- Email: [email protected]
Total dose, fraction size, and tumor volume in the local control of Hodgkin's disease
Inl. .I. Radiation Oncology Biol. Pbys., Vol. 15, pp. 25-28 Printed in the U.S.A. All rights resawd.
Copyri&t
0360-3016/88 $3.00 + .OO 0 1988 Pergamon Press plc
??Original Contribution
TOTAL DOSE, FRACTION SIZE, AND TUMOR VOLUME THE LOCAL CONTROL OF HODGKIN’S DISEASE KEVIN L. SCHEWE,
IN
M.D.,’ JUDY REAVIS, M.S.,’ LARRY E. KUN, M.D.* AND JAMES D. Cox,
M.D.3
‘MedicalCollege of Wisconsin, Milwaukee,WI; ‘St. Jude Children’s Research Hospital, Memphis, TN; and ‘Columbia-Presbyterian Medical Center, New York, NY There are importantuncertaintiesin the radiationtherapy of Hodgkin’s disease. These uncertaintiesare related to dose-fractionationand total dose necessary to control subclinical disease (prophylacticirradiation),small-size tumors, and especially bulky tumors such as those frequently encountered in the mediastinum. Data are lacking on total dose and tumor control as a function of tumor volume. A retrospective study was undertaken of patients with Hodgkin’s disease Stages I-III treated with radiation therapy alone at the Medical College of Wisconsin Affiliated Hospitals between 1970 and 1982. Detailed dose calculations of off-axis points were made to assign precise minimum doses to 1,304 separate lymph node regions. Treatment volumes received individual fractions of 150 cGy to 300 cGy and total doses of 30 Gy to 42 Gy. Tumor control was correlated with tumor size at presentation, fractionation schedule, and total dose. The results confirm the absence of a dose-response relationship for tumor control, between 30 Gy and 42 Gy total dose. In addition, there is no apparent difference in total dose required for larger tumors relative to that required for small tumors. Hodgkin’s disease, Tumor size, Total dose, Fractionation.
of Wisconsin was undertaken to investigate the following questions: (a) Can one identify a dose-response relationship between 30-45 Gy? (b) Is there a relationship between the size of nodal disease and irradiation control or tumoricidal dose?
INTRODUflION
The treatment goal in radiation therapy is to deliver a tumoricidal dose of ionizing radiation without exceeding normal tissue tolerance within the treatment volume, (uncomplicated cures). Dose-response data from various epithelial tumors such as head and neck and breast carcinoma demonstrate a relationship between the volume of tumor and the dose required for control.‘,5 In general, for epithelial tumors, larger volumes of disease require higher doses for control. Also, time-dose relationships do exist and are of importance when assessing control. However, very radiosensitive tumors such as seminoma and nodular lymphoma do not have such dose response relationships. There are some exceptions in the treatment of huge tumors, where doses of 25 Gy to 30 Gy are adequate to achieve tumor control. Hodgkin’s disease has been considered to have a doseresponse between 30-45 GY.~ However, independent studies of patients treated for Hodgkin’s disease appear to have shown no dose response above 30 GY.~,‘*This retrospective study of Hodgkin’s disease involving treatment with radiation therapy alone at the Medical College
METHODS
AND
MATERIALS
Patient characteristics From January 1970 to December 1982, 125 patients with Hodgkin’s disease Stages I-III were treated with radiation therapy with curative intent at the Medical College of Wisconsin and its affiliated hospitals. Thirteen patients with Stage III disease received combined modality therapy and are excluded from this study. In total, 112 patients received radiation therapy alone as their initial management. All patients were followed for a minimum of 29 months after treatment with a median follow-up of 89 months and a maximum of 183 months. The ages of the patients at initial presentation ranged from 6 years 8 months to 77 years, with the median age being 28 years. Of the 112 patients, 48% (54) were female and 52% (58) were male.
Reprint requests to: Kevin L. Schewe, M.D., Department of Radiation Oncology, Bishop Clarkson Memorial Hospital, 44th and Dewey Ave., P.O. Box 3328, Omaha, Nebraska 68 103-0328.
Accepted for publication 1 February 1988.
