Phase II Trial of Conventional Versus IMRT Whole Brain Radiotherapy for Brain Metastases

SCHEMA For Patients with MRI Evidence of Brain Metastasis within 1 Month Prior to
Registration

Prior to Treatment Start Confirmation of patient's insurance coverage prior to receiving
study-related procedures to e ensure that treatment with IMRT will not be denied.

Radiation Therapy

1. MRI with Fused CT Simulation

2. Neurocognitive Function Testing

3. Quality of Life Assessment

Arm 1 Whole brain radiation therapy delivered via IMRT (37.5 Gy to the brain tumors, 30 Gy
to the uninvolved brain in 15 fractions), mean dose of less than 18 Gy to the scalp

Arm 2 Conventional whole brain radiation therapy (37.5 Gy to the brain tumors and uninvolved
brain in 15 fractions)

Patient Population: (See Section 3.0 for Eligibility) At least one radiologically diagnosed
brain metastasis associated with a histologically proven diagnosis of a nonhematopoietic
malignancy. Patients must be classified as RTOG RPA class I or RPA class II

1.0 INTRODUCTION

1.1 Adverse Effects of Whole-Brain Radiotherapy (WBRT)

Whole brain radiotherapy (WBRT) remains the standard treatment approach for patients with
multiple brain metastases. WBRT has been shown to achieve rapid palliation of neurological
symptoms and improves overall survival compared to corticosteroids alone for patients with
multiple brain metastases 1. Additionally, adjuvant WBRT has been shown to improve local
control and time to neurocognitive function decline in patients with limited (1 to 4) brain
metastases that are treated with surgery or stereotactic radiosurgery 2-4. Despite
significant technical advances in radiation delivery and increased survival in tumors that
demonstrate sensitivity to systemic therapies, conventional WBRT has not materially changed
in the past 50 years 5. Conventional WBRT is generally well tolerated, save for alopecia,
fatigue and short-term neurocognitive decline in patients with a short life expectancy (≤ 6
months) 6-8. In a recent randomized trial at MD Anderson, WBRT after SRS increased the risk
of neurocognitive decline ≥ 5 points as assessed by the Hopkins Verbal Learning Test at 4
months after treatment compared to SRS alone (49% vs. 23%, p<0.05) 9. In other studies,
development of subsequent brain metastases are a significant contributor to cognitive
decline 2, 10-11. In long term survivors (≥ 12 months), irreversible neurocognitive decline
has been reported in up to 11% of patients treated with conventional WBRT, although this
study utilized daily radiation doses ≥3 Gy per day that are no longer in common use 12. The
decline in cognitive function assessed by mini mental status examination may take up to 3
years to manifest 2.

Extensive research has investigated methods of improving the efficacy of WBRT. This included
increasing the radiation dose, hyperfractionated radiation schedules and combining WBRT with
drug therapy 13-16. Currently, the standard WBRT radiation dose schedule is 30 to 37.5 Gy in
10 to 15 fractions. One promising approach to improve local control and survival has been
combining WBRT with stereotactic radiosurgery for patients with 1 to 4 metastases 6, 17. An
emerging strategy to reduce the toxicity of WBRT is to administer surgery or SRS alone for
patients with limited brain metastases 4, 9. Although this results in a higher risk of brain
relapse, some of these recurrences can be salvaged with repeat SRS and/or WBRT. For the
majority of patients with brain metastases who require WBRT, little research has focused on
improving the therapeutic ratio of WBRT by reducing its toxicity 18.

1.2 Rationale for selective targeting of brain metastases in WBRT

In general, radiation oncologists approach patients by selectively targeting the gross tumor
plus margin for microscopic extension and setup uncertainty to the prescription dose while
administering a lower dose to areas of subclinical risk 19. This strategy is extensively
utilized in brain, head and neck, lung, gastrointestinal, breast, gynecologic, hematologic
and genitourinary tumors. In many centers, over half of the patients are treated with IMRT
to improve dose distributions to increase efficacy and/or reduce toxicity. Despite the
critical physiological role played by the uninvolved brain, the reason this paradigm has not
been extended to WBRT likely relates to the general poor prognosis of patients with brain
metastases with a median survival of 4 months 20. Selectively targeting brain metastases
requires more physician, physicist and radiation therapist effort and the investment of
increased resources and cost of treatment for this patient population may be unjustified
unless the improved dose distribution translates into significant clinical benefit. In the
era of accountable care, determining the cost effectiveness of IMRT vs. conventional WBRT is
necessary.

