Hyperfractionated conformal radiotherapy in locally advanced prostate cancer: Results of a dose escalation study

ELSEVIER

l Phase I/II Clinical Trials

Int. J. Radiation Oncology Biol. Phys.. Vol. 34, No. 3. pp. 655-662, 1996 Copyright 0 1996 Else&r Science Inc. Printed in the USA. All rights reserved

OXtO-3016/96 $15.00 + .I0

0360~3016( 95)02202-3

HYPERFRACTIONATED CONFORMAL RADIOTHERAPY IN LOCALLY ADVANCED PROSTATE CANCER: RESULTS OF A DOSE ESCALATION STUDY

JEFFREY D. FOFWAN, M.D., MARIE DUCLOS, M.D., FALAH SHAMSA, PH.D., ARTHUR T. PORTER, M.D. AND COLIN ORTON, PH.D.

Department of Radiation Oncology, Wayne State University, Detroit, MI

Purpose: This study was initiated to assess the incidence of chronic complications and histoiogic and biochemical control following hyperfractionated conformal radiotherapy in patients with locally advanced prostate cancer. Methods and Materials: Between October 1991 and October 1994,49 patients with locaily advanced prostate cancer were entered on the first two dose levels of a prospective dose-escalation study using hyperfiaction- ated three dimensional conformai radiotherapy. The first 25 patients received a minimum tumor dose of 78 Gy to the prostate and semhmi vesicles in 6 weeks at 1.3 Gy, b.i.d. No increase in chronic toxicity compared with conventional radiotherapy was noted; therefore, an additional 24 patients were treated to a minimum tumor dose of 82.8 Gy to the prostate and seminal vesicles in 7 weeks at 1.15 Gy, b.i.d. Toxicity was scored according to the Radiation Therapy Oncology Group morbidity grading scale. Eflicacy was assesmd through scheduled postradiation prostate speciiic antigen values and ultrasound-guided biopsies. The median follow-up for the entire group was 20 months. Results: The hyperfractionated external radiation was well tolerated with minhnai acute morbidity. At 30 months, the actuarial probability of Grade 2 gastrohttestinal toxicity was 17 % . At 30 months, the actuarial probability of Grade 2 genitourhmry toxicity was 16%. There was no statisticaiiy significant difference between the two dose levels. No Grade 3 or 4 gastrointestinal or genitourinary toxicity was noted. At 12 months, 84% of patients had a prostate speciiic antigen 5 4; and 53%; s 1 rig/ml. At 12 months, 71% of patients had post radiation biopsies that were either negative (55% ) or showed a marked therapeutic effect (16%). Conclusion: The use of hyperfractionated conformai radiotherapy facilitated dose escalation with no in- crease in chronic toxicity compared to standard doses. The initial tumor response based on prostate specific antigen measurements and postradiation biopsies is highly encouraging. Based on these rest&s, an increase in dose to 87.4 Gy has been planned according to the schema of thii ongoing dose escalation study.

Hyperfractionation, Locally advanced prostate cancer, Dose escalation.

INTRODUCTION

The incidence of residual or recurrent disease following

Recent studies have indicated that up to 50% of pa-

external beam radiation therapy for locally advanced

tients, without clinical evidence of recurrence, have re-

prostate cancer has been shown to be higher than pre- viously reported (7, 8, 13, 14, 23). Following radiation, evidence of local failure has been documented clinically

sidual disease identified by subsequent prostate biopsy

in up to 50% of patients by digital rectal examination, without the additional information available through his- tologic evaluation or serum prostate specific antigen lev- els (20).

(7). In the past, postradiation biopsies had been re- stricted to patients undergoing transurethral resection for

Several reports in the literature suggest that the total

progressive urinary outlet obstruction, or obtained in se-

radiation dose may have an impact on the local eradica-

lected patients, such as those with clinical evidence of failure. However, because approximately 50% of pa- tients with locally advanced disease have clinical evi-

tion of prostate cancer. Doses in excess of 70 Gy appear

dence of failure, and half of those with clinical local control have histologic evidence of residual disease, only 20%-25% of these patients may achieve local control following conventional treatment with standard dose ra- diotherapy alone.

