- Email: [email protected]
Hip stiffness following mixed conformal neutron and photon radiotherapy: A dose-volume relationship
Int. .I. Radiation
Oncology
Biol.
Phys., Vol. 35. No. 4. pp. 693-699, 1996 Copyright 0 1996 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/96 $15.00 + .OO
PII: SO360-3016(96)00136-S
ELSEVIER
0 Clinical Original Contribution HIP STIFFNESS FOLLOWING PHOTON RADIOTHERAPY:
MIXED CONFORMAL NEUTRON AND A DOSE-VOLUME RELATIONSHIP
PAUL J. CHUBA, PH.D., M.D., RENU SHARMA, B.S., C.M.D., MARK YUDELEV, MS., MARIE DUCLOS, M.D., FALAH SHAMSA, PH.D., SUSAN GIACALONE, R.N., COLIN G. ORTON, PH.D., RICHARD L. MAUGHAN, PH.D. AND JEFFREY D. FORMAN, M.D. Department of Radiation Oncology, Wayne State University, Gershenson Radiation Oncology Center, Detroit, MI, USA Purpose: To determine the relationship between dose, volume, and the incidence of hip stiffness in patients who received conformal neutron irradiation for nrostate cancer. Methods and Materials: A series of dose-searching studies using neutron irradiation for prostate cancer were performed to determine the optimal dose, fraction size, field size, technique, and proportions of photon and neutron dose. Neutron doses ranged fro& 9 to 20 Gy and photon doses ranged frbmb to 38 Gy: Data were analyzed by using a hip stiffness grading scale. Results: Hip stiffness was recorded on follow-up examination in 30% of patients (40 out of 132) treated with fast neutrons or mixtures of fast neutron and photon radiation for prostate cancer. Hip stiEness was categorized as none (Grade 0,92 patients), mild (Grade 1,24 patients), moderate (Grade 2,10 patients), or severe (Grade 3,6 patients). The incidence of hip stiffness differed significantly by dose and volume in the five dose levels studied @ < 0.001). Conclusions: By using a mixture of conformal neutron and photon irradiation and limiting the total neutron dose to less than 13 Gy, hip stiffness toxicity could be reduced to acceptable levels. Radiotherapy side effects, Fast-neutron radiation
INTRODUCTION To improve the therapeutic ratio for fast neutron irradiation for prostate cancer, it is necessary to increase local tumor eradication without an increase in normal tissue complications as compared to photon irradiation. Although prior experiences have confirmed an improvement in local control, the reported complication rates have been prohibitively high (15). The severity of these chronic complications depends on the total dose delivered, the dose fractionation, the interfraction interval, the volume, and the quality of the radiation, among other factors (6, 11, 13, 20). In general, the late effects of high linear energy transfer radiations such as fast neutrons may be greater than x-rays when similar acute toxicities are reached (2, 5, 8-10, 13, 14, 21, 28). No simple formula exists to predict normal tissue tolerances for fast neutron irradiation based on experience with photon radiation. For this reason dose-seeking studies for fast neutron radiotherapy must be conducted separately.
Randomized clinical trials have demonstrated the usefulness of fast neutron treatment and mixed fast neutron/ photon treatment for prostate cancer (15, 22). Reduced toxicity was shown to be related to the use of shaped neutron beams with improved depth dose characteristics. However, the use of shaped neutron fields in the Radiation Therapy Oncology Group experience still resulted in a 10% rate of severe (Grade 3 and 4) morbidity (22). More recent data indicate that conformal photon therapy is associated with decreased morbidity for prostate cancer (19, 23). To test the hypothesis that conformal neutron radiotherapy would result in a decreased rate of morbidity, three prospective dose-finding studies have been conducted at this institution. The goal of these studies was to help define the optimal dose and volume of conformal neutron irradiation for prostate cancer. The early results of these trials suggested that hip stiffness was a dose-limiting toxicity in pelvic neutron radiotherapy. In this report, combined neutron and photon doses leading to hip stiffness are analyzed. Clinical follow-up and normal tissue
Reprint requests to: Jeffrey D. Forman, M.D., Gershenson Radiation Oncology Center, 3990 John R., Detroit, MI 482012097. Acknowledgements-We thank Tongming He for expert assis-
tance with the GRATIStm 3D treatment-planning system, Prof. M. Joiner (UK) for stimulating discussions, and Gail Frasure C.T.R. and Chris Zuniga for data retrieval. Acceptedfor publication 1 March 1996. 693
I. J. Radiation Oncology
694
0 Biology
Volume 35. Number 3. 1996
0 Physics
Table I. Patient characteristics SNP-9 Neutron dose* Photon dose Composite dose’ Stage Mean age Median follow-up Number of patients Number treated to large field* African American Caucasian
9 GY 38 Gy 69.5 Gy A. B 66 10 months 51 0 16 35
SNP-IO IO Gy 38 Gy 73 Gy
A. B 66 4 months 27 0 3 24
and doses by group MNP-13
MNP- I5
N-20
13 Gy 25 Gy 70.5 Gy C, D 60 16 months 5 4 0 5
1s Gy 18 Gy 70.5 Gy C. D 65 16 months 43 36 17 26
70 Gy 0 Gy 70 Gy (1. D 63 5 months 6 5 0 6
* Doses quoted are the prescribed dose to the tumor volume. ’ Composite doses were calculated for this purpose by multiplying the neutron dose by RBE = 3.5 and adding the photon dose. t Larger pelvic fields were treated as described in Methods and Materials. SNP = sequential neutron photon, MNP = mixed neutron photon, N = neutron only.
dose information were used in an attempt to define the critical dose-volume relationships necessary to avoid hip stiffness.