25
I. J. Radiation Oncology 0 Biology 0 Physics
26
Patients were staged according to the AJC modification of the Ann Arbor Staging Classification.* Fortyseven (42%) patients were clinically staged and sixty-five (58%) patients were pathologically staged. The histopathologic subtypes were classified according to the Rye modification of the Lukes and Butler system.” The distribution of initial stage and histopathologic subtypes is shown in Table 1. Radiotherapy technique In the first 3 years of the study, standard treatment
involved reduced fractionation of 3 fractions per week with 300 cGy. The patients treated with this regimen were irradiated with higher energy (25 MVX) equipment (Bolus was consistently used for superficial nodal regions). They were not treated with both anterior and posterior (API PA) ports each day as they are in our current 5 fraction per week schedule. Two patterns of fractionation were used: a. 300 300 300 b. 300 300
cGy cGy cGy cGy cGy
day day day day day
(AP) (PA) (AP) (AP) (AP)
1 3 5, etc. 1 3
We currently do not treat with 3 fractions per week. In 1973 the time-dose relationship evolved to our current standard fractionation schedule of 5 fractions per week with 150- 180 cGy. Continuous split course therapy was used in the 5 fraction per week schedule.7,12 In the split course technique, the mantle (upper extended field), was given a median dose of 18 Gy. Then para-aortic nodes and spleen or splenic pedicle were given 18 Gy. Next, the mantle was again treated to a specified completion dose with modification of custom blocks in the lung, hilar, or mediastinal areas if there had been interval regression of bulky disease. The para-aortic and spleen or splenic pedicle fields were then completed. Pelvic treatment, because of nodal disease, stage, or histology was started 2-4 weeks after completion of the para-aortic
Table 1. Patient population Stage
LD
by stage and histology
MC
LP
NS
Totals
IA IB
7 1
6
11
24 1>
25
IIB IIA
11
6
437
“‘: >
67
12 1
li720
IIIA IIIB
1
4 2
Totals
1
25
12
74
112
Note: Histopathologic subtypes as classified according to the Rye modification ofthe Lukes and Butler system are: Lymphocyte Deplete (LD), Mixed Cellularity (MC), Lymphocyte Predominant (LP), and Nodular Sclerosis (NS). Forty-seven (47) patients were clinically staged and 65 patients were pathologically staged.
July 1988, Volume 15, Number I
fields. In most of the continuous schedule patients, an “inverted Y” field encompassing the para-aortic nodes, spleen, or splenic pedicle and the pelvic nodes was used 2-4 weeks following the mantle field. All patients were treated with supervoltage equipment having energies of 6 MVX, 18 MVX, and 25 MVX. Doses at 1304 nodal sites were calculated by “off-axis irregular field” data. The following sites were analyzed for nodal size and dose: 1. Mantle: cervical, supraclavicular, axillary, mid-mediastinum, and low mediastinum. 2. Para-aortic and spleen or splenic pedicle. 3. Pelvis: iliac and inguinal. Parameters analyzed
Local control was correlated with fractionation schedule and total dose (less than 30, 30.1-33, 33.1-36, 36. l39,39.1-42, and greater than 42 Gy). We have correlated total dose with local control as affected by nodal size at the time of initial presentation. Nodal sizes were divided into subgroups of subclinical disease, 0.5 cm.-3.0 cm. and greater than 3.0 cm. Size was defined by actual physical measurements, X ray studies (chest X rays, CT scans, and lymphography) and operative reports. Response of nodal sites and permanent control (recurrence rates) were reviewed. Analysis of the local lymph node control rates were undertaken with corrections for the inherent inhomogeneity within the mantle field. Nodal recurrences were categorized as local, marginal, or out of treatment field. The patients with local failure had recurrent disease inside the treatment field. A marginal failure was recurrence at the radiation field edge defined by the shielding. Failures out of the treatment field were in unirradiated areas. Only local and marginal recurrences were counted as failures for correlation with irradiation parameters. RESULTS Of the 112 patients, there were 33 recurrences. Table 2 demonstrates recurrences as they relate to number of fractions per week. A greater infield recurrence rate (22.2%) was seen in the patients treated using our original 3 fractions per week, 300 cGy/fraction regimen. This compares with 8.6% for the 5 fractions per week, 150180 cGy/fraction schedule currently used (p > 0.05). Tables 3 and 4 demonstrate recurrence rates correlated with nodal size at presentation and total dose. From Table 3, our infield failure rate varied from 1.6% to 6.3% for nodal groups within the mantle field with doses above 30 Gy. In nodal groups with overt disease, (>0.5 cm nodal size), there were 2/93 (2.2%) infield failures with doses above 39 Gy, and lo/261 (3.8%) infield failure in the 30.1-39 Gy dose range. With doses above 30.1 Gy failure rates were not significantly reduced when using higher total doses. Control rates for larger tumor volumes were not improved with higher total doses.
Hodgkin’s
disease 0 K. L.