The recent identification of long-term survivors of metastatic cancer treated with more
effective local and systemic therapies is slowly changing these attitudes 21-22. There are
distinct subgroups of patients with brain metastases with a more reasonable prognosis. For
instance RTOG recursive partitioning analysis (RPA) group 1 patients have a median survival
of 7 months 20. Patients with single brain metastases undergoing surgery followed by WBRT
have a median survival of 10 months 4. A recent study from Japan further classified RTOG RPA
class II to Class IIa, IIb and IIc based on the presence of favorable factors including
performance status, number of brain tumors, primary tumor controlled or active and
extracranial metastases 23. Survival for patients with RPA class IIa, IIb and IIc was 16 to
20 months, 8 months and 4 to 5 months respectively with long-term survivors in each
subgroup.

There have been preliminary efforts to apply IMRT to improve whole brain radiation. This
concept was first proposed by Kao, et al in 2005 24. Two avenues of research were proposed.
One approach is to selectively spare parts of the brain that are critical for neurocognitive
function, such as the hippocampal stem cells in the subgranular zone 25. This approach has
been extensively tested by investigators at the University of Wisconsin. The risk of brain
metastases in the hippocampal avoidance zone is less than 5% 26-27. RTOG 0933 is an ongoing
phase II trial testing IMRT WBRT to a total dose of 30 Gy in 10 fractions while limiting the
mean hippocampal avoidance zone dose to less than 10 Gy 25. The major criticism of this
approach is that the hippocampal avoidance zone is only one of several regions of the brain
are involved with memory processing and retention 28. Using the RTOG 0933 technique,
potentially functional brain tissue including the limbic circuit and neural stem cell region
will receive the full prescription dose even if clinically uninvolved with metastases.
Pending further study, this approach remains experimental and should not be performed
outside the context of controlled clinical trials.

A second strategy is to selectively boost areas of gross disease while simultaneously
treating the uninvolved brain with standard radiation doses 24. This strategy is currently
being tested in countries with socialized health systems such as the United Kingdom and
Canada as a cost-effective alternative to stereotactic radiosurgery boost. A published
report from England reported the feasibility of treating gross tumors to 40 Gy in 10
fractions while treating the uninvolved brain to 30 Gy in 10 fractions 29.

A third application of whole brain radiation is selective sparing of the scalp 30. The
clinical target is the whole brain but unintended radiation to the scalp can result in
temporary or permanent alopecia. Due to the dose distribution of conventional WBRT, the
vertex of the scalp receives a particularly high dose. Preliminary work suggests that IMRT
can limit the mean scalp dose to 16 to 18 Gy and these reduced doses may shorten the
duration of temporary alopecia and possibly reduce the risk of permanent alopecia 30-32.

A final strategy has not yet been explored. Rather than increasing the dose of radiation to
the identified brain tumors, we can reduce the dose to the uninvolved brain to reduce acute
and long term side effects. In the setting of prophylactic WBRT for small cell and non-small
cell lung cancer, lower radiation doses of 25 to 30 Gy in 10 to 15 fractions are considered
standard treatment 33-35. In a randomized trial of prophylactic cranial irradiation for
small cell lung cancer, there was no evidence of improved disease control with 36 Gy vs. 25
Gy 34. With the exception of a single report demonstrating subtle effects on neurocognitive
function as assessed by Hopkins Verbal Learning Test, there is little evidence that WBRT to
25 to 30 Gy in 10 to 15 fractions results in neurocognitive decline with prophylactic WBRT
vs. observation in multiple randomized controlled trials 35-36. Higher doses of WBRT have
been shown to reduce verbal memory function 37. Additionally, there are robust data from
randomized trials reproducibly demonstrating a significant reduction in the incidence of
subsequent brain metastases in regions of the brain with no dominant mass appreciable on MRI
prior to treatment 33, 35, 38. A theoretical disadvantage of limiting WBRT to 30 Gy in 15
fractions is that this dose may be inadequate to prevent relapse in relatively
radioresistant tumors. However, as discussed earlier, some investigators are now
administering 0 Gy to uninvolved sites by deferring WBRT due to concerns of toxicity 11.