METHODS
AND MATERIALS
Follow-up data from three prospective Phase II protocols in which fast neutron therapy or mixtures of photons and neutrons were used in the radiotherapy of prostate cancer patients at Harper Hospital were studied. Neutron and photon treatments were delivered with fully conforma1 beam shaping in all cases. All patients had completed radiotherapy at the time of analysis. Neutrons were produced from a superconducting cyclotron with a 48 MeV deuteron beam incident on an internal beryllium target (18). The source-to-axis distance was 182.9 cm, D,,, was 0.9 cm, and the percent depth dose at 10 cm was 63.9%. Photon fields were treated with 15 MV photons. Patients were simulated conventionally using polyurethane cradle immobilization, bladder contrast, and urethrogram. Treatment planning computed tomography scan immediately following simulation was used to provide axial image data for use in treatment planning system (26, 27). Conformal fields were designed to provide a minimum margin of 1.5 Table 2. Hip stiffness grading scale Grade
Severity
0
none
1
mild
2
moderate
3
severe
4
disabling
Description No hip stiffness or limitation of range of motion Mild hip stiffness or soreness or limitation of range of motion Interference with motion of hip joint, moderate pain Interference with normal daily activity, severe pain Severe limitation of ambulation or requiring surgery
cm from the gross tumor volume to the block edge. For each axial CT image, four anatomes were contoured for use in constructing dose-volume histograms; the femoral head (FH), the adductor muscle group (AM), the iliopsoas muscle group (IPM), and the gluteus medius and minimus together with the tensor fascia lata (GM). In one protocol 80 patients with early stage, low to intermediate grade prostate cancer (Tl -2NxM0, Gleason score < 8) were treated with sequential neutron followed by photon (SNP) irradiation between April 1993 and May 1994. Two dose levels were tested. Either 9.0 (5 1 patients) or 10.0 (29 patients) neutron Gy were delivered in 1.O Gy daily fractions to the prostate (boost volume) followed by 38 photon Gy in 2.0 Gy daily fractions to the prostate, seminal vesicles, and periprostatic lymph nodes. Photon treatments were delivered using a four-field box; however, for the neutron treatment, opposed lateral fields were paired with conformal nonaxial right and left anteriorinferior-superior oblique fields as previously described (16, 17, 19, 26). Forty-nine patients with locally advanced prostate cancer (T3-T4, NO-Nl, Gleasons score 5 7) were treated with concurrently mixed neutron photon radiation (MNP group) between October 1992 and April 1994. One patient was lost to follow-up. Two dose levels were used. Fortythree patients received a total of 15 neutron Gy to the prostate and seminal vesicles delivered in 1 Gy fractions 3 days per week (Mondays, Wednesdays, and Fridays) plus 18 photon Gy delivered in 1.8 Gy fractions 2 days per week (Tuesdays and Thursdays). Pelvic lymph nodes were treated similarly except that 0.6 Gy daily neutron fractions were used. Five patients received a total of 13 neutron Gy and 25.2 photon Gy using similar fractionation. The dose differential for treating pelvic lymph nodes was accomplished by using a partial transmission block (tungsten rods) (16, 26). Six patients with locally advanced prostate cancer were treated with neutrons alone (N group) between October 1992 and March 1994. These patients received 12.5 Gy
Hip stiffness following mixed conformal radiotherapy
0 P. J. CHUBA
695
et al.
Table 3. Incidenceof hip stiffnessby group SNP-9 Total patients Grade0 Grade 1 Grade2 Grade3 Grade > 0 Grade> 1
SNP-10
MNP-13
MNP-15
N-20
51
21
5
43
6
42(83%)
21(78%) 5 (19%)
3 (60%)
25 (58%) 7(16%) 6 (14%) 5 (12%) 18(42%)
1(17%) 3(50%) 1(17%)
9 (18%) 0 (0%) 0 (0%) 9(18%)
0 (0%)
0 (0%)
1 (4%) 0 (0%)
0 (0%)
6 (22%) 0(4%)
2(40%) 2(40%)
2 (40%)
11 (26%)
1 (17%) 5(83%) 2(330/o)
SNP = sequentialneutronphoton. MNP = mixed neutronphoton,N = neutrononly
to the prostate, seminal vesicles, and pelvic lymph nodes in 1.25 Gy daily fractions followed by a nonaxial boost to the prostate and seminal vesicles of 7.5 Gy in 1.25 Gy daily fractions. Characteristics of patients making up the five dose levels investigated are shown in Table 1. The median followup for the MNP protocol was 16 months (range: 4 to 24 months), for the SNP protocol 10 months (range: 3 to 16 months), and for the neutron only patients 5 months (range: 3 to 24 months). As the neutron dose increased the photon dose decreased so that the composite neutron plus photon dose (to tumor) for all patients was estimated at 69.5 to 73 Gy assuminga relative biologic effectiveness of 3.5. All patients in the two SNP groups were treated with small fields, but only nine patients in the two MNP groups and N-20 groups combined had small volumes treated (Table 1). A scale for grading hip stiffness was developed (Table 2). Mild, moderate, severe, and disabling hip stiffness was
assigned a grade of 1 to 4 according to the scale. We applied chi-square tests to assessassociationsamong the relative incidence of graded complications between groups. Fisher’s exact test was used for evaluating associations when the cell sizes in the contingency tables was small. The Mantel-Haenszel method was used to adjust for confounding effects.
RESULTS Hip stiffness ( Grade 0; Table 3) was observed in 9 out of 51 (18%) patients who received 9 neutron Gy (SNP9) 6 out of 27 (22%) who received 10 neutron Gy (SNPlo), 2 out of 5 (40%) who received 13 neutron Gy (MNP13), 18 out of 43 (41%) who received 15 neutron Gy (MNP- 15) and 5 out of 6 (83%) who received 20 neutron Gy (N-20). A bar graph illustrating the relative incidence of any hip stiffness (> Grade 0) or severe (Grade 3) hip stiffness in the five groups is shown in Fig. 1. The differ-
100% 90% 80%
42%
22% 18%
r-l
SNP-9
SNP-10
MNP-13
MNP-15
N-20
Fig. 1. Relative incidenceof hip stiffnessby patient group. The incidenceof any hip stiffness(> Grade0) in the five patientgroupsconsideredis illustratedwith an openbar. Incidenceof severe(Grade3) hip stiffnessisillustrated with shadedbar.
696
I. J. Radiation Oncology 0 Biology 0 Physics
ence in the incidence of complications between these five dose-level groups was statistically significant (p < 0.01). Moderate to severe hip stiffness was also related to the proportion of neutron dose (Table 3, Figure l), and was statistically significant @ < 0.01). One of 80 patients (I .2%) receiving less than 13 neutron Gy complained of Grade 2 hip stiffness, however Grade 2 or Grade 3 hip stiffness was recorded for 2 out of 5 (40%) patients who received 13 neutron Gy, in 11 out of 43 (26%) patients who received 15 neutron Gy, and in 2 out of 6 (33%) patients who received 20 neutron Gy. As shown in Table 4, 23 out of 45 (51%) of all patients who were treated to pelvic lymph node regions but only 17 out of 87 (20%) of all patients treated to small fields complained of hip stiffness. The effect of volume was statistically significant whether the entire series @ < 0.001, five dose levels) or only the three dose levels MNP-13, MNP-15, and N-20 were considered (JJ < 0.001, patients receiving at least 13 neutron Gy). Results were compared in three subcategories: low neutron dose/low volume (78 SNP patients), high neutron dose/low volume (9 patients from MNP-13, MNP- 15, and N-20 dose levels), and high neutron dose/high volume (45 patients from MNP-13, MNP-15, and N-20 dose levels). In the SNP-9 and SNP10 groups combined, all hip stiffness was Grade 1, with a single Grade 2 complication for a total of 14 out of 78 (19%) patients. Of the nine patients who received 13 Gy or more of neutron treatment and small fields, two out of nine (22%) had hip stiffness (one Grade 1 and one Grade
Table 4. Incidence
of hip stiffness in patients treated with large* or small fields
Low neutron dose/low volume SNP-9 Grade 1 Grade 2 Grade 3 Total
9 0 0 9/51
(18%) (0%) (0%) (18%)
High neutron dose/low volume MNP-13 Grade I Grade 2 Grade 3 Total
0 0 0 O/l
(0%) (0%) (0%) (0%)
High neutron dose/high volume MNP-13 Grade 1 Grade 2 Grade 3 Total
0 2 0 2/4
(0%) (50%) (0%) (50%)
SNP-10 5 1 0 6/27
(19%) (4%) (0%) (22%)
MNP-15 1 (14%) 0 (0%) 0 (0%) l/7 (14%)
MNP-15 6 6 5 17/36
(17%) (17%) (14%) (47%)
N-20 0 (0%) 0 (0%) 1 (0%) l/l (100%)
N-20 3 1 0 4/5
(60%) (20%) (0%) (80%)
* Large Fields were treated as described under Methods and Materials. SNP = sequential neutron photon, MNP= mixed neutron photon, N = neutron only.