Table 2. Recurrence-fractions/week 5 fractions/week
3 fractions/week
In-Field Marginal Out-of-field
8 (8.6%) 4 (4.3%) 14(15.1%)
4 (22.2%) 1(5.6%) 2(11.1%)
Totals
26 (28%)
7 (38.9%)
Note: Ninety-three (93) patients were treated with five fractions per week, and 18 patients were treated with three fractions per week. One patient treated four days per week is excluded. Absolute numbers are shown with percentages in parenthesis.
failure occurred in the 36. I-
39 Gy dose range and two infield failures occurred in 39.1-42 Gy dose range. For doses above 30 Gy, the pelvic infield failure rate was 1.7% (3 nodal groups out of 179 treated). Para-aortic nodes were not assessed for size. Marginal recurrences (MR) occurred in the para-aortic fields at the dose ranges of: 1 MR I 30 Gy; 2 MR at 33.1-36 Gy; and 1 MR at 36.1-39 Gy. No local or marginal recurrences were observed in the irradiated spleen or splenic pedicle. Table 5 summarizes all infield failures as they relate to total dose. Overall, no increased incidence of infield failure was observed with total doses over 30 Gy. However, there was a significant (p < 0.0 1) increase in infield failures for doses less than 30 Gy versus doses greater than 30 Gy. Table 6 summarizes marginal and infield failures above 30.1 Gy related to nodal size at presentation. Overall, there was a trend toward higher recurrence rates with increasing nodal size at presentation which did not reach statistical significance.
DISCUSSION Dose-response data for several epithelial tumors clearly demonstrate a relationship between the volume of tumor and the dose required for control. Dose-response data for Hodgkin’s disease reflects a different tu-
Table 3. Mantle field: Recurrence rates* nodal size vs. dose Subclinical 530 Gray 30.1-33 33.1-36
3/13IF(23.1%) O/7 3/145 IF(Z.I%)
36.1-39
4/244 IF (I .6%) l/244 (MR (0.4%) o/115 O/16
39.1-42 >42 Totals
I l/540 (2.0%)
et al.
27
Table 4. Pelvic field: Recurrence rates* nodal size vs. dose
Type of recurrence
In the pelvis, one infield
SCHEWE
0.5-3 cm. O/5 2/45 IF (4.4%) 3145 MR (6.7%)
23.
0.5-3 cm.
23.1 cm.
O/8 O/4 21107 MR o/19 2119 IF O/9
-
-
O/6 l/5 IF -
oj O/I -
* Separate nodal recurrences are recorded per site. MR = Marginal recurrence; IF = In-field recurrence (local).
mor biology. There have been two schools of thought regarding dose-response relationships in Hodgkin’s disease. In 1966, Kaplan8 published a dose-response curve for Hodgkin’s disease indicating a nearly linear dose-response for doses up to and including 40 Gy. Fletcher and Shukovsky’ later pointed out deficiencies in Kaplan’s dose-response assessment and published a sigmoid-response curve for Hodgkin’s disease derived from the
same data. Their sigmoid response curve showed steeply rising probability of tumor control from about 20% control at 1,500 roentgens to greater than 90% control at 3,000 roentgens. They concluded that for megavoltage irradiation, doses of 3,500 cGy given up to 6 weeks will achieve at least a 95% control rate and that doses as high as 6,000 cGy do not prevent the appearance of a few recurrences. The Patterns of Care Study by Hanks et aL6 did not show a dose-response curve for infield failures above 3,000 cGy. Thus, our data concur with the observations by Fletcher and Shukovsky and the Patterns of Care Study. We could not demonstrate a dose-response beyond 30 Gy. We examined failure rates based on nodal size at presentation and total dose delivered. There was no apparent difference in total dose required for larger tumors relative to that required for small tumors. Consistent control was obtained with total doses of 30-40 Gy using our current standard fractionation schedule of 5 fractions per week with 150- 180 cGy per fraction. There is a correlation with higher recurrence rates and increasing nodal size at presentation. However, as this study demonstrates, using higher total doses (>40 Gy) for larger nodal volumes does not appear to improve local control. Thar
I cm.
O/2 012 l/27 IF (3.7%) l/27 MR (3.7%)
3/123 (IF (2.4%) l/50 IF (2.0%) O/4
4/64 IF (6.3%) l/28 IF (3.6%) O/II
9/227 (4.0%)
7/134(5.2%)
??Separate nodal recurrences (eg: cervical and supraclavicular corded per site. MR = Marginal recurrence; IF = In-field recurrence (local).
~30 Gray 30.1-33 33.1-36 36.1-39 39.1-42 >42
CO.5 cm.