Reprint requests to: Jeffrey D. Forman, M. D., Department of Radiation Oncology, Wayne State University, 3990 John R,

Detroit, MI 48201. E-mail: [email protected]

655

656 I. J. Radiation Oncology l Biology 0 Physics

to improve the local control rates (9). This has been seen with external beam doses in excess of 70 Gy, as well as when external beam irradiation has been supplemented by temporary interstitial implants ( 15). However, the proximity of critical normal structures, such as the bladder and rectum, has limited the total doses used. Even at standard dose levels (I 70 Gy), serious late sequelae have been reported in up to 5% of patients at 5 years when treated with conformal irradiation ( 17).

Preliminary data for photon dose escalation using stan- dard fractionation ( 1.8-2.0 Gy), and conformally de- signed fields, have led to mixed results. Two groups have been able to safely exceed the 75Gy dose level ( 12, 17, 18). However, Schultheiss et al. (19), despite using conformal irradiation techniques, found a 14% incidence of severe chronic toxicity at 78 Gy, precluding further attempts at dose escalation.

The need for improved local control combined with evidence that conformal irradiation alone may not allow for safe dose escalation, supplied the impetus to explore hyperfractionation as a method to improve the therapeutic ratio (5, 22). There was prior evidence that a hyperfrac- tionated schedule in bladder cancer could allow for safe dose escalation (3). Calculations by Fowler et al. (5) also suggest that changing from 2-Gy fractions daily to two fractions of 1.2 Gy per day, would almost always be worthwhile for locally advanced prostate cancer. They estimated an increase in local control in excess of 20%. In this report, we present interim results on a study de- signed to test the feasibility of dose escalation using hy- per-fractionated three dimensional conformal radiotherapy in locally advanced carcinoma of the prostate.

METHODS AND MATERIALS

Between October 1991 and October 1994,49 patients with locally advanced adenocarcinoma of the prostate were treated in the Wayne.State University Department of Radiation Oncology with a hyperfractionated, three dimensional conformal radiotherapy technique. All pa- tients were entered on this Institutional Review Board (IRB) approved phase I/II study. Informed consent was obtained from all patients prior to treatment. Eligibility included patients with a large primary tumor (T3, T4), and/or a poorly differentiated histology (Gleason score > 8 ) . Eighty-two percent of patients were clinical Stage T3 and 51% had a Gleason score of 8, with a mean pretreatment prostate specific antigen (PSA) level of 47.5 rig/ml and a median pretreatment PSA of 26.8 rig/ml. The clinical characteristics of these patients are summarized in Table 1. The mean age of the patients was 66.4 years (range 43-83 years).

Pretreatment evaluation included a complete history and physical examination, complete blood count, bio- chemical screening profile, serum prostatic acid phospha- tase (PAP), and PSA. Radiographic evaluation included

Volume 34, Number 3, 1996

Table 1. Clinical characteristics

Race: Caucasian 29 (60%)

Stage:

Gleason score:

Pretreatment PSA:

Pretreatment hormones:

African-American TlC T2 A/B T2C T3 6 7 8 9-10 <4 4-10

10-20 20-50 >50 Yes No

20 (40%) 3 (6%) 3 (6%) 3 (6%)

40 (82%) 6 (12%)

13 (27%) 25 (51%)

5 (10%) 2 (4%) 6 (12%)

11 (22%) 20 (40%) 10 (20%) 28 (57%) 21 (43%)

a transrectal prostatic ultrasound, pelvic and abdominal computerized tomography (CT), radionuclide bone scan, and plain radiographs to confirm bone scan abnormalities. Pretreatment gastrointestinal and genitourinary symptoms were recorded for all patients.