Volume 35. Number 4. 1996
3). Of 45 patients who received 13 neutron Gy or more and larger pelvic field treatment, 23 (51%) exhibited hip stiffness. In the latter group nine Grade 2 and five Grade 3 complications were recorded. When the relationship of dose level to hip stiffness (Table 3) was adjusted for the (confounding) effect of volume (Table 4) or vice-versa. the p-value was reduced (p = 0.07) to borderline levels. implying that the dose and volume effects were strongly associatedwith each other as well as with complications. Two axial CT images with femoral heads (FH), and three extrapelvic muscle groups illustrated are shown in Figure 2. The possible anatomic significance of the volumes of these normal tissues irradiated was investigated by determining the dose distributions to these structures by constructing dose-volume histograms. Dose-volume histograms (DVHs) for these four anatomes using the two different field sizes used in the MNP and N-20 groups are shown in Figure 3. The cumulative DVHs to the femoral heads, the adductor, and the iliopsoas muscle groups were quite similar whether or not large pelvic fields were included in the planning. For the gluteus muscle group however, the DVH curves differed markedly according to field size. Small field treatment resulted in significant sparing of the gluteus muscle group (Fig. 3).
DISCUSSION Connective tissue is vulnerable to radiation damage. The tolerance doses(TD5/5 and TD50/5) for femoral head necrosis for photon radiotherapy have been establishedat 52 and 65 Gy, respectively (7). Tolerance dosesto muscle are not well characterized, although skeletal muscle makes up a large proportion of body massand is extremely well vascularized. On a cellular level, it is thought that microvasculature injury is reflected secondarily in late effects to the connective tissue stroma. Alternatively, direct radiation damage to parenchymal cells may lead to formation of granulation tissue. Hip stiffness in pelvic radiotherapy has been reported previously (4), and has been assumedto be related to fibrosis of the soft tissue adjacent to the joint. In a series reported by Duncan ef al., 18 patients had received 16 neutron Gy in 20 fractions over 4 weeks (4). Their machine was low in energy and had a relatively high relative biologic effectiveness. Five patients had severe hip stiffness. By comparison to the 132 patients considered here. 40 (30%) described hip stiffness, 6 were severe. The patients treated with greater dosesof neutron radiation exhibited a significantly greater incidence and severity of hip stiffness (Table 1). Of the six patients with severe hip stiffness described here, two had radiologic findings in the femoral head. It, therefore, appeared that severe hip stiffness was partly related to a direct effect on bone. For less severe hip stiffness, no osteoradionecrosis was evident and an effect on muscle was investigated. Using DVHs. it was shown that
Hip stiffness following mixed conformal radiotherapy 0 P. J. CHUBA
et al.
697
a
b
Fig. 2. Location of the femoral head, gluteus, adductor, and iliopsoas anatomes. Axial computed tomograms at the level of the femoral head (a) or the pubic symphysis (b) are shown. The GM muscle group (gluteus medius, gluteus minimus, and tensor fascia lata; most lateral), the AM muscle group (adductor magnus, adductor longus, adductor brevis; most medial), and the IPM muscle group (iliopsoas associated; most anterior), are outlined. Anteroposterior and right lateral beam edges are indicated for reference.
the reduced incidence of hip stiffness in patients treated with smaller fields could be attributed to sparing of the GM muscle group (Fig. 3). This group is located laterally, in the region of relative dose maxima for the neutron treatments. Because the patient groups differed in field arrangements, field sizes, and in the sequencing of photon and
neutron treatments, as well as in the proportion of the total dose delivered with photon or neutron treatments, any of these factors could have been responsible for the differences in hip stiffness incidence. Data presented here show that both the total neutron dose (or proportion of the dose delivered with neutrons) and the volume of normal tissue irradiated were significant factors. The incidence of hip
698
I. J. Radiation Oncology 0 Biology 0 Physics
0
20
40
60
80
Volume 3.5,Number 4. 1996
100
0
20
40
60
100
80
80 60
b.
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
20
40
60
80
100
100 80 60
d.
0
Patient 1
Patient 2
Fig. 3. Cumulative dose-volume histograms for (a) femoral head (FH), (b) gluteus muscle (GM), (c) adductor muscle (AM), and (d) iliopsoas muscle (IPM) anatomes in two patients from the Mm-15 group. Plots of percent volume irradiated vs. percent dose are shown for four anatomes in two individual patients. Dose-volume histograms using large fields are shown with solid lines, small fields with dashed lines.
stiffness was greater when the proportion of dose delivered with neutrons was greater (Table 3) or when larger pelvic fields were used (Table 4). Although the effect of volume could not be completely separated from the effect of dose level, patients who received high neutron dosehigh volume treatment had a higher incidence (5 1%) and much greater severity of complications (Table 4). When the dose effects were adjusted for the effects of volume, they remained statistically significant. The composite doses (Table 1) for mixed neutron photon treatments calculated by using an RBE of 3.5 are not useful for predicting normal tissue effects. An estimate of the actual composite bioeffective dose (20) delivered to the hip joint will require knowledge of the dose per fraction at depth and the actual relative biologic effectiveness.
For mixed neutron/photon treatments, the influence of the increase in the neutron portion of the dose on the composite dose is greater with higher RBE. Several groups have reported methods for modeling dose-effect relationships for complications (1, 3, 11, 12,24, 25) and by using a composite bioeffect dose for mixed treatments, the dose-response for hip stiffness can be fitted to a sigmoid curve using a logistic function. Such an analysis can be used to predict an RBE for hip stiffness and other-sideeffects for the fast neutron therapy (M. Yudelev, oral communication, March 1995). In conclusion, hip stiffness was found to be a doselimiting toxicity for conformal neutron irradiation ofprostate cancer. As we make progress in defining the clinical, biochemical, and histologic local control in these patients,
Hip
stiffness following mixed conformal radiotherapy 0 P. J. CHUBA
the effect of these dose limitations on improving the therapeutic ratio will become evident. By using conformal mixed neutron/photon doses of less than 13 neutron Gy, the hip stiffness incidence was low, and severe hip stiff-
et ul.
699
ness was avoided. A volume effect was demonstrated with significantly reduced toxicity with sparing of the GM muscle group. Severe hip stiffness appeared to be related to a direct effect on bone.