Table 5. Summary of in-field failures related to total dose Dose range in gray
Number of nodal groups
130 30.1-36 36.1-39 >39
28 344 456 252
Recurrences 2 12 4
Percent recurrence 10.7 1.7 2.6 1.6
failures) are re-
p < 0.01 for doses 530 Gray versus doses >30 Gray i.e. 3128 versus 2211052.
28
I. J. RadiationOncology 0 Biology 0 Physics Table 6. Summary of marginal and in-field failures above 30.1 gray related to nodal size at presentation
Nodal size
Number of nodal groups
Recurrences
Percent recurrence
Subclinical 0.5-3.0 cm. >3.1 cm.
685 233 134
12 10 7
1.8 4.3 5.2
p > 0.05
for all comparisons.
et al. I2 examined infield failures in Hodgkin’s disease related to dose and tumor size. In patients with disease more than 6 cm, 3/3 failed at doses of 3 l-33 Gy, but no clear dose-response existed above 34 Gy and recurrences were noted at or above 40 Gy. Failure to control Hodgkin’s disease with radiation therapy may be more a function of treatment volume than total dose. Adequate treatment volumes are necessary to encompass the true extent of tumor. Patients with a large mediastinal mass or hilar disease may have micro-
July 1988, Volume 15, Number I
scopic tumor extension into the lung parenchyma and/ or pleura that is not radiographically detectable. Previous studies3*’’have identified “bulky” tumors in the mediastinum which seem to have recurred more frequently. The present study does not corroborate significantly more frequent recurrences with tumors greater than 3 cm. It is entirely possible that insufficient margins rather than an insufficient total dose may have accounted for higher failure rates with large mediastinal tumors. The University of Minnesota has demonstrated improved results in these patients by utilizing whole-lung or half-lung radiation. Their recurrence-free survival is equivalent to that reported to occur with combined modality treatment.’ Large volumes are necessary when treating mantle, para-aortic, and pelvic fields. Higher doses to large volumes do not increase local control but increase the risk for complications. Based upon fundamental dose-complication data4 one must conclude that nodal doses in excess of 40 Gy add only to morbidity and not local control.
REFERENCES 1. Arrigada, R., Mouriesse, H., Sarrazin, D., Clark, R.M., De-
boer, G.: Radiotherapy alone in breast cancer. I. Analysis of tumor parameters, tumor dose and local control: The experience of the Gustave-Roussy Institute and the Princess Margaret Hospital. Int. J. Radiat. Oncol. Biol. Phys. ll(10):
1751-1757,
1985.
2. Beahrs, O.H., Myers, M.H.: Hodgkin’s and Non-Hodgkin’s lymphoma. Manual for Staging of Cancer, 2nd edition. Philadelphia, J.B. Lippincott. 1983, pp. 227-232. 3. Carmel, R.J., Kaplan, H.S.: Mantle irradiation in Hodgkin’s disease. An analysis of technique, tumor eradication, and complications. Cancer 37: 28 13-2825, 1976. 4. Fazekas, J.T., Cox, J.D., Turner, W.M.: Irradiation of Stage I and II Hodgkin’s Disease. Am. J. Roentgenol. 123: 154-162,
1975.
5. Fletcher, G.H., Shukovsky, L.J.: The interplay of radiocurability and tolerance in the irradiation of human cancers. J. Radio. Electrol. 56: 383-400, 1975. 6. Hanks, G.E., Kinzie, J.J., White, R.L., Herring, D.F., Kramer, S.: Patterns of Care Outcome Studies: Results of
the national practice in Hodgkin’s disease. Cancer Q(4): 569-573,
1983.
7. Johnson, R.E., Ruhl, U., Johnson, S.K., Glover, M.: Split course radiotherapy of Hodgkin’s disease. Cancer 37(4): 1713-1717, 1976. 8. Kaplan, H.S.: Evidence for a tumoricidal dose level in the radiotherapy of Hodgkin’s disease. Cancer Rex 26(part I):
1221-1224, 1966. 9. Levitt, S.H., Chung, K.K.L., Aeppli, D.M., Bloomfield, C.D.: Radical treatment of Hodgkin’s disease with radiation therapy: Results of a 15-year clinical trial. Radiology 162: 623-630,1987.
10. Lukes, R.J., Butler, J.J.: The pathology and nomenclature of Hodgkin’s disease. Cancer Rex 26: 1063- 108 1, 1966. 11. Mauch, P., Goodman, R., Hellman, S.: The significance of mediastinal involvement in early stage Hodgkin’s disease. Cancer42:
1039-1045,1978.
12. Thar, T.L., Million, R.R., Hausner, R.J., McKetty, Marlene H.B.: Hodgkin’s disease, Stages I and II: Relationship of recurrence to size of disease, radiation dose, and number ofsites involved. Cancer43(3): 1101-l 105, 1979.