The use of preirradiation (neoadjuvant) hormonal ther- apy was not dictated by the protocol. Therefore, hormonal therapy was not uniformly delivered or randomly allo- cated. In total, 21 patients did not receive neoadjuvant hormones and 28 patients did. However, at the 78-Gy dose level, only 8 of 25 patients received hormones com- pared with 20 of 24 at the 82.8-Gy dose level. This was due to the fact that evidence of a benefit from the use of neoadjuvant therapy became available only after accrual at the 78-Gy level was completed ( 14). All subsequent patients were offered neoadjuvant as part of their treat- ment. Thus, comparisons of efficacy between the two dose levels must be viewed cautiously.

Neoadjuvant therapy consisted of 4 months of treat- ment with an LH-RI-I agonist. Some patients also received an oral antiandrogen during the first month. Simultaion and CT scanning was performed at the time of the second LH-RI-I injections. Radiation treatment began at the time of the third injection. No patient received planned adju- vant therapy following the completion of radiation.

Prior to treatment, patients underwent conventional x- ray simulation in the supine position. Patients were immo- bilized in a custom-made alpha cradle cast. The simula- tion was accomplished with oral, intravesicle, and in- traurethraI contrast according to a previously reported technique (4). At the time of the simulation, the isocenter was placed approximately in the geographic center of the prostate. Provisional simulation films were taken for the orthogonal [anterior-posterior ( AP) , right lateral (RL)] , and oblique angles [right (RSO) and left superior obliques (LSO) and right (RAISO) and left anterior infe- rior to superior oblique (LAISO)]. For the RSO and LSO fields, the gantry was horizontal at 90” and 270”,

Hyperfractionated conformal radiotherapy l J. D. FORMAN er a/. 651

respectively; with a 30” ( + 5”) pedestal rotation (patient head toward gantry head). For the RAISO and LAISO fields, the gantry was at t 50” with a 30” pedestal rotation (patient foot to gantry head).

The patients then underwent CT scanning in the simu- lated position using the alpha cradle and localization marks. Consecutive CT images with a 5-mm slice thick- ness and table feed were obtained from 2-cm above the vertex of the bladder to the ischial tuberosites. Additional CT images (6-8 cm) were obtained above and below these scans with a slice thickness of 1 cm.

The outline of the prostate, seminal vesicles, pelvic lymph nodes, pelvic bones, femoral heads, urethra, blad- der, rectum, and skin were digitized into the three dimen- sional planning system. Beam apertures were then de- signed. At the first dose level, 78 Gy, the patients received four field box irradiation (AP/PA/RL/LL) throughout the treatment (4). The pelvic lymph nodes received 45 Gy in 25 fractions and the prostate-seminal vesicles (PSV) received 78 Gy in 60 fractions of 1.3 Gy each. This was accomplished with a partial transmission filter over the prostate and seminal vesicles in the morning treatment such that the pelvic lymph nodes received 1.8 Gy, com- pared to 1.3 Gy for the PSV. In the afternoon, the pelvic lymph nodes were shielded. After 5 weeks, the PSV re- ceived 1.3 Gy b.i.d. for a total dose of 78 Gy in 6 weeks. The minimum margin from the gross tumor volume (GTV) to the block edge was 1.5 cm. The target dose was prescribed to the maximum isodose that completely encompassed the GTV. Thus, the minimum tumor dose to the GTV was 78 Gy. The maximum or central axis dose was up to 4% higher (e.g., 81 Gy). The volumes were displayed with the beam’s eye view (BEV) tech- nique. The BEVs were registered to the simulation films by aligning the bony pelvic anatomy, and the location of the target volumes, normal structures, and treatment portal were transposed onto the simulator films. These were the templates for cerrobend block fabrication and for port- film verification.