REFERENCES 1. Battermann, J. J.; Hart, G. A. M.; Breur, K. Dose-effect relations for tumour control and complication rate after fast neutron therapy for pelvic tumours. Br. J. Radiol. 54:899904; 1981. 2. Catterall, M.; Errington, R. D.; Bewley, D. K. A comparison of clinical and laboratory data on neutron therapy for locally advanced tumors. Int. J. Radiat. Oncol. Biol. Phys. 13:17831791; 1987. 3. Cohen, L.; Schultheiss, T. E.; Hendrickson, F. R.; Mansell, J.; Saroja. K. R.; Lennox, A. Normal tissue reactions and complications following high-energy neutron beam therapy: I. Crude response rates. Int. J. Radiat. Oncol. Biol. Phys. 1673-78; 1989. 4. Duncan, W.; Arnott, S. J.; Jack, W. J. L.; Orr, J. A.; Kerr, G. R.; Williams, J. R. Results of two randomised trials of neutron therapy in rectal adenocarcinoma. Radiother. Oncol. 8: 191-198; 1987. 5. Duncan, W.; Williams, J. R.; Kerr, G. R.; Amott, S. J.; Quilty, P. M.; Rodger, A.; Macdougall, R. H.; Jack, W. J. L. An analysis of the radiation related morbidity observed in a randomized trial of neutron therapy for bladder cancer. Int. J. Radiat. Oncol. Biol. Phys. 122085-2092; 1986. 6. Ellis, F.; Weatherbum, H. RBE and clinical response in radiotherapy with neutron beams. Br. J. Radiol. 57:817; 1984. 7. Emami, B.; Lyman, J.; Brown, A.; Coia, L.; Goitein, M.; Munzenrider. J. E.; Shank, B.; Solin, L. J.; Wesson, M. Tolerance of normal tissue to therapeutic radiation. Int. J. Radiat. Oncol. Biol. Phys. 21:109-122; 1991. 8. Errington, R. D.; Ashby, D.; Gore, S. M.; Abrams, K. R.: Myint. S.; Bonnett, D. E.; Blake, S. W.; Saxton, T. E. High energy neutron treatment for pelvic cancers: Study stopped because of increased morbidity. Br. Med. J. 302: 1045-105 1; 1991. 9. Griffin, T. W.: Martz, K. L.; Laramore, G. E.; Thomas, F. J.; Maor. M. H.; Hendrickson, F. R.; Parker, R. G.; Richter, M. P.; Davis. L. W. High energy (42-66 MeV reactions) fast neutron dose optimization studies in the head and neck. thorax, upper abdomen, pelvis, and extremities. Radiother. Oncol. 19:307-316; 1990. 10. Griffin, T.; Pajak, T.; Laramore, G.; Davis, L. Analysis of neutron radiotherapy treatment complications. Bull. Cancer (Paris) 73:582-586; 1986. 11. Johns. H.; Joiner, M. C. A single method for fitting curves to dose-effect data for functional damage. Int. J. Radiat. Biol. 60:533-541: 1991. 12. Joiner, M. C.; Denekamp, J.: Maughan, R. L. The use of ‘top-up’ experiments to investigate the effect of very small doses per fraction in mouse skin. Int. J. Radiat. Biol. 49:565580: 1986. 13. Laramore, G. E.; Austin Seymour, M. M. Fast neutron radiotherapy in relation to the radiation sensitivity of human organ systems. Adv. Biol. 15:153-192; 1992. 14. Laramore, G. E.: Griffeth. J. T.; Boespflug, M.; Pelton, J. G.: Griffin, T. W.: Griffin, B. R.; Russell, K. J.; Kroh, W. Fast neutron radiotherapy results for sarcomas of soft tissue. bone, and cartilage. Am. J. Clin. Oncol. 12:320326; 1989.
15. Laramore, G. E.; Krall, G. E.; Thomas, F. J.; Griffin. T. W.; Maor, M. H.; Hendrickson, F. R. Fast neutron radiotherapy for locally advanced prostate cancer: Results of an RTOG randomized study. Int. J. Radiat. Oncol. Biol. Phys. 11:1621-1627; 1985. 16. Leibel, S. A.; Kutcher, G. J.; Mohan, R.; Harrison, L. B.: Armstrong, J. G.: Zelefsky, M. J.; Losasso, T. J.; But-man, C. M.; Mageras, G. S.; Chui, C. S.: Brewster, L. J.: Masterson, M. E.; Lo, Y. C.; Ling, C. C.; Fuks, F. Three-dimensional conformal radiation therapy at the Memorial Sloan Kettering Cancer Center. Semin. Radiat. Oncol. X74-289; 1992. 17. Marsh, L. H.: Ten Haken, R. K.; Sandler, H. M. A customized nonaxial external beam technique for treatment of prostate carcinoma. Cancer 45:724-727; 1992. 18. Maughan, R. L.; Powers, W. E.; Blosser, H. G. A superconducting cyclotron for neutron radiotherapy. Med. Phys. 21:779-785; 1994. 19. Mesina, C. F.; Sharma, R.; Rissman, L. S.; Geering, L.; He, T.; Forman, J. D. Comparison of a conformal nonaxial boost with a four-field boost technique in the treatment of adenocarcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 30:427430; 1994. 20. Orton, C. G. Recent developments in time-dose modelling. Aust. Phys. Eng. Sci. Med. 14:57-&t; 1991. 21. Parker, R. G.; Bumison, C. M.; Sadeghi, A. Normal-tissue tolerance to fast-neutron teletherapy. Front. Radiat. Ther. Oncol. 23:185-193; 1989. 22. Russell, K. J.; Laramore, G. E.; Krall, J. M.; Thomas, F. J.; Maor. M. H.; Hendrickson, F. R.; Krieger, J. N.; Griffin, T. W. Eight years experience with neutron radiotherapy in the treatment of stages C and D prostate cancer: Updated results of the RTOG 7704 randomized clinical trial. Prostate 1 I :183-193: 1987. 23. Sandler, H. M.; Perez-Tamayo, C.; Ten Haken, R. K.; Lichter, A. S. Dose escalation for stage C (T3) prostate cancer: Minimal rectal toxicity observed using conformal therapy. Radiother. Oncol. 23:53-54; 1991. 24. Schultheiss, T. E.; Cohen, L.: Mansell, J. Normal tissue reactions and complications following high-energy neutron beam therapy II: Complication rates adjusted for censoring. Int. J. Radiat. Oncol. Biol. Phys. 18:165-171; 1990. 25. Schultheiss, T. E.; Stephens. L. C.; Ang, K. K.: Price, R. E.; Peters, L. J. Volume effects in rhesus monkey spinal cord. Int. J. Radiat. Oncol. Biol. Phys. 29:67-72; 1994. 26. Sharma, R.; Warmelink, C.; Yudelev, M. S.; Mesina, C. F.; Maughan, R. L.; Forman, J. D. Description of a 3D conforma1 neutron and photon radiotherapy technique for prostate cancer. Med. Dosim. 30:45-53: 1995. 27. Simpson, J. R.; Purdy, J. A.; Manolis, J. M.; Pilepich, M. V.; Burman, C.; Forman, J.; Fuks, Z.; Cheng, E.; Chu, J.; Matthews, J.; Mohan, R.; Solin, L.; Tepper. J.; Urie, M. Three-dimensional treatment planning considerations for prostate cancer. Int. J. Radiat. Oncol. Biol. Phys. 21:243-252: 1991. 28. Stone, R. S. Neutron therapy and specific ionization. Am. J. Roentgenol. 59:771-785: 1940.
