national averages)
● M. C. SMITT et al.
337
Table 4. RT treatment processes (percentage, national averages) All Academic Hospital Freestanding
Procedure
All
Academic
Hospital
Freestanding
ESR CXR LAG CXR or chest CT chest CT abdomen CT LAG or abdCT marrow bx marrow bx I/IIA gallium scan laparotomy*
29 92 50 100 88 87 94 81 80 26 46
21 97 57 100 83 86 97 87 85 44 42
37 88 45 100 91 84 92 79 78 23 50
15 98 54 100 86 93 97 81 80 15 50
Linac# Simulation Ports Ports each change Indiv blocks* Field sep calc Special blocks
other special blocks (57% vs. 94 and 94%, p , 0.001). Freestanding facilities without electron capability generally had the lowest rates of performance in these areas. Only 43% of these facilities performed simulations; 73% used individually shaped blocks. The mean supradiaphragmatic dose was 36.74 Gy and the mean subdiaphragmatic dose was 33.81 Gy. For Stage I/II patients treated with RT alone, the mean supradiaphragmatic dose was 38.43 Gy (median 39.16, range 24.90 – 48.00) and the mean subdiaphragmatic dose was 34.51 Gy (median 35.83, range 25.48 – 42.00). Seventy-two percent of these patients received treatment to both sides of the diaphragm. Laparotomy did not appear to reduce the dose or volume of RT administered among these Stage I/II patients. The mean subdiaphragmatic dose was 34.64 Gy and 34.24 Gy with or without laparotomy respectively, and the mean supradiaphragmatic dose was higher among patients undergoing laparotomy 39.42 Gy vs. 36.91 Gy. Twenty-one percent of Stage I/II patients treated with RT alone who underwent laparotomy had fields restricted to one side of the diaphragm vs. 40% of those who did not undergo laparotomy. For Stage I/IIA patients who had undergone laparotTable 3. Percentage of patients undergoing laparotomy %
IA IB IIA IIB IIIA IIIB
38 0 60 24 62 23
Findings of laparotomy
%
Negative Spleen .5 nodules
75 95 63 87
92 99 100
100 98 99
91 78 100
79 99 39 94
83 96 86 94
48 88 35 57
,0.001 0.003 ,0.001 ,0.001
* Individualized blocks constructed. # Use of linear accelerator for treatment, absolute values. 1 Freestanding facilities vs academic and hospital-based facilities.
* indicates use of absolute value where national average data not available. 1 Tables 2–5 were modified and reprinted from Smitt MC, Buzydlowski J, Hoppe RT. Over 20 years of progress in radiation oncology: Hodgkin’s disease. Seminars in Radiation Oncology 7(2):127–34, 1997.
By Final Stage
94 94 99.3
p1
93.5 0.8
omy, 63% were treated with a standard mantle and subdiaphragmatic field, 21% were treated to one side of the diaphragm alone, and the remainder had some modification of the mantle field along with subdiaphragmatic treatment. Thirty-six percent of patients were treated with planned chemotherapy. The use of chemotherapy by stage is shown in Table 5. Combined modality therapy was used in the majority of patients with B symptoms, bulky mediastinal disease, or stage III disease. ABVD was used in 32% of CMT cases, MOPP was used in 22%, and an alternating regimen was employed in 25%. Patients treated at freestanding radiation facilities appeared more likely to receive MOPP chemotherapy (47%) than those treated at other facilities (15–20%) although this difference was not statistically significant. Eighty percent of patients undergoing CMT completed chemotherapy prior to radiation. The mean supradiaphragmatic radiation dose among Stage I/II patients receiving combined modality therapy was 34.53 Gy (median 35.24, range 20.40 – 45.00) and the mean subdiaphragmatic dose was 31.27 Gy (median 30.60, range 20.00 – 36.60). Of these patients, 87% had treatment limited to one side of the diaphragm.