At the second dose level, 82.8 Gy, elective pelvic lymph node irradiaton was eliminated. This decision was made because of concerns relating to enhanced toxicity at this dose level as well as to make the experience more analogous to other ongoing dose-escalation programs that did not include pelvic irradiation ( 12). In addition, to limit the dose to the rectum, bladder, and femoral heads, multiple oblique noncoplanar fields were used. The pro- cess of simulation, treatment planning, and contouring was unchanged. The twice daily treatment volume con- sisted of the prostate and seminal vesicles, and was the same in the morning and afternoon. The morning treat- ment of 1.15 Gy was delivered through AP/PA/RL/LL fields (Fig. 1). The afternoon treatment consisted of RSO/LSO/RAISO/LAISO fields (Fig. 2). The block margins were nonuniform for this eight-field noncoplanar treatment, with a minimum margin of 1.2 cm from the

(4

(b)

Fig. 1. (a) Sagital, CT-reconstructed, view of conformal four- field beam arrangement. (b) Axial CT view at the level of the prostate of conformal four-field plan. Anterior, posterior, and right and left lateral.

gross tumor volume to the block edge. For both dose levels, the two daily fractions were separated by 6 to 8 h. The change in fraction size from 1.3 to 1.15 Gy was due to the possible confounding effects of shortening overall treatment time on either local control or toxicity. At 1.15 Gy per fraction, the treatment required 36 days, which is identical to that of our standard fractionation schedules.

658 I. J. Radiation Oncology l Biology l Physics Volume 34, Number 3, 1996

lb)

Fig. 2. (a) Coronal, computed tomography (CT) reconstructed view, of conformal noncoplanar beam arrangment: right (RAISO) and left anterior inferior to superior oblique (LAISO), and right (RSO) and left superior oblique (LSO). Note that no beams exit into each other. (b) Axial CT view at the level of the prostate showing the intersection of the four noncoplanar beams. (c) Axial CT view at the level of the prostate showing the composite isodose distribution for all eight beams: anterior, posterior, right and left lateral, RAISO, LAISO, RSO, and LSO.

Hyperfractionated confomal radiotherapy 0 J. D. FORMAN ef a/. 659

The minimum dose to the GTV was 82.8 Gy with a maximum dose within the GTV of up to 86 Gy.

Patients were evaluated at least weekly during treat- ment. The first follow-up visit was 1 month following treatment and then at months 3, 6, 9, 12, 16, 20, and 24. Patients were seen twice yearly thereafter. At each visit, there was a physical exam including digital rectal exami- nation, complete blood count (CBC ) , blood chemistries, serum PAP, and PSA levels. Posttreatment biopsies were obtained 12 months after treatment. Toxicity was graded according to the Radiation Therapy Oncology Group (RTOG) morbidity grading scale. Chronic complications were defined as those developing more than 3 months after the completion of irradiation. Times to last follow- up, to local recurrence, or to developing a radiation com- plication were calculated from the date of last treatment. Life-table probabilities of the above observations were calculated by the methods of Kaplan and Meier ( 10).

RESULTS

Acute toxicity All 49 patients completed the planned course of treat-

ment without interruption of the treatment course. The average duration of treatment was 41.7 days (78 Gy ) and 50.2 days (82.8 Gy) for the two dose levels, respectively. Although the increase in dose was expected to result in an increase in the acute treatment effects, we observed no significant worsening of treatment-related reactions compared with standard fractionation. Medications to re- lieve gastrointestinal (GI) and genitourinary (GU) symp- toms were freely prescribed however, no patient experi- enced higher than Grade 2 acute GI or GU toxicity.

Late complications The median follow-up time for the 78-Gy dose level

was 30 months (range 9-36) and 12 months (range 9- 20) for the 82.8-Gy level. No patient has been lost to follow-up. No Grade 3 or 4 complications have been seen. Six patients ( 12%) had seven Grade 2 complications. Two Grade 2 GU complications were observed at 16 and 30 months. Five Grade 2 GI complications were noted from 9 to 30 months after treatment. Five of the six com- plications have been seen in the 78 Gy dose group. At 30 months, the actuarial risk of Grade 2 GI and GU mor- bidity were 17% and 16%, respectively.