Oncology
Biol.
Phys., Vol. 35. No. 4. pp. 693-699, 1996 Copyright 0 1996 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/96 $15.00 + .OO
PII: SO360-3016(96)00136-S
ELSEVIER
0 Clinical Original Contribution HIP STIFFNESS FOLLOWING PHOTON RADIOTHERAPY:
MIXED CONFORMAL NEUTRON AND A DOSE-VOLUME RELATIONSHIP
PAUL J. CHUBA, PH.D., M.D., RENU SHARMA, B.S., C.M.D., MARK YUDELEV, MS., MARIE DUCLOS, M.D., FALAH SHAMSA, PH.D., SUSAN GIACALONE, R.N., COLIN G. ORTON, PH.D., RICHARD L. MAUGHAN, PH.D. AND JEFFREY D. FORMAN, M.D. Department of Radiation Oncology, Wayne State University, Gershenson Radiation Oncology Center, Detroit, MI, USA Purpose: To determine the relationship between dose, volume, and the incidence of hip stiffness in patients who received conformal neutron irradiation for nrostate cancer. Methods and Materials: A series of dose-searching studies using neutron irradiation for prostate cancer were performed to determine the optimal dose, fraction size, field size, technique, and proportions of photon and neutron dose. Neutron doses ranged fro& 9 to 20 Gy and photon doses ranged frbmb to 38 Gy: Data were analyzed by using a hip stiffness grading scale. Results: Hip stiffness was recorded on follow-up examination in 30% of patients (40 out of 132) treated with fast neutrons or mixtures of fast neutron and photon radiation for prostate cancer. Hip stiEness was categorized as none (Grade 0,92 patients), mild (Grade 1,24 patients), moderate (Grade 2,10 patients), or severe (Grade 3,6 patients). The incidence of hip stiffness differed significantly by dose and volume in the five dose levels studied @ < 0.001). Conclusions: By using a mixture of conformal neutron and photon irradiation and limiting the total neutron dose to less than 13 Gy, hip stiffness toxicity could be reduced to acceptable levels. Radiotherapy side effects, Fast-neutron radiation
INTRODUCTION To improve the therapeutic ratio for fast neutron irradiation for prostate cancer, it is necessary to increase local tumor eradication without an increase in normal tissue complications as compared to photon irradiation. Although prior experiences have confirmed an improvement in local control, the reported complication rates have been prohibitively high (15). The severity of these chronic complications depends on the total dose delivered, the dose fractionation, the interfraction interval, the volume, and the quality of the radiation, among other factors (6, 11, 13, 20). In general, the late effects of high linear energy transfer radiations such as fast neutrons may be greater than x-rays when similar acute toxicities are reached (2, 5, 8-10, 13, 14, 21, 28). No simple formula exists to predict normal tissue tolerances for fast neutron irradiation based on experience with photon radiation. For this reason dose-seeking studies for fast neutron radiotherapy must be conducted separately.
Randomized clinical trials have demonstrated the usefulness of fast neutron treatment and mixed fast neutron/ photon treatment for prostate cancer (15, 22). Reduced toxicity was shown to be related to the use of shaped neutron beams with improved depth dose characteristics. However, the use of shaped neutron fields in the Radiation Therapy Oncology Group experience still resulted in a 10% rate of severe (Grade 3 and 4) morbidity (22). More recent data indicate that conformal photon therapy is associated with decreased morbidity for prostate cancer (19, 23). To test the hypothesis that conformal neutron radiotherapy would result in a decreased rate of morbidity, three prospective dose-finding studies have been conducted at this institution. The goal of these studies was to help define the optimal dose and volume of conformal neutron irradiation for prostate cancer. The early results of these trials suggested that hip stiffness was a dose-limiting toxicity in pelvic neutron radiotherapy. In this report, combined neutron and photon doses leading to hip stiffness are analyzed. Clinical follow-up and normal tissue
Reprint requests to: Jeffrey D. Forman, M.D., Gershenson Radiation Oncology Center, 3990 John R., Detroit, MI 482012097. Acknowledgements-We thank Tongming He for expert assis-
tance with the GRATIStm 3D treatment-planning system, Prof. M. Joiner (UK) for stimulating discussions, and Gail Frasure C.T.R. and Chris Zuniga for data retrieval. Acceptedfor publication 1 March 1996. 693
I. J. Radiation Oncology
694
0 Biology
Volume 35. Number 3. 1996
0 Physics
Table I. Patient characteristics SNP-9 Neutron dose* Photon dose Composite dose’ Stage Mean age Median follow-up Number of patients Number treated to large field* African American Caucasian
9 GY 38 Gy 69.5 Gy A. B 66 10 months 51 0 16 35
SNP-IO IO Gy 38 Gy 73 Gy
A. B 66 4 months 27 0 3 24
and doses by group MNP-13
MNP- I5
N-20
13 Gy 25 Gy 70.5 Gy C, D 60 16 months 5 4 0 5
1s Gy 18 Gy 70.5 Gy C. D 65 16 months 43 36 17 26
70 Gy 0 Gy 70 Gy (1. D 63 5 months 6 5 0 6
* Doses quoted are the prescribed dose to the tumor volume. ’ Composite doses were calculated for this purpose by multiplying the neutron dose by RBE = 3.5 and adding the photon dose. t Larger pelvic fields were treated as described in Methods and Materials. SNP = sequential neutron photon, MNP = mixed neutron photon, N = neutron only.
dose information were used in an attempt to define the critical dose-volume relationships necessary to avoid hip stiffness.