DISCUSSION With the exception of gallium scans, patterns of assessment and staging for Hodgkin’s disease were fairly uniform among the facility strata in the 1989 survey. Further improvement in workup in all strata would be achieved by routine incorporation of the ESR, a well-documented prognostic factor, into staging evaluation and in reduction of use of bone marrow biopsy for low risk stage I/IIA patients. In comparison with the 1973 and 1983 results, this survey documents an increased use of computerized tomography in staging with a concomitant decline in the use of lymphangiograms and laparotomy (14). In 1983, 56% of patients underwent chest CT vs. 88% in 1989. During that same period, use of LAG dropped from 70% to 50% and
338
I. J. Radiation Oncology
●
Biology
●
Physics
Table 5. Use of combined modality therapy Stage All IA IB IIA IIB IIIA IIIB BulkyMed*
% 36 11 43 30 68 57 100 90
Type
%
MOPP ABVD Alternating Other/Unknown
22 32 25 21
Sequence CT Before RT CT After RT Sandwich Concurrent
80 9 2 2
* Bulky mediastinal disease defined as mediastinal mass . 8 cm where mediastinum was bulkiest site of disease.
performance of laparotomy from 77% to 46%. A relatively low yield of positive laparotomy findings (6.5%) was observed because planned combined modality therapy without laparotomy was more often employed in high risk patients in the 1989 survey and it is likely that some laparotomy staged patients with PSIIIA disease were not referred for radiation treatment but were managed instead with chemotherapy alone. The use of laparotomy was not associated with a reduction in supra- or subdiaphragmatic radiation doses or radiation treatment volumes. In part this reflects selection factors in the use of laparotomy and in the design of radiotherapy fields. However, along with the lack of proven survival benefit for use of laparotomy (5), these findings document and support continued reduction in the use of staging laparotomy for Hodgkin’s disease. Radiotherapy treatment processes have clearly improved at the national level since the 1973 and 1983 surveys. For example, only 45% of patients in 1973 and 84% of patients in 1983 were treated on a linear accelerator as compared to 94% in 1989. The importance of radiotherapy processes on outcome was demonstrated in a multivariate analysis of Stage I/II patients treated with radiotherapy alone (2). Equipment type and the use of individually shaped blocks were found to be independent prognostic factors for relapse along with stage and extent of treatment field. Outcome data from 1973 showed patients treated at freestanding facilities to have significantly poorer relapse-free survival (6). Of concern in the 1989 survey is the persistence of differences
Volume 43, Number 2, 1999
in radiotherapy processes among facility types. Freestanding facilities as a group, and particularly those without electron capability, have a substandard level of performance in several areas. Whether further substantial facility upgrading can be encouraged in the current cost-conscious medical climate is unclear. It is conceivable that insurers could direct patients, by contracting arrangements, to facilities with high standards of care, such as those that utilize dedicated simulation and individualized blocking. However, if cost is the only consideration in contracting, facility upgrading may be discouraged. Future PCS surveys may examine the use of staging procedures, treatment facility types, and outcomes based on payor information. The mean supradiaphragmatic and subdiaphragmatic radiation doses have declined in the PCS surveys since 1973. The mean supradiaphragmatic dose was 39.90 cGy in 1973, 39.12 cGy in 1983, and 36.74 Gy in 1989. The mean subdiaphragmatic dose was 38.62 cGy in 1973, 35.18 cGy in 1983, and 33.81 Gy in 1989. Decreases from 1973 to 1983 most likely reflect the publication of the PCS doseresponse data; changes from 1983 also reflect the impact of increasing use of combined modality treatments. The risk of major bowel complications has been related to use of radiation doses greater than 3500 cGy, especially if a laparotomy is performed (7). Decreases in dose should diminish radiation-related treatment toxicity, provided that the radiation dose-response curve for complications is not substantially affected by chemotherapy. It has been previously reported that almost 20% of cases treated at either academic or freestanding facilities did not undergo daily treatment of both anterior and posterior fields (8). More universal treatment of both anterior and posterior fields daily could diminish late effects of treatment. Reductions in radiation volumes are also important, particularly in regards to secondary carcinogenesis (9) and cardiovascular mortality (10). Further surveys may address the use of more limited fields of radiation in the context of CMT or based on risk profiles. The 1973–1989 surveys document an increasing use of combined modality treatment. Thirteen percent of patients in 1973 received planned CMT, twenty-two percent in 1983, and thirty-six percent in 1989. This trend was primarily evident for poorer prognosis patients, specifically those with B symptoms or Stage III disease. Traditional MOPP chemotherapy use dropped from 59% in 1983 to 22% in 1989 while use of ABVD or alternating regimens went from 3% to 57%. However, treatment at a freestanding radiation facility appeared to be associated with increased likelihood of receiving MOPP chemotherapy. Recently, new chemotherapy programs have been described which are designed to limit acute and long-term toxicities of treatment (11, 12). In addition, abbreviated versions of traditional chemotherapy may be effective in low risk patients (13, 14). Future surveys will be important in evaluating the application of these toxicity-reducing strategies.