Copyri&t
0360-3016/88 $3.00 + .OO 0 1988 Pergamon Press plc
??Original Contribution
TOTAL DOSE, FRACTION SIZE, AND TUMOR VOLUME THE LOCAL CONTROL OF HODGKIN’S DISEASE KEVIN L. SCHEWE,
IN
M.D.,’ JUDY REAVIS, M.S.,’ LARRY E. KUN, M.D.* AND JAMES D. Cox,
M.D.3
‘MedicalCollege of Wisconsin, Milwaukee,WI; ‘St. Jude Children’s Research Hospital, Memphis, TN; and ‘Columbia-Presbyterian Medical Center, New York, NY There are importantuncertaintiesin the radiationtherapy of Hodgkin’s disease. These uncertaintiesare related to dose-fractionationand total dose necessary to control subclinical disease (prophylacticirradiation),small-size tumors, and especially bulky tumors such as those frequently encountered in the mediastinum. Data are lacking on total dose and tumor control as a function of tumor volume. A retrospective study was undertaken of patients with Hodgkin’s disease Stages I-III treated with radiation therapy alone at the Medical College of Wisconsin Affiliated Hospitals between 1970 and 1982. Detailed dose calculations of off-axis points were made to assign precise minimum doses to 1,304 separate lymph node regions. Treatment volumes received individual fractions of 150 cGy to 300 cGy and total doses of 30 Gy to 42 Gy. Tumor control was correlated with tumor size at presentation, fractionation schedule, and total dose. The results confirm the absence of a dose-response relationship for tumor control, between 30 Gy and 42 Gy total dose. In addition, there is no apparent difference in total dose required for larger tumors relative to that required for small tumors. Hodgkin’s disease, Tumor size, Total dose, Fractionation.
of Wisconsin was undertaken to investigate the following questions: (a) Can one identify a dose-response relationship between 30-45 Gy? (b) Is there a relationship between the size of nodal disease and irradiation control or tumoricidal dose?
INTRODUflION
The treatment goal in radiation therapy is to deliver a tumoricidal dose of ionizing radiation without exceeding normal tissue tolerance within the treatment volume, (uncomplicated cures). Dose-response data from various epithelial tumors such as head and neck and breast carcinoma demonstrate a relationship between the volume of tumor and the dose required for control.‘,5 In general, for epithelial tumors, larger volumes of disease require higher doses for control. Also, time-dose relationships do exist and are of importance when assessing control. However, very radiosensitive tumors such as seminoma and nodular lymphoma do not have such dose response relationships. There are some exceptions in the treatment of huge tumors, where doses of 25 Gy to 30 Gy are adequate to achieve tumor control. Hodgkin’s disease has been considered to have a doseresponse between 30-45 GY.~ However, independent studies of patients treated for Hodgkin’s disease appear to have shown no dose response above 30 GY.~,‘*This retrospective study of Hodgkin’s disease involving treatment with radiation therapy alone at the Medical College
METHODS
AND
MATERIALS
Patient characteristics From January 1970 to December 1982, 125 patients with Hodgkin’s disease Stages I-III were treated with radiation therapy with curative intent at the Medical College of Wisconsin and its affiliated hospitals. Thirteen patients with Stage III disease received combined modality therapy and are excluded from this study. In total, 112 patients received radiation therapy alone as their initial management. All patients were followed for a minimum of 29 months after treatment with a median follow-up of 89 months and a maximum of 183 months. The ages of the patients at initial presentation ranged from 6 years 8 months to 77 years, with the median age being 28 years. Of the 112 patients, 48% (54) were female and 52% (58) were male.
Reprint requests to: Kevin L. Schewe, M.D., Department of Radiation Oncology, Bishop Clarkson Memorial Hospital, 44th and Dewey Ave., P.O. Box 3328, Omaha, Nebraska 68 103-0328.
Accepted for publication 1 February 1988.