EfJicacy Tumor status was assessed at each follow-up visit by

digital rectal exam, serum PSA, and PAP levels. Twelve months following the completion of treatment, transrectal ultrasound-guided biopsy was performed and six 18- gauge cores were obtained from the dominant hypoechoic or color doppler regions of suspicion in each sextent. All postradiation biopsies were reviewed by a single patholo- gist. Twenty-two of the first 31 1Zmonth biopsies were

either negative (55%) or showed marked therapeutic ef- fect ( 16%). Marked therapeutic effect indicated that any residual tumor consisted of isolated glands or single cells having abundant vacuolated cytoplasm. Nuclei were small and pyknotic or in some cases open, prominent nucleoli.

Twelve of 20 patients at the 78-Gy dose level (60% ) and 10 of 11 patients at the 82.8 Gy dose level (91%) have had negative or marked effect biopsies. Of the 17 patients who had biopsies but did not receive neoadjuvant hormonal therapy, 47% were either negative (6 patients) or showed a marked therapeutic response (2 patients). Of the 14 patients who had biopsies and received neoadju- vant therapy, 11 were negative (79%) and 3 (21%) showed a marked therapeutic response. These results are summarized in Table 2.

With a median follow-up for all patients of 20 months, 59% remain clinically and biochemically free of recur- rence. Six patients ( 12%) have a rising PSA or biochemi- cal evidence of failure, four patients (8%) have local progression, and eight patients ( 16%) have developed distant metastasis. In total, 84% of patients achieved a PSA nadir level 5 4 rig/ml; and 53% achieved a PSA nadir level 5 1 rig/ml. Of patients with a minimum fol- low-up of 12 months, 45% have a PSA level 5 1 rig/ml at the time of last follow-up. Although the patients who received neoadjuvant hormonal therapy reached nadir PSA values of 1 and 4 more often, this was not true at the time of last follow-up (Table 3 ) .

DISCUSSION

The use of radiation dose escalation in the management of locally advanced prostate cancer remains investiga- tional with several questions remaining to be answered ( 1) . These include whether dose escalation is needed (6, 11, 20, 23), whether dose escalation is possible ( 12, 17, 18), and whether dose escalation could result in a net improvement in the rates of complication-free survival (2 1) . In addition, if radiation dose escalation is found to be possible, should it then be applied to patients with early stage disease in whom an improvement in local control might be expected to more directly translate into improved survival outcomes?

Beginning with whether radiation dose escalation is needed in locally advanced prostate cancer, published data suggest that the majority of patients treated with standard doses (65-70 Gy ) of photon irradiation have histologic or biochemical evidence of residual disease following radiotherapy (8, 14,24). Based on digital rectal examination alone, 25%-50% of patients with locally advanced prostate cancer have evidence of local progres- sion ( 14, 20, 23). In addition, as methods for detecting residual local disease have improved with the use of transrectal ultrasound (with or without biopsies), as well as postradiation serum PSA determinations, it is clear that many more patients have evidence of residual disease

660 I. J. Radiation Oncology 0 Biology 0 Physics Volume 34, Number 3, 1996

Table 2. Biopsy results as a function of dose and the use of neoadjuvant hormonal therapy

78 Gy 83 Gy

No hormones Hormones No hormones Hormones Total

Negative Marked effect Positive Not done

6/15 (40%) 415 (80%) l/2 (osO%)

719 (78%) 17/31 (55%) l/15 (7%) l/5 (20%) 219 (22%) 5/31 (16%) 805 (53%) 0 l/2 (50%) 0 9/31 (29%)

2 3 2 11 18

following radiation ( 14,24). Positive postradiation biops- ies have been reported in 75% of patients with locally advanced disease, which is consistent with the percentage of patients found to have rising PSA levels following conventional radiotherapy (7, 8). Therefore, in locally advanced prostate cancer, standard doses of irradiation alone result in a low probability of cure, with many pa- tients harboring residual disease in the prostate itself.

Assuming that improvements in local control are neces- sary, a number of investigators have explored methods of radiation dose escalation in an attempt to demonstrate whether they are possible. These include once/day photon irradiation (9, 12, 17), proton irradiation (20,2 1 ), inter- stitial irradiation with either low or high dose rate (2, 15)) high linear energy transfer (LET) irradiation ( 16)(e.g., neutrons), and hyperfractionated photon irradiation (4).