METHODS
AND MATERIALS
Follow-up data from three prospective Phase II protocols in which fast neutron therapy or mixtures of photons and neutrons were used in the radiotherapy of prostate cancer patients at Harper Hospital were studied. Neutron and photon treatments were delivered with fully conforma1 beam shaping in all cases. All patients had completed radiotherapy at the time of analysis. Neutrons were produced from a superconducting cyclotron with a 48 MeV deuteron beam incident on an internal beryllium target (18). The source-to-axis distance was 182.9 cm, D,,, was 0.9 cm, and the percent depth dose at 10 cm was 63.9%. Photon fields were treated with 15 MV photons. Patients were simulated conventionally using polyurethane cradle immobilization, bladder contrast, and urethrogram. Treatment planning computed tomography scan immediately following simulation was used to provide axial image data for use in treatment planning system (26, 27). Conformal fields were designed to provide a minimum margin of 1.5 Table 2. Hip stiffness grading scale Grade
Severity
0
none
1
mild
2
moderate
3
severe
4
disabling
Description No hip stiffness or limitation of range of motion Mild hip stiffness or soreness or limitation of range of motion Interference with motion of hip joint, moderate pain Interference with normal daily activity, severe pain Severe limitation of ambulation or requiring surgery
cm from the gross tumor volume to the block edge. For each axial CT image, four anatomes were contoured for use in constructing dose-volume histograms; the femoral head (FH), the adductor muscle group (AM), the iliopsoas muscle group (IPM), and the gluteus medius and minimus together with the tensor fascia lata (GM). In one protocol 80 patients with early stage, low to intermediate grade prostate cancer (Tl -2NxM0, Gleason score < 8) were treated with sequential neutron followed by photon (SNP) irradiation between April 1993 and May 1994. Two dose levels were tested. Either 9.0 (5 1 patients) or 10.0 (29 patients) neutron Gy were delivered in 1.O Gy daily fractions to the prostate (boost volume) followed by 38 photon Gy in 2.0 Gy daily fractions to the prostate, seminal vesicles, and periprostatic lymph nodes. Photon treatments were delivered using a four-field box; however, for the neutron treatment, opposed lateral fields were paired with conformal nonaxial right and left anteriorinferior-superior oblique fields as previously described (16, 17, 19, 26). Forty-nine patients with locally advanced prostate cancer (T3-T4, NO-Nl, Gleasons score 5 7) were treated with concurrently mixed neutron photon radiation (MNP group) between October 1992 and April 1994. One patient was lost to follow-up. Two dose levels were used. Fortythree patients received a total of 15 neutron Gy to the prostate and seminal vesicles delivered in 1 Gy fractions 3 days per week (Mondays, Wednesdays, and Fridays) plus 18 photon Gy delivered in 1.8 Gy fractions 2 days per week (Tuesdays and Thursdays). Pelvic lymph nodes were treated similarly except that 0.6 Gy daily neutron fractions were used. Five patients received a total of 13 neutron Gy and 25.2 photon Gy using similar fractionation. The dose differential for treating pelvic lymph nodes was accomplished by using a partial transmission block (tungsten rods) (16, 26). Six patients with locally advanced prostate cancer were treated with neutrons alone (N group) between October 1992 and March 1994. These patients received 12.5 Gy
Hip stiffness following mixed conformal radiotherapy
0 P. J. CHUBA
695
et al.
Table 3. Incidenceof hip stiffnessby group SNP-9 Total patients Grade0 Grade 1 Grade2 Grade3 Grade > 0 Grade> 1
SNP-10
MNP-13
MNP-15
N-20
51
21
5
43
6
42(83%)
21(78%) 5 (19%)
3 (60%)
25 (58%) 7(16%) 6 (14%) 5 (12%) 18(42%)
1(17%) 3(50%) 1(17%)
9 (18%) 0 (0%) 0 (0%) 9(18%)
0 (0%)
0 (0%)
1 (4%) 0 (0%)
0 (0%)
6 (22%) 0(4%)
2(40%) 2(40%)
2 (40%)
11 (26%)
1 (17%) 5(83%) 2(330/o)
SNP = sequentialneutronphoton. MNP = mixed neutronphoton,N = neutrononly
to the prostate, seminal vesicles, and pelvic lymph nodes in 1.25 Gy daily fractions followed by a nonaxial boost to the prostate and seminal vesicles of 7.5 Gy in 1.25 Gy daily fractions. Characteristics of patients making up the five dose levels investigated are shown in Table 1. The median followup for the MNP protocol was 16 months (range: 4 to 24 months), for the SNP protocol 10 months (range: 3 to 16 months), and for the neutron only patients 5 months (range: 3 to 24 months). As the neutron dose increased the photon dose decreased so that the composite neutron plus photon dose (to tumor) for all patients was estimated at 69.5 to 73 Gy assuminga relative biologic effectiveness of 3.5. All patients in the two SNP groups were treated with small fields, but only nine patients in the two MNP groups and N-20 groups combined had small volumes treated (Table 1). A scale for grading hip stiffness was developed (Table 2). Mild, moderate, severe, and disabling hip stiffness was
assigned a grade of 1 to 4 according to the scale. We applied chi-square tests to assessassociationsamong the relative incidence of graded complications between groups. Fisher’s exact test was used for evaluating associations when the cell sizes in the contingency tables was small. The Mantel-Haenszel method was used to adjust for confounding effects.
RESULTS Hip stiffness ( Grade 0; Table 3) was observed in 9 out of 51 (18%) patients who received 9 neutron Gy (SNP9) 6 out of 27 (22%) who received 10 neutron Gy (SNPlo), 2 out of 5 (40%) who received 13 neutron Gy (MNP13), 18 out of 43 (41%) who received 15 neutron Gy (MNP- 15) and 5 out of 6 (83%) who received 20 neutron Gy (N-20). A bar graph illustrating the relative incidence of any hip stiffness (> Grade 0) or severe (Grade 3) hip stiffness in the five groups is shown in Fig. 1. The differ-
100% 90% 80%
42%
22% 18%
r-l
SNP-9
SNP-10
MNP-13
MNP-15
N-20
Fig. 1. Relative incidenceof hip stiffnessby patient group. The incidenceof any hip stiffness(> Grade0) in the five patientgroupsconsideredis illustratedwith an openbar. Incidenceof severe(Grade3) hip stiffnessisillustrated with shadedbar.
696
I. J. Radiation Oncology 0 Biology 0 Physics
ence in the incidence of complications between these five dose-level groups was statistically significant (p < 0.01). Moderate to severe hip stiffness was also related to the proportion of neutron dose (Table 3, Figure l), and was statistically significant @ < 0.01). One of 80 patients (I .2%) receiving less than 13 neutron Gy complained of Grade 2 hip stiffness, however Grade 2 or Grade 3 hip stiffness was recorded for 2 out of 5 (40%) patients who received 13 neutron Gy, in 11 out of 43 (26%) patients who received 15 neutron Gy, and in 2 out of 6 (33%) patients who received 20 neutron Gy. As shown in Table 4, 23 out of 45 (51%) of all patients who were treated to pelvic lymph node regions but only 17 out of 87 (20%) of all patients treated to small fields complained of hip stiffness. The effect of volume was statistically significant whether the entire series @ < 0.001, five dose levels) or only the three dose levels MNP-13, MNP-15, and N-20 were considered (JJ < 0.001, patients receiving at least 13 neutron Gy). Results were compared in three subcategories: low neutron dose/low volume (78 SNP patients), high neutron dose/low volume (9 patients from MNP-13, MNP- 15, and N-20 dose levels), and high neutron dose/high volume (45 patients from MNP-13, MNP-15, and N-20 dose levels). In the SNP-9 and SNP10 groups combined, all hip stiffness was Grade 1, with a single Grade 2 complication for a total of 14 out of 78 (19%) patients. Of the nine patients who received 13 Gy or more of neutron treatment and small fields, two out of nine (22%) had hip stiffness (one Grade 1 and one Grade
Table 4. Incidence
of hip stiffness in patients treated with large* or small fields
Low neutron dose/low volume SNP-9 Grade 1 Grade 2 Grade 3 Total
9 0 0 9/51
(18%) (0%) (0%) (18%)
High neutron dose/low volume MNP-13 Grade I Grade 2 Grade 3 Total
0 0 0 O/l
(0%) (0%) (0%) (0%)
High neutron dose/high volume MNP-13 Grade 1 Grade 2 Grade 3 Total
0 2 0 2/4
(0%) (50%) (0%) (50%)
SNP-10 5 1 0 6/27
(19%) (4%) (0%) (22%)
MNP-15 1 (14%) 0 (0%) 0 (0%) l/7 (14%)
MNP-15 6 6 5 17/36
(17%) (17%) (14%) (47%)
N-20 0 (0%) 0 (0%) 1 (0%) l/l (100%)
N-20 3 1 0 4/5
(60%) (20%) (0%) (80%)
* Large Fields were treated as described under Methods and Materials. SNP = sequential neutron photon, MNP= mixed neutron photon, N = neutron only.