25
I. J. Radiation Oncology 0 Biology 0 Physics
26
Patients were staged according to the AJC modification of the Ann Arbor Staging Classification.* Fortyseven (42%) patients were clinically staged and sixty-five (58%) patients were pathologically staged. The histopathologic subtypes were classified according to the Rye modification of the Lukes and Butler system.” The distribution of initial stage and histopathologic subtypes is shown in Table 1. Radiotherapy technique In the first 3 years of the study, standard treatment
involved reduced fractionation of 3 fractions per week with 300 cGy. The patients treated with this regimen were irradiated with higher energy (25 MVX) equipment (Bolus was consistently used for superficial nodal regions). They were not treated with both anterior and posterior (API PA) ports each day as they are in our current 5 fraction per week schedule. Two patterns of fractionation were used: a. 300 300 300 b. 300 300
cGy cGy cGy cGy cGy
day day day day day
(AP) (PA) (AP) (AP) (AP)
1 3 5, etc. 1 3
We currently do not treat with 3 fractions per week. In 1973 the time-dose relationship evolved to our current standard fractionation schedule of 5 fractions per week with 150- 180 cGy. Continuous split course therapy was used in the 5 fraction per week schedule.7,12 In the split course technique, the mantle (upper extended field), was given a median dose of 18 Gy. Then para-aortic nodes and spleen or splenic pedicle were given 18 Gy. Next, the mantle was again treated to a specified completion dose with modification of custom blocks in the lung, hilar, or mediastinal areas if there had been interval regression of bulky disease. The para-aortic and spleen or splenic pedicle fields were then completed. Pelvic treatment, because of nodal disease, stage, or histology was started 2-4 weeks after completion of the para-aortic
Table 1. Patient population Stage
LD
by stage and histology
MC
LP
NS
Totals
IA IB
7 1
6
11
24 1>
25
IIB IIA
11
6
437
“‘: >
67
12 1
li720
IIIA IIIB
1
4 2
Totals
1
25
12
74
112
Note: Histopathologic subtypes as classified according to the Rye modification ofthe Lukes and Butler system are: Lymphocyte Deplete (LD), Mixed Cellularity (MC), Lymphocyte Predominant (LP), and Nodular Sclerosis (NS). Forty-seven (47) patients were clinically staged and 65 patients were pathologically staged.
July 1988, Volume 15, Number I
fields. In most of the continuous schedule patients, an “inverted Y” field encompassing the para-aortic nodes, spleen, or splenic pedicle and the pelvic nodes was used 2-4 weeks following the mantle field. All patients were treated with supervoltage equipment having energies of 6 MVX, 18 MVX, and 25 MVX. Doses at 1304 nodal sites were calculated by “off-axis irregular field” data. The following sites were analyzed for nodal size and dose: 1. Mantle: cervical, supraclavicular, axillary, mid-mediastinum, and low mediastinum. 2. Para-aortic and spleen or splenic pedicle. 3. Pelvis: iliac and inguinal. Parameters analyzed
Local control was correlated with fractionation schedule and total dose (less than 30, 30.1-33, 33.1-36, 36. l39,39.1-42, and greater than 42 Gy). We have correlated total dose with local control as affected by nodal size at the time of initial presentation. Nodal sizes were divided into subgroups of subclinical disease, 0.5 cm.-3.0 cm. and greater than 3.0 cm. Size was defined by actual physical measurements, X ray studies (chest X rays, CT scans, and lymphography) and operative reports. Response of nodal sites and permanent control (recurrence rates) were reviewed. Analysis of the local lymph node control rates were undertaken with corrections for the inherent inhomogeneity within the mantle field. Nodal recurrences were categorized as local, marginal, or out of treatment field. The patients with local failure had recurrent disease inside the treatment field. A marginal failure was recurrence at the radiation field edge defined by the shielding. Failures out of the treatment field were in unirradiated areas. Only local and marginal recurrences were counted as failures for correlation with irradiation parameters. RESULTS Of the 112 patients, there were 33 recurrences. Table 2 demonstrates recurrences as they relate to number of fractions per week. A greater infield recurrence rate (22.2%) was seen in the patients treated using our original 3 fractions per week, 300 cGy/fraction regimen. This compares with 8.6% for the 5 fractions per week, 150180 cGy/fraction schedule currently used (p > 0.05). Tables 3 and 4 demonstrate recurrence rates correlated with nodal size at presentation and total dose. From Table 3, our infield failure rate varied from 1.6% to 6.3% for nodal groups within the mantle field with doses above 30 Gy. In nodal groups with overt disease, (>0.5 cm nodal size), there were 2/93 (2.2%) infield failures with doses above 39 Gy, and lo/261 (3.8%) infield failure in the 30.1-39 Gy dose range. With doses above 30.1 Gy failure rates were not significantly reduced when using higher total doses. Control rates for larger tumor volumes were not improved with higher total doses.
Hodgkin’s
disease 0 K. L.