Three groups have pioneered the use of standard frac- tionation conformal photon dose escalation in the man- agement of prostate cancer (9, 12, 17). All used different beam arrangements, volumes of irradiation, margin defi- nition, and prescription points. Schultheiss et al. ( 19), using a four-field box technique found their 14% rate of Grade 3 toxicity at 78 Gy to be prohibitively high and have discontinued further efforts at photon dose escala- tion. Leibel et al. ( 12) and Sandler et al, ( 18), have been able to reach doses of 81 and 76 Gy, respectively, without excessive toxicity. Multiple axial and/or nonaxial confor- mal fields were used in both studies. Further follow-up is anticipated but increased toxicity has already been seen with the higher photon doses in the latter study.

Proton irradiation has been evaluated as a method of conformal boost (20). In a randomized trial reported by Shipley et al. (21), photon and proton irradiation to total doses of 75.6 Gy resulted in increased toxicity without proven improvement in local control (except for poorly differentiated tumors) or survival. Evidence that the vol- ume of rectal irradiation was directly related to the inci-

dence of complications may supply additional impetus for increasingly conformal treatment delivery.

The ultimate conformal boost may be an interstitial radiotherapy boost, most commonly using ‘921r (2). The largest experience, with a minimum 5-year follow-up, has been reported by Porter et al. ( 15). In their experience, 11% of implanted patients suffered severe complications requiring surgical repair. Although improvements in nee- dle positioning, real-time dosimetric optimization, and re- duced volumes of irradiation have all contributed to re- ducing the chronic toxicity of this technique, further fol- low-up is necessary to confirm these observations. However, the net result of interstitial boosts would appear to be an improvement in clinical and histologic tumor eradication.

Improvement in local control for locally advanced pros- tate cancer has been demonstrated twice with high LET irradiation in the form of fast neutrons ( 16). In both RTOG studies (77-04 and 85-23) clinical local control was significantly improved through the use of neutrons ( 16). Severe complications can apparently be reduced through the use of conformal neutron beams, however, this technology is available in only two to three centers in the country. It was with this background, a need for improved local control, that is, methods of improving control with moderate to severe toxicity but with the re- ward of improved survival, that hyperfractionated irradia- tion was evaluated.

There are several reasons why the therapeutic ratio of external irradiation can be increased by hyperfractionated radiotherapy (3,4, 5, 22). The dose-response curves for tumor and late reacting tissues indicated that higher doses can be achieved without an increase in morbidity by using fractional doses < 1.8 to 2.0 Gy. This observation had been confirmed previously in the management of bladder cancer using a total dose of 84 Gy (1 Gy, three times daily) with which an improvement in local control and

Table 3. PSA response as a function of radiation dose and the use of neoadjuvant hormones

Without hormones Hormones Total

Nadir <4 nglml 15/21 (71%) 26128 (93%) 41/49 (84%) <l @ml lo/21 (48%) 16/28 (57%) 26149 (53%)

At last follow-up <4 rig/ml 10/20 (50%) 14/27 (52%) 24147 (49%) <l n&ml 9/20 (45%) 13/27 (48%) 22/47 (45%)

Hyperfractionated conforrnal radiotherapy l J. D, FORMAN et nl. 661

survival was demonstrated with no significant increase in normal tissue toxicity compared to a group that received 64 Gy (2 Gy/fraction) (3).

In the current study, a dose was calculated to create a fractionation scheme that would increase the total dose without increasing complications compared to standard fractionation. Using an (Y lp of 3 and 10 Gy for the normal tissues and tumor, respectively, it was expected that the two completed dose levels, 78 and 82.8 Gy would be safe. The observation that no Grade 3 or 4 complications have been observed confirms this prediction and may also reflect the benefit of the highly conformal field design and the elimination of elective pelvic lymph node irradia- tion (83 Gy). This compares favorably to the 7% Grade 3 complication rate seen in the RTO. The use of noncopla- nar beams has been shown to reduce the volume of rectal and bladder irradiation and result in a reduction in the probability of normal tissue complications ( 12, 17 ).