Volume 35. Number 4. 1996
3). Of 45 patients who received 13 neutron Gy or more and larger pelvic field treatment, 23 (51%) exhibited hip stiffness. In the latter group nine Grade 2 and five Grade 3 complications were recorded. When the relationship of dose level to hip stiffness (Table 3) was adjusted for the (confounding) effect of volume (Table 4) or vice-versa. the p-value was reduced (p = 0.07) to borderline levels. implying that the dose and volume effects were strongly associatedwith each other as well as with complications. Two axial CT images with femoral heads (FH), and three extrapelvic muscle groups illustrated are shown in Figure 2. The possible anatomic significance of the volumes of these normal tissues irradiated was investigated by determining the dose distributions to these structures by constructing dose-volume histograms. Dose-volume histograms (DVHs) for these four anatomes using the two different field sizes used in the MNP and N-20 groups are shown in Figure 3. The cumulative DVHs to the femoral heads, the adductor, and the iliopsoas muscle groups were quite similar whether or not large pelvic fields were included in the planning. For the gluteus muscle group however, the DVH curves differed markedly according to field size. Small field treatment resulted in significant sparing of the gluteus muscle group (Fig. 3).
DISCUSSION Connective tissue is vulnerable to radiation damage. The tolerance doses(TD5/5 and TD50/5) for femoral head necrosis for photon radiotherapy have been establishedat 52 and 65 Gy, respectively (7). Tolerance dosesto muscle are not well characterized, although skeletal muscle makes up a large proportion of body massand is extremely well vascularized. On a cellular level, it is thought that microvasculature injury is reflected secondarily in late effects to the connective tissue stroma. Alternatively, direct radiation damage to parenchymal cells may lead to formation of granulation tissue. Hip stiffness in pelvic radiotherapy has been reported previously (4), and has been assumedto be related to fibrosis of the soft tissue adjacent to the joint. In a series reported by Duncan ef al., 18 patients had received 16 neutron Gy in 20 fractions over 4 weeks (4). Their machine was low in energy and had a relatively high relative biologic effectiveness. Five patients had severe hip stiffness. By comparison to the 132 patients considered here. 40 (30%) described hip stiffness, 6 were severe. The patients treated with greater dosesof neutron radiation exhibited a significantly greater incidence and severity of hip stiffness (Table 1). Of the six patients with severe hip stiffness described here, two had radiologic findings in the femoral head. It, therefore, appeared that severe hip stiffness was partly related to a direct effect on bone. For less severe hip stiffness, no osteoradionecrosis was evident and an effect on muscle was investigated. Using DVHs. it was shown that
Hip stiffness following mixed conformal radiotherapy 0 P. J. CHUBA
et al.
697
a
b
Fig. 2. Location of the femoral head, gluteus, adductor, and iliopsoas anatomes. Axial computed tomograms at the level of the femoral head (a) or the pubic symphysis (b) are shown. The GM muscle group (gluteus medius, gluteus minimus, and tensor fascia lata; most lateral), the AM muscle group (adductor magnus, adductor longus, adductor brevis; most medial), and the IPM muscle group (iliopsoas associated; most anterior), are outlined. Anteroposterior and right lateral beam edges are indicated for reference.
the reduced incidence of hip stiffness in patients treated with smaller fields could be attributed to sparing of the GM muscle group (Fig. 3). This group is located laterally, in the region of relative dose maxima for the neutron treatments. Because the patient groups differed in field arrangements, field sizes, and in the sequencing of photon and
neutron treatments, as well as in the proportion of the total dose delivered with photon or neutron treatments, any of these factors could have been responsible for the differences in hip stiffness incidence. Data presented here show that both the total neutron dose (or proportion of the dose delivered with neutrons) and the volume of normal tissue irradiated were significant factors. The incidence of hip
698
I. J. Radiation Oncology 0 Biology 0 Physics
0
20
40
60
80
Volume 3.5,Number 4. 1996
100
0
20
40
60
100
80
80 60
b.
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
20
40
60
80
100
100 80 60
d.
0
Patient 1
Patient 2
Fig. 3. Cumulative dose-volume histograms for (a) femoral head (FH), (b) gluteus muscle (GM), (c) adductor muscle (AM), and (d) iliopsoas muscle (IPM) anatomes in two patients from the Mm-15 group. Plots of percent volume irradiated vs. percent dose are shown for four anatomes in two individual patients. Dose-volume histograms using large fields are shown with solid lines, small fields with dashed lines.
stiffness was greater when the proportion of dose delivered with neutrons was greater (Table 3) or when larger pelvic fields were used (Table 4). Although the effect of volume could not be completely separated from the effect of dose level, patients who received high neutron dosehigh volume treatment had a higher incidence (5 1%) and much greater severity of complications (Table 4). When the dose effects were adjusted for the effects of volume, they remained statistically significant. The composite doses (Table 1) for mixed neutron photon treatments calculated by using an RBE of 3.5 are not useful for predicting normal tissue effects. An estimate of the actual composite bioeffective dose (20) delivered to the hip joint will require knowledge of the dose per fraction at depth and the actual relative biologic effectiveness.
For mixed neutron/photon treatments, the influence of the increase in the neutron portion of the dose on the composite dose is greater with higher RBE. Several groups have reported methods for modeling dose-effect relationships for complications (1, 3, 11, 12,24, 25) and by using a composite bioeffect dose for mixed treatments, the dose-response for hip stiffness can be fitted to a sigmoid curve using a logistic function. Such an analysis can be used to predict an RBE for hip stiffness and other-sideeffects for the fast neutron therapy (M. Yudelev, oral communication, March 1995). In conclusion, hip stiffness was found to be a doselimiting toxicity for conformal neutron irradiation ofprostate cancer. As we make progress in defining the clinical, biochemical, and histologic local control in these patients,
Hip
stiffness following mixed conformal radiotherapy 0 P. J. CHUBA
the effect of these dose limitations on improving the therapeutic ratio will become evident. By using conformal mixed neutron/photon doses of less than 13 neutron Gy, the hip stiffness incidence was low, and severe hip stiff-
et ul.
699
ness was avoided. A volume effect was demonstrated with significantly reduced toxicity with sparing of the GM muscle group. Severe hip stiffness appeared to be related to a direct effect on bone.