Table 2. Recurrence-fractions/week 5 fractions/week
3 fractions/week
In-Field Marginal Out-of-field
8 (8.6%) 4 (4.3%) 14(15.1%)
4 (22.2%) 1(5.6%) 2(11.1%)
Totals
26 (28%)
7 (38.9%)
Note: Ninety-three (93) patients were treated with five fractions per week, and 18 patients were treated with three fractions per week. One patient treated four days per week is excluded. Absolute numbers are shown with percentages in parenthesis.
failure occurred in the 36. I-
39 Gy dose range and two infield failures occurred in 39.1-42 Gy dose range. For doses above 30 Gy, the pelvic infield failure rate was 1.7% (3 nodal groups out of 179 treated). Para-aortic nodes were not assessed for size. Marginal recurrences (MR) occurred in the para-aortic fields at the dose ranges of: 1 MR I 30 Gy; 2 MR at 33.1-36 Gy; and 1 MR at 36.1-39 Gy. No local or marginal recurrences were observed in the irradiated spleen or splenic pedicle. Table 5 summarizes all infield failures as they relate to total dose. Overall, no increased incidence of infield failure was observed with total doses over 30 Gy. However, there was a significant (p < 0.0 1) increase in infield failures for doses less than 30 Gy versus doses greater than 30 Gy. Table 6 summarizes marginal and infield failures above 30.1 Gy related to nodal size at presentation. Overall, there was a trend toward higher recurrence rates with increasing nodal size at presentation which did not reach statistical significance.
DISCUSSION Dose-response data for several epithelial tumors clearly demonstrate a relationship between the volume of tumor and the dose required for control. Dose-response data for Hodgkin’s disease reflects a different tu-
Table 3. Mantle field: Recurrence rates* nodal size vs. dose Subclinical 530 Gray 30.1-33 33.1-36
3/13IF(23.1%) O/7 3/145 IF(Z.I%)
36.1-39
4/244 IF (I .6%) l/244 (MR (0.4%) o/115 O/16
39.1-42 >42 Totals
I l/540 (2.0%)
et al.
27
Table 4. Pelvic field: Recurrence rates* nodal size vs. dose
Type of recurrence
In the pelvis, one infield
SCHEWE
0.5-3 cm. O/5 2/45 IF (4.4%) 3145 MR (6.7%)
23.
0.5-3 cm.
23.1 cm.
O/8 O/4 21107 MR o/19 2119 IF O/9
-
-
O/6 l/5 IF -
oj O/I -
* Separate nodal recurrences are recorded per site. MR = Marginal recurrence; IF = In-field recurrence (local).
mor biology. There have been two schools of thought regarding dose-response relationships in Hodgkin’s disease. In 1966, Kaplan8 published a dose-response curve for Hodgkin’s disease indicating a nearly linear dose-response for doses up to and including 40 Gy. Fletcher and Shukovsky’ later pointed out deficiencies in Kaplan’s dose-response assessment and published a sigmoid-response curve for Hodgkin’s disease derived from the
same data. Their sigmoid response curve showed steeply rising probability of tumor control from about 20% control at 1,500 roentgens to greater than 90% control at 3,000 roentgens. They concluded that for megavoltage irradiation, doses of 3,500 cGy given up to 6 weeks will achieve at least a 95% control rate and that doses as high as 6,000 cGy do not prevent the appearance of a few recurrences. The Patterns of Care Study by Hanks et aL6 did not show a dose-response curve for infield failures above 3,000 cGy. Thus, our data concur with the observations by Fletcher and Shukovsky and the Patterns of Care Study. We could not demonstrate a dose-response beyond 30 Gy. We examined failure rates based on nodal size at presentation and total dose delivered. There was no apparent difference in total dose required for larger tumors relative to that required for small tumors. Consistent control was obtained with total doses of 30-40 Gy using our current standard fractionation schedule of 5 fractions per week with 150- 180 cGy per fraction. There is a correlation with higher recurrence rates and increasing nodal size at presentation. However, as this study demonstrates, using higher total doses (>40 Gy) for larger nodal volumes does not appear to improve local control. Thar
I cm.
O/2 012 l/27 IF (3.7%) l/27 MR (3.7%)
3/123 (IF (2.4%) l/50 IF (2.0%) O/4
4/64 IF (6.3%) l/28 IF (3.6%) O/II
9/227 (4.0%)
7/134(5.2%)
??Separate nodal recurrences (eg: cervical and supraclavicular corded per site. MR = Marginal recurrence; IF = In-field recurrence (local).
~30 Gray 30.1-33 33.1-36 36.1-39 39.1-42 >42
CO.5 cm.
Table 5. Summary of in-field failures related to total dose Dose range in gray
Number of nodal groups
130 30.1-36 36.1-39 >39
28 344 456 252
Recurrences 2 12 4
Percent recurrence 10.7 1.7 2.6 1.6
failures) are re-
p < 0.01 for doses 530 Gray versus doses >30 Gray i.e. 3128 versus 2211052.