1.3 Feasibility of Selective Avoidance of Uninvolved Brain and Scalp during IMRT

Based on this body of published evidence, we started utilizing IMRT for selected patients
with brain metastases in June 2012. Our planning objectives are to deliver 37.5 Gy in 15
fractions to the gross tumor(s) + 5 mm margin while limiting radiation dose to 30 Gy. A
secondary goal is to limit the mean scalp dose to 16 to 18 Gy. No effort was made to achieve
additional sparing of the hippocampal stem cells since definitive data demonstrating a
benefit has not yet been published. Based on the feasibility and promising preliminary
experience, we propose a prospective randomized trial to determine whether there are
significant benefits for WBRT delivered via IMRT.

1.4 Neurocognitive Function and Quality of Life Assessment

Although more extensive and sensitive neurocognitive tests such as the Hopkins Verbal
Learning Test are available, in the context of resources available to a high-quality
community hospital program, we will limit our neurocognitive function assessment to serial
mini-mental status examinations. Mini-mental status examination (MMSE) is the most widely
used global mental status measure in medical settings and requires less than 10 minutes to
complete. This assessment tool has been extensively validated in nearly 2,000 patients with
brain tumors treated on RTOG protocols 39.

Quality of life will be assessed using the EORTC QLQ - BN20, which is an extensively
validated one page patient-reported survey of 20 questions. The EORTC-QLQ-BN20 is designed
for use with patients undergoing chemotherapy or radiotherapy, and is composed of 20
questions assessing visual disorders, motor dysfunction, communication deficit, various
disease symptoms (eg, headaches and seizures), treatment toxicities (eg, hair loss), and
future uncertainty. The EORTC QLQ - BN20 has robust psychometric properties resulting from
rigorous testing, development, and external validity 40.

Within 2 weeks prior to WBRT, all patients will undergo a baseline quality of life
assessment.

After completion WBRT, all patients will undergo quality of life assessments every 3 months
for 6 months and then every 4 months after whole brain radiotherapy. Quality of life
assessments will be scored by a blinded reviewer to avoid potential bias.

1.5 Summary

In summary, preclinical and clinical evidence suggests that radiation dose received by
uninvolved portions of the brain during WBRT plays a critical role in causing
radiation-induced alopecia and neurocognitive decline without improving survival. Extensive
data from randomized trials suggests a benefit in reduced brain relapse from elective
treatment of uninvolved brain. Although other approaches are being pursued, reducing the
radiation dose to levels utilized for prophylactic cranial irradiation is an attractive
alternative to conventional WBRT that non-specifically irradiates the entire brain or
eliminating WBRT entirely. We hypothesize that IMRT-WBRT will reduce the incidence and
duration of alopecia while reducing the incidence of neurocognitive deficit to the
acceptable levels observed in prophylactic cranial irradiation.

ELIGIBILITY CHECKLIST

Inclusion Criteria:

- Evidence of at least one brain metastasis on a gadolinium contrast-enhanced MRI

- Pathologic/histological/cytologic proof of a diagnosis of a non-hematopoietic
malignancy within 5 years of study entry.

- Patient ≥18 years of age?

- Fall into RTOG Recursive Partition Analysis (RPA) class I or II.

- Karnofsky Performance Score ≥70. (See Appendix III)

- Biopsy done at least 1 week prior to registration. (This requirement does not apply
to stereotactic biopsies.)

Exclusion Criteria:

- Contraindication to MR imaging such as implanted metal devices or foreign bodies,
severe claustrophobia.

- Creatinine level > 1.4 mg/dl drawn ≤30 days prior to study entry.

- Severe, active co-morbitities.

- Unstable angina, and/or congestive heart failure requiring hospitalization within the
last 6 months.

- Transmural myocardial infarction within the last 6 months

- Acute bacterial or fungal infection requiring intravenous antibiotics at the time of
registration

- Hepatic insufficiency resulting in clinical jaundice and/or coagulation defects

- Uncontrolled, clinically significant cardiac arrhythmias

- Pregnancy