REFERENCES 1. Battermann, J. J.; Hart, G. A. M.; Breur, K. Dose-effect relations for tumour control and complication rate after fast neutron therapy for pelvic tumours. Br. J. Radiol. 54:899904; 1981. 2. Catterall, M.; Errington, R. D.; Bewley, D. K. A comparison of clinical and laboratory data on neutron therapy for locally advanced tumors. Int. J. Radiat. Oncol. Biol. Phys. 13:17831791; 1987. 3. Cohen, L.; Schultheiss, T. E.; Hendrickson, F. R.; Mansell, J.; Saroja. K. R.; Lennox, A. Normal tissue reactions and complications following high-energy neutron beam therapy: I. Crude response rates. Int. J. Radiat. Oncol. Biol. Phys. 1673-78; 1989. 4. Duncan, W.; Arnott, S. J.; Jack, W. J. L.; Orr, J. A.; Kerr, G. R.; Williams, J. R. Results of two randomised trials of neutron therapy in rectal adenocarcinoma. Radiother. Oncol. 8: 191-198; 1987. 5. Duncan, W.; Williams, J. R.; Kerr, G. R.; Amott, S. J.; Quilty, P. M.; Rodger, A.; Macdougall, R. H.; Jack, W. J. L. An analysis of the radiation related morbidity observed in a randomized trial of neutron therapy for bladder cancer. Int. J. Radiat. Oncol. Biol. Phys. 122085-2092; 1986. 6. Ellis, F.; Weatherbum, H. RBE and clinical response in radiotherapy with neutron beams. Br. J. Radiol. 57:817; 1984. 7. Emami, B.; Lyman, J.; Brown, A.; Coia, L.; Goitein, M.; Munzenrider. J. E.; Shank, B.; Solin, L. J.; Wesson, M. Tolerance of normal tissue to therapeutic radiation. Int. J. Radiat. Oncol. Biol. Phys. 21:109-122; 1991. 8. Errington, R. D.; Ashby, D.; Gore, S. M.; Abrams, K. R.: Myint. S.; Bonnett, D. E.; Blake, S. W.; Saxton, T. E. High energy neutron treatment for pelvic cancers: Study stopped because of increased morbidity. Br. Med. J. 302: 1045-105 1; 1991. 9. Griffin, T. W.: Martz, K. L.; Laramore, G. E.; Thomas, F. J.; Maor. M. H.; Hendrickson, F. R.; Parker, R. G.; Richter, M. P.; Davis. L. W. High energy (42-66 MeV reactions) fast neutron dose optimization studies in the head and neck. thorax, upper abdomen, pelvis, and extremities. Radiother. Oncol. 19:307-316; 1990. 10. Griffin, T.; Pajak, T.; Laramore, G.; Davis, L. Analysis of neutron radiotherapy treatment complications. Bull. Cancer (Paris) 73:582-586; 1986. 11. Johns. H.; Joiner, M. C. A single method for fitting curves to dose-effect data for functional damage. Int. J. Radiat. Biol. 60:533-541: 1991. 12. Joiner, M. C.; Denekamp, J.: Maughan, R. L. The use of ‘top-up’ experiments to investigate the effect of very small doses per fraction in mouse skin. Int. J. Radiat. Biol. 49:565580: 1986. 13. Laramore, G. E.; Austin Seymour, M. M. Fast neutron radiotherapy in relation to the radiation sensitivity of human organ systems. Adv. Biol. 15:153-192; 1992. 14. Laramore, G. E.: Griffeth. J. T.; Boespflug, M.; Pelton, J. G.: Griffin, T. W.: Griffin, B. R.; Russell, K. J.; Kroh, W. Fast neutron radiotherapy results for sarcomas of soft tissue. bone, and cartilage. Am. J. Clin. Oncol. 12:320326; 1989.
15. Laramore, G. E.; Krall, G. E.; Thomas, F. J.; Griffin. T. W.; Maor, M. H.; Hendrickson, F. R. Fast neutron radiotherapy for locally advanced prostate cancer: Results of an RTOG randomized study. Int. J. Radiat. Oncol. Biol. Phys. 11:1621-1627; 1985. 16. Leibel, S. A.; Kutcher, G. J.; Mohan, R.; Harrison, L. B.: Armstrong, J. G.: Zelefsky, M. J.; Losasso, T. J.; But-man, C. M.; Mageras, G. S.; Chui, C. S.: Brewster, L. J.: Masterson, M. E.; Lo, Y. C.; Ling, C. C.; Fuks, F. Three-dimensional conformal radiation therapy at the Memorial Sloan Kettering Cancer Center. Semin. Radiat. Oncol. X74-289; 1992. 17. Marsh, L. H.: Ten Haken, R. K.; Sandler, H. M. A customized nonaxial external beam technique for treatment of prostate carcinoma. Cancer 45:724-727; 1992. 18. Maughan, R. L.; Powers, W. E.; Blosser, H. G. A superconducting cyclotron for neutron radiotherapy. Med. Phys. 21:779-785; 1994. 19. Mesina, C. F.; Sharma, R.; Rissman, L. S.; Geering, L.; He, T.; Forman, J. D. Comparison of a conformal nonaxial boost with a four-field boost technique in the treatment of adenocarcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 30:427430; 1994. 20. Orton, C. G. Recent developments in time-dose modelling. Aust. Phys. Eng. Sci. Med. 14:57-&t; 1991. 21. Parker, R. G.; Bumison, C. M.; Sadeghi, A. Normal-tissue tolerance to fast-neutron teletherapy. Front. Radiat. Ther. Oncol. 23:185-193; 1989. 22. Russell, K. J.; Laramore, G. E.; Krall, J. M.; Thomas, F. J.; Maor. M. H.; Hendrickson, F. R.; Krieger, J. N.; Griffin, T. W. Eight years experience with neutron radiotherapy in the treatment of stages C and D prostate cancer: Updated results of the RTOG 7704 randomized clinical trial. Prostate 1 I :183-193: 1987. 23. Sandler, H. M.; Perez-Tamayo, C.; Ten Haken, R. K.; Lichter, A. S. Dose escalation for stage C (T3) prostate cancer: Minimal rectal toxicity observed using conformal therapy. Radiother. Oncol. 23:53-54; 1991. 24. Schultheiss, T. E.; Cohen, L.: Mansell, J. Normal tissue reactions and complications following high-energy neutron beam therapy II: Complication rates adjusted for censoring. Int. J. Radiat. Oncol. Biol. Phys. 18:165-171; 1990. 25. Schultheiss, T. E.; Stephens. L. C.; Ang, K. K.: Price, R. E.; Peters, L. J. Volume effects in rhesus monkey spinal cord. Int. J. Radiat. Oncol. Biol. Phys. 29:67-72; 1994. 26. Sharma, R.; Warmelink, C.; Yudelev, M. S.; Mesina, C. F.; Maughan, R. L.; Forman, J. D. Description of a 3D conforma1 neutron and photon radiotherapy technique for prostate cancer. Med. Dosim. 30:45-53: 1995. 27. Simpson, J. R.; Purdy, J. A.; Manolis, J. M.; Pilepich, M. V.; Burman, C.; Forman, J.; Fuks, Z.; Cheng, E.; Chu, J.; Matthews, J.; Mohan, R.; Solin, L.; Tepper. J.; Urie, M. Three-dimensional treatment planning considerations for prostate cancer. Int. J. Radiat. Oncol. Biol. Phys. 21:243-252: 1991. 28. Stone, R. S. Neutron therapy and specific ionization. Am. J. Roentgenol. 59:771-785: 1940.