Hodgkin’s PCS 1989 Process Survey
● M. C. SMITT et al.
339
REFERENCES 1. Smitt MC, Buzydlowski J, Hoppe RT. Over 20 years of progress in radiation oncology: Hodgkin’s disease. Semin Radiat Oncol 1997;7:127–134. 2. Hoppe RT, Hanlon AL, Hanks GE, et al. Progress in the treatment of Hodgkin’s disease in the United States, 1973 versus 1983. The Patterns of Care Study. Cancer 1994;74: 3198 –3203. 3. Kinzie JJ, Hanks GE, Maclean CJ, et al. Patterns of Care Study: Hodgkin’s disease relapse rates and adequacy of portals. Cancer 1983;52:2223–2226. 4. Owen JB, Sedransk J, Papak TF. National averages for process and outcome in radiation oncology: methodology of the Patterns of Care Study. Sem Radiat Oncol 1997;7:101–107. 5. Carde P, Hagenbeek A, Hayat M, et al. Clinical staging versus laparotomy and combined modality with MOPP versus ABVD in early-stage Hodgkin’s disease: the H6 twin randomized studies from the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Group. J Clin Oncol 1993;11:2258 –2272. 6. Hanks GE. Hodgkin’s disease. Successful dissemination of high technology treatment throughout the United States. Cancer Surveys 1985;4:477– 485. 7. Coia LR, Hanks GE. Complications from large field intermediate dose infradiaphragmatic radiation: an analysis of the
8. 9. 10. 11. 12.
13. 14.
Patterns of Care outcome studies for Hodgkin’s disease and seminoma. Int J Radiat Oncol Biol Phys 1988;15:29 –35. Hughes DB, Smith AR, Hoppe RT, et al. Treatment planning for Hodgkin’s disease: a Patterns of Care Study. Int J Radiat Oncol Biol Phys 1995;33:519 –524. Hancock SL, Tucker MA, Hoppe RT. Breast cancer after treatment of Hodgkin’s disease JNCI 1993;85:25–31. Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin’s disease. JAMA 1993;270:1949 –1955. Bartlett NL, Rosenberg SA, Mason J, et al. Brief chemotherapy and adjuvant radiotherapy for bulky or advanced stage Hodgkin’s disease. J Clin Oncol 1995;13:1080 –1088. Horning SJ, Hoppe RT, Mason J, et al. Stanford-Kaiser Permanente G1 study for clinical stage I to IIA Hodgkin’s disease: subtotal lymphoid irradiation versus vinblastine, methotrexate, and bleomycin chemotherapy and regional irradiation. J Clin Oncol 1997;15:1736 –1744. Bonfante V, Santoro A, Viviani S, et al. ABVD plus radiotherapy (subtotal nodal vs. involved field) in early-stage Hodgkin’s disease. (Abstr.) Proc ASCO 1994;13:373. Klasa RJ, Connors J, Hoskins P, et al. Early stage Hodgkin’s disease: impact of brief chemotherapy together with radiotherapy without staging laparotomy. (Abstr.) Proc ASCO 1994; 13:372.