The relative improvement in local control as a result of this dose escalation will depend on the steepness of the dose-response curve at these dose levels. If we assume a negative postradiation biopsy rate of 25%-30% at 1 year in locally advanced patients as the baseline for standard dose patients, the 55% negative biopsy rates at 78 and

83 Gy seen in this study suggest a steep dose-response curve for local control. In addition, with biopsies done at only l-year posttreatment, many of the marked effect biopsies may become negative. In evaluating the sequelae of treatment, adequate follow-up is critical because 50% of photon complications are seen by 1 yeax and 90% by 2 years. Therefore, the low complication rate seen in the present study may be viewed as reasonably accurate. In addition, experience that dose escalation may shorten the latent period for chronic toxicity affirms this conclusion.

This study demonstrated that hyperfractionation al- lowed for safe dose escalation. On the basis of these encouraging results, we plan to proceed to the next dose level of 87.4 Gy at 1.15 Gy b.i.d. In addition, this escala- tion has resulted in a demonstrable improvement in histo- logic local control. The potential beneficial effects of neo- adjuvant hormonal therapy in conjunction with hyperfrac- tionated photon irraditaion is encouraging and similar to that seen in the RTOG experience ( 14). Further studies will demonstrate whether hyperfractionated conformal ra- diotherapy (HYCORT) will improve the rates of compli- cation-free survival in patients with locally advanced dis- ease and/or be used to improve local control in patients with organ-confined disease.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

Bagshaw, M. A. Current conflicts in the management of prostate cancer. Int. J. Radiat. Oncol. Biol. Phys. 12: 172 I- 1727; 1986. Brindle, J. S.; Martinez, A.; Schray, M.; Edmundson, G.; Benson, R. C.; Zincke, H.; Diokno, A.; Gonzalez, J. Pelvic lymphadenectomy and transperineal interstitial implanta- tion of IR 192 combined with external beam radiotherapy for bulky stage C prostatic carcinoma. Int. J. Radiat. Oncol. Phys. 17:1063-1066; 1989. Edsmyr, F.; Andersson, L.; Esposti, P.L.; L&brand, B.; Nilsson, B. Irradiation therapy with multiple small fractions per day in urinary bladder cancer. Radiother. Oncol. 4: 197- 203; 1985. Forman, J. D.; Orton, C.; Ezzell, G.; Porter, A. T. Hyper- fractionated three dimensional conformal radiotherapy for locally advanced adenocarcinoma of the prostate: Prelimi- nary results of a biologic dose escalation study. J. Ra- diother. Oncol. 27:203-208; 1993. Fowler, J. F.; Ritter, M. A rationale for fractionation for slowly proliferating tumors such as prostatic adenocarci- noma. Int. J. Radiat. Oncol. Biol. Phys. 32:521-529; 1995. Fuks, Z.; Whitmore W. F.; Leibel, S.; Nori, D.; Hilaris, B. Local control is associated with decreased distant failure in early stage carcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 17(Supp 1):121; 1989. Freiha, F. S.; Bagshaw, M. A. Carcinoma of the prostate: Results of post-irradiation biopsy. Prostate 5: 19-25; 1984. Hanks, G. E.; Lee, W. R.; Corn, B.; Schultheiss, T. E. Response to “only 20% of irradiated prostate cancer pa- tients are cured and the remainder may be made worse.” Int. J. Radiat. Oncol. Biol. Phys. 30:217; 1994. Hanks, G. E.; Martz K. L.; Diamond, J. J. The effect of dose on local control of prostate cancer. Int. J. Radiat. Oncol. Biol. Phys. 15:1299-1305; 1988.

10.

II.

12.

13.

14.

15.

16.

17.