28
I. J. RadiationOncology 0 Biology 0 Physics Table 6. Summary of marginal and in-field failures above 30.1 gray related to nodal size at presentation
Nodal size
Number of nodal groups
Recurrences
Percent recurrence
Subclinical 0.5-3.0 cm. >3.1 cm.
685 233 134
12 10 7
1.8 4.3 5.2
p > 0.05
for all comparisons.
et al. I2 examined infield failures in Hodgkin’s disease related to dose and tumor size. In patients with disease more than 6 cm, 3/3 failed at doses of 3 l-33 Gy, but no clear dose-response existed above 34 Gy and recurrences were noted at or above 40 Gy. Failure to control Hodgkin’s disease with radiation therapy may be more a function of treatment volume than total dose. Adequate treatment volumes are necessary to encompass the true extent of tumor. Patients with a large mediastinal mass or hilar disease may have micro-
July 1988, Volume 15, Number I
scopic tumor extension into the lung parenchyma and/ or pleura that is not radiographically detectable. Previous studies3*’’have identified “bulky” tumors in the mediastinum which seem to have recurred more frequently. The present study does not corroborate significantly more frequent recurrences with tumors greater than 3 cm. It is entirely possible that insufficient margins rather than an insufficient total dose may have accounted for higher failure rates with large mediastinal tumors. The University of Minnesota has demonstrated improved results in these patients by utilizing whole-lung or half-lung radiation. Their recurrence-free survival is equivalent to that reported to occur with combined modality treatment.’ Large volumes are necessary when treating mantle, para-aortic, and pelvic fields. Higher doses to large volumes do not increase local control but increase the risk for complications. Based upon fundamental dose-complication data4 one must conclude that nodal doses in excess of 40 Gy add only to morbidity and not local control.
REFERENCES 1. Arrigada, R., Mouriesse, H., Sarrazin, D., Clark, R.M., De-
boer, G.: Radiotherapy alone in breast cancer. I. Analysis of tumor parameters, tumor dose and local control: The experience of the Gustave-Roussy Institute and the Princess Margaret Hospital. Int. J. Radiat. Oncol. Biol. Phys. ll(10):
1751-1757,
1985.
2. Beahrs, O.H., Myers, M.H.: Hodgkin’s and Non-Hodgkin’s lymphoma. Manual for Staging of Cancer, 2nd edition. Philadelphia, J.B. Lippincott. 1983, pp. 227-232. 3. Carmel, R.J., Kaplan, H.S.: Mantle irradiation in Hodgkin’s disease. An analysis of technique, tumor eradication, and complications. Cancer 37: 28 13-2825, 1976. 4. Fazekas, J.T., Cox, J.D., Turner, W.M.: Irradiation of Stage I and II Hodgkin’s Disease. Am. J. Roentgenol. 123: 154-162,
1975.
5. Fletcher, G.H., Shukovsky, L.J.: The interplay of radiocurability and tolerance in the irradiation of human cancers. J. Radio. Electrol. 56: 383-400, 1975. 6. Hanks, G.E., Kinzie, J.J., White, R.L., Herring, D.F., Kramer, S.: Patterns of Care Outcome Studies: Results of
the national practice in Hodgkin’s disease. Cancer Q(4): 569-573,
1983.
7. Johnson, R.E., Ruhl, U., Johnson, S.K., Glover, M.: Split course radiotherapy of Hodgkin’s disease. Cancer 37(4): 1713-1717, 1976. 8. Kaplan, H.S.: Evidence for a tumoricidal dose level in the radiotherapy of Hodgkin’s disease. Cancer Rex 26(part I):
1221-1224, 1966. 9. Levitt, S.H., Chung, K.K.L., Aeppli, D.M., Bloomfield, C.D.: Radical treatment of Hodgkin’s disease with radiation therapy: Results of a 15-year clinical trial. Radiology 162: 623-630,1987.
10. Lukes, R.J., Butler, J.J.: The pathology and nomenclature of Hodgkin’s disease. Cancer Rex 26: 1063- 108 1, 1966. 11. Mauch, P., Goodman, R., Hellman, S.: The significance of mediastinal involvement in early stage Hodgkin’s disease. Cancer42:
1039-1045,1978.
12. Thar, T.L., Million, R.R., Hausner, R.J., McKetty, Marlene H.B.: Hodgkin’s disease, Stages I and II: Relationship of recurrence to size of disease, radiation dose, and number ofsites involved. Cancer43(3): 1101-l 105, 1979.