Trial Information

A Phase I/II Study of Escalated-Dose Short-Course Hypofractionated Radiotherapy (55 Gy/16 Fractions/4 Weeks) for Localized Prostate Cancer

Purpose:

- To determine the acute and late effects of high-dose short-course hypofractionated
radiotherapy of 55 Gy/16 fraction/4 weeks in the treatment of localized prostate
cancer.

- To provide clinical data for confirmation of current linear-quadratic radiobiological
model of fractionation sensitivity in the treatment of prostate cancer by external beam
radiotherapy.

Radiotherapy for Prostate Cancer:

In 2003, 18,800 Canadians were expected to be diagnosed with prostate cancer, 2000 of whom
were in Alberta. Prostate cancer represents a significant workload in most Canadian cancer
centres. At the Cross Cancer Institute (CCI), approximately 400 patients underwent a
radical course of external beam radiotherapy in 2002. Typically, each course of radiation
consisted of 35 fractions given over 7 weeks. Assuming an ideal efficiency of patient
turnover on a given linear accelerator, 3000 machine hours would be required to treat 400
patients.

At present, clinical trials are on-going to confirm that an increase in radiation dose given
over 39-44 fractions per patient will be more efficacious than the current standard of 35-39
fractions. Such an increase in radiation dose will require an increase in treatment
capacity by 10% without increasing the number of patients treated.

As an alternative to increasing the number of fractions to deliver a greater total dose, the
radiation dose given in each fraction can be increased. Accordingly the total dose is
decreased to maintain acceptable toxicity. Such treatment schedule is referred to as
"hypofractionation".

Hypofractionation in prostate cancer was studied in the Phase III NCIC Clinical Trials Group
PR-5 study. The long term outcomes were reported at ASTRO in November of 2003. The results
of the PR-5 study, together with a similar Australian randomized trial reported in 2003,
strongly support the biological model of a high fractionation sensitivity of prostate
cancer, which can be exploited using hypofractionated radiotherapy.

Hypofractionation - Choice of Dose-Fractionation and Treatment Planning:

Hypofractionated radiotherapy schedules are designed based on the premise of biological
equivalent dose, modeled by the linear quadratic method. Possible schedules to achieve
equivalent results can be given over 10 to 30 fractions. In 2002 and 2003, investigators
from Toronto, Ontario and Cleveland, Ohio reported promising preliminary results based on 20
and 28 fractions per treatment course. This technique takes advantage of modern CT-based
treatment planning systems and technologies to precisely target the prostate with maximal
sparing of adjacent normal tissues.

In 2003, investigators from TBCC conducted a dosimetry study to determine the equivalent
hypofractionated dose given over 16 fractions. The results, presented by this principal
investigator at the European Cancer Conference in September 2003 and at the Canadian
Association of Radiation Oncologists scientific meeting in October 2003, showed that a dose
of 55 Gy in 16 fractions given over four weeks is a biologically equivalent course of
treatment compared to current standard treatment over seven to eight weeks at the CCI.

The hypothesis being tested in this study is that hypofractionated radiotherapy of 55 Gy/16
fractions/4 weeks will be well tolerated with acute and late toxicity comparable to
conventional radiotherapy. The secondary objective is to determine the rate of biochemical
control and the need for complex treatment planning (IMRT) as opposed to 3D conformal
therapy.

Patient Population:

Patients with low and intermediate risk prostate cancer are eligible for this study.
Specifically, patients with clinical T1 or T2, Gleason score 6 and PSA <20 and patients with
clinical T1-T2, Gleason score 7 and PS <15 are eligible.

Treatment:

Neoadjuvant hormonal therapy up to six months is allowed. Radiation dose prescribed is 55
Gy in 16 fractions over four weeks, four fractions per week. CTV is the prostate +/- 0.5-1
cm of seminal vesicles. PTV is the CTV plus 10 mm margin in all directions except posterior
(5 mm).

To ensure precise targeting using daily on-line imaging and correction of positioning prior
to treatment, fiducial markers will be inserted into the prostate under transrectal
ultrasound (TRUS) guidance. Three inert gold markers (0.98 mm diameter x 5 mm length) will
be implanted into the prostate apex, mid-gland and base. The procedure will follow similar
guidelines as for a TRUS prostate biopsy.

Treatment planning guidelines are provided for 3D planning as well as IMRT. All patients
are planned by 3D conformal technique and if the dosimetry is unsatisfactory, IMRT will be
used. Failure to meet all dose-constraint criteria for normal tissues will exclude the
patient from the trial. Details of treatment planning are provided in the full study
protocol.

Radiation Delivery:

With the use of fiducial markers, isocentre placements (ie: targeting) correction will be
made to maintain a target error of no greater than 3 mm in the right-left and
superior-inferior directions, and no greater than 2 mm in the anterior-posterior directions.
All deviations will be recorded.

Treatment Outcomes:

Other outcomes include acute and late GU/bladder and acute GI/rectum toxicity as well as
patient-assessed quality of life (Prostate Cancer Index). The percentage of patients
requiring IMRT treatment will be determined. The changes in rectal mucosa will be assessed
by endoscopy for a subset of patients who agree to undergo sigmoidoscopy at baseline, 18
months and then 36 months after radiotherapy. Tumor control will be determined as PSA
relapse free survival, using the ASTRO consensus definition for PSA failure.

Statistics and Safety Monitoring:

The sample six is one of convenience. Given the limitations imposed by the time required
for treatment planning and completing trials, the investigators estimate that one
patient/month can be accrued and treated at each of the three centres. They anticipate 72
patients will be enrolled in the trial over a two year period. This will allow estimation
of the late toxicity rate with reasonable precision.

Approximate confidence interval (CI) width assuming minimal sample size (72), based on exact
binomial interval:

Gr 3, 4 toxicity rate 0.05: 0.1: 0.15: 0.2; 95% CI width 0.12: 0.16: 0.18: 0.20.

Given the potential for increased toxicity due to the larger dose/fraction, the
investigators will implement early stopping rules for unacceptable acute toxicity. Although
the late toxicity is the endpoint of interest, acute toxicity will be used as a proxy due to
the length of time needed to determine late toxicities. Acute GI and GU toxicity have been
reported to be significant predictors of late toxicity by several investigators.

The stopping rules were determined following the method of Thall, Simon and Estey (JCO 1996)
(20). The investigators assume that the apriori distribution for the toxicity rate can be
described by Betal (1.3, 20) ie: mean toxicity rate = 0.05, 95% probability interval for
toxicity rate is between 0.00 an d0.10. The trial would be stopped if the posterior
distribution was such that :

Pr (toxicity rate > 0.10/data) >=0.80 - that is, if it is very likely that the toxicity rate
>10%.

Data Coordination:

This study will be conducted in three cancer centres, namely the Tom Baker Cancer Centre,
Calgary; the Cross Cancer Institute, Edmonton; and the Saskatoon Cancer Centre, Saskatoon.
All patient related data will be collected in the individual institution and forwarded to
the TBCC, the coordinating centre. Grade 3 or 4 toxicity and serious adverse events will be
monitored and discussed among the co-investigators. Data entry and analysis will be
performed at TBCC with statistical support from the Division of Population Health and
Information at the centre.