1016 Validation of radiation therapy oncology group (RTOG) recursive partitioning analysis (RPA) classes for patients with brain metastases: A single center experience with 647 cases

Proceedings

of the 41st Annual

ASTRO

Meeting

265

Materials & Methods: 112 patients presenting with ED after RT were identified from a prostate data base. 45 patients agreed to be evaluated for study entry. After exclusions because of current nitrate therapy and patient refusals, 34 patients with ED were included in this prospective study. All patients had received radical RT for adenocarcinoma of the prostate. Mean age of the cohort was 71 years (range, 54.79), that of sexual partners was 61.5 years (range, 34.78). 24% (8/34) had pre-RT androgen ablation (median 4 months, range, 4-6). 24% (8/34) were under hormonal manipulation on study entry. Sildenafil 100 mg was given once a week; for 6 consecutive weeks. Erectile function was evaluated weekly using the International Index of Erectile Function scale, and side effects were prospectively assessed. A re-evaluation was performed at least 3 months post-study to assess compliance, response durability and side effects. Results: 91% (31/34) patients completed the study. No patient discontinued treatment because of insufficient response. Response was gradual, with 38% (13/34). 56% (19/34), 65% (22/34), and 68% (21/31) responding at weeks 1 to 4, respectively. No increase in response rate was noted after 4 weeks. 4 of 8 (50%) patients receiving current hormonal therapy did not respond. 35% (12/34) recovered erectile function sufficient for intercourse in the absence of sildenafil. Adverse effects were: flushing (26%), transient headache (16%) and dyspepsia (11%). No patient reported priapism. With a median follow-up of 16 weeks post-study (range, 12-20) 45% of the patients with an improved erectile function reported continued use of sildenafil. For these patients reported efficacy remained identical to that noted at the end of the study period. Reasons for discontinuation were: sexual partner’s refusal (20%), loss of interest (15%), intercurrent illness (IO%), improved erectile capacity without sildenafil (5%) and fear of side effects (5%). No patient stopped sildenafil because of insufficient response. Conclusions: In men presenting with ED after RT for prostate cancer, these data suggest that sildenafil is generally well tolerated and is effective in about two-thirds of cases. Interestingly, despite the favorable therapeutic index, about half the patients discontinued regular use of sildenafil, most often because of apparent lack of interest or partner’s refusal.

10

i 5

PROGNOSTIC AGGRESSIVE CLASSES?

FACTORS TREATMENT

IN BRAIN METASTASES: SHOULD ACCORDING TO RECURSIVE

Nieder C’, Nestle U*, Motaref B*, Walter K’, Niewald Ma, Schnabel K* UT M. D. Anderson Cancer Center, Houston, TX, USAl; The Saarland University

PATIENTS PARTITIONING

Hospital,

BE SELECTED ANALYSIS

Homburg/Saar,

Purpose: To determine whether or not Radiation Therapy Oncology Group (RTOG) RPA-derived patients with brain metastases are generally applicable and can be recommended as rational strategy future clinical trials. Inclusion of time to non-CNS death as additional endpoint besides of death from in further valuable information, as survival limitation due to uncontrolled extracranial disease can be

FOR (RPA)

Germany2

prognostic classes for for patient selection for any cause might result explored.

Materials and Methods: We performed a retrospective analysis of prognostic factors for survival and time to non-CNS death in 528 patients treated at a single institution (1983-1997) with radiotherapy or surgery plus radiotherapy for brain metastases (BM). Results: Median overall survival was 2.9 months. Karnofsky-performance status (KPS) was the most important prognostic factor for survival in uni- and multivariate analyses (Cox regression analysis). We did not find other variables splitting patients with KPS less than 70% in different prognostic groups. However, advanced age, multiple BM, presence of extracranial metastases, and uncontrolled primary tumor each predicted shorter survival in patients with KPS of at least 70%. When grouped into the original RTOG RPA classes, our data set split into 3 prognostically different subgroups with median survival times of 10.5, 3.5, and 2 months, respectively (p less than 0.05). Only 3% of patients fell into the most favourable group. Median time to non-CNS death was 4.1 months (12.9 months in class 1: 4.9 months in class II, and 3.8 months in class III, respectively, p greater than 0.05 for class II vs. III). However, it was 6.6-13.3 months in class II subgroups with either solitary BM, or controlled primary tumor, or absence of extracranial metastases, or primary breast cancer. Status of primary tumor and extracranial metastases were most important in multivariate analysis of prognostic factors for time to non-CNS death. Conclusion: published by seem to profit appears to be benefit from

10 16

Despite some differences, this analysis essentially confirmed Gaspar et al. (IJROBP 37:745-751, 1997), when survival was from aggressive treatment strategies and should be included very limited. Considering time to non-CNS death, our results increased local control of BM. Yet, RPA-derived classes do

the value of RPA-derived prognostic classes, as chosen as endpoint. RPA class I patients most likely in appropriate clinical trials. However, their number suggest that other subgroups of patients also might not allow adequate selection of these patients.

VALIDATION OF RADIATION THERAPY ONCOLOGY GROUP (RTOG) RECURSIVE PARTITIONING ANALYSIS (RPA) CLASSES FOR PATIENTS WITH BRAIN METASTASES: SINGLE CENTER EXPERIENCE WITH 647 CASES

Lutterbach

.I, Stancu

University

Hospital

E, Witucki Freiburg,

G, Guttenberger Freiburg,

R, Frommhold

A

H

Germany

Purpose: To determine whether the results of the Recursive Partitioning Analysis (RPA) from the Radiation Therapy Oncology Group (RTOG) studies on brain metastases (Gaspard L et al: Int .I Radiat Oncol Biol Phys 37(4):745-751, 1997) are applicable to our patients and to search for further prognostic factors. Materials and Methods: From 1985.1996 postoperative or primary whole brain radiotherapy was delivered to 647 consecutive patients with brain metastases at our institution. 387 patients were male. 260 were female. Median age was 58 years. Pretreatment Karnofsky Performance Status (KPS) was 2 70 in 323 patients. 168 patients had uncontrolled primary tumors, 406 patients had extracerebral metastases. 240 patients bad a single lesion, multiple metastases were found in 407 cases. All patients underwent computed tomography (n = 473) or magnetic resonance imaging (n = 174). 212 patients had postoperative radiotherapy, 435 patients had primary radiotherapy. The whole brain was treated with a cobalt-60 machine or a linear

266

I. J. Radiation

Oncology

* Biology