Kaplan, E.L.; Meier, P. Nonparametric estimation ;rorn in- complete observations. J. Am. Stat. Assoc. 53:457-481; 1958. Kuban, D.; El-Mahdi, A. M.; Schellhammer, P. F. Effect of local tumor control on distant metastasis and survival in prostatic adenocarcinoma. Urology 30:420-426; 1987. Liebel S. A.; Kutcher, G. J.; Mohan. R.; Harrison L. B.; Armstrong, J. G.; Zelefsky, M. J.; LoSasso. T. J.; Burman, C. M.; Mageras, G. S.; Chui, C. S.: Brewster, L. J.; Mas- terson, 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. 2(4):274-289; 1992. Perez, C. A.; Garcia. D.; Simpson, J. R.; Zivnuska, F.; Lockett, M. A. Factors influencing outcome of definitive radiotherapy for localized carcinoma of the prostate. Ra- diother. Oncol. 16: l-21 ; 1989. Pilepich, M.; Sause, W.; Shipley, W. U.; Krall, J. M. Andro- gen deprivation with radiation therapy compared with radi- ation therapy alone for locally advanced prostatic carci- noma: A randomized comparative trial of the raidaiton ther- apy oncology group. Urology 4514; 1995. Porter, A. T.; Fortnan, J. D. Prostate Brachytherapy: An overview. Cancer 71:950-958; 1993. Russell, K. J.; Caplan, R. J.; Laramore, G. E.; Bumison, C. M.; Maor, M. H.; Taylor, M. E.; Zink, S.; Davis, L. W.; Griffin, T. W. Photon vs. fast neutron external beam radiotherapy in the treatment of locally advanced prostate cancer: Results of a randomized prospective trial. Int. J. Radiat. One. Biol. Phys. 28:47-54; 1994. Sandler, H.; McLaughlin, W.; Ten Haken, R.; Addison, H.; Forman, J. D.; Lichter, A. 3D Conformal radiotherapy for the treatment of prostate cancer: Low risk of chronic rectal

662 1. J. Radiation Oncology 0 Biology 0 Physics Volume 34, Number 3, 1996

18.

19.

20.

21.

morbidity observed in a large series of patients. I. J. Radiat. Oncol. Biol. Phys. 27 (Supp 1): 135, 1993. 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. Schultheiss, T. E.; Hanks, G.; Hunt, M. A.; Lee, W. R. Incidence of and factors related to late complications in conformal and conventional radiation treatment of cancer of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 32:643- 649; 1995.

22.

Shipley, W. U.; Prout, G. R.; Coachman, N. M.; McManus, P. L.; Healey, E. A.; Althausen, A. F.; Heney, N. M.; Park- hurst, E. C.; Young H. H., II.; Shipley, W.; Kaufman, S. D. Radiation therapy for localized prostate carcinoma: Experi- ence at the Massachusetts General Hospital ( 1973- 198 1). NC1 Monograph 7:67-73; 1988. Shipley, W. U.; Verhey, L. J.; Munzemider, J. E.; Suit,

23.

24.

H. D.; Urie, M. M.; McManus, P. L.; Young, R. H.; Shipley, J. W.; Zietman, A. L.; Biggs, P. J.; Heney, N. M.; Goitein, M. Advanced prostate cancer: The results of a randomized comparative trial of high dose irradiation boosting with conformal protons compared with conventional dose irradi- tion using photons alone. Int. J. Radiation Oncology Biol. Phys. 32:3- 12; 1995. Thames, H. D., Jr.; Withers, H. R.; Peters, L. J.; Flectcher, G. H. Changes in early and late radiation responses with altered dose fractionation: Implications for dose-survival relationships. Int. J. Radiat. Oncol. Biol. Phys. 8:219-226; 1982. Zagars, G. K.; von Eschenbach, A. C.; Johnson, D. E.; Oswald, M. J. Stage C adenocarcinoma of the prostate: An analysis of 551 patients treated with external beam radia- tion. Cancer 60: 1489- 1499; 1987. Zeitman, A. L.; Shipley, W. U.; Coen, J. J. Prostate: New insights into outcome from repeat biopsy and prostate spe- cific antigen followup. J. Urol. 5:1806-1815; 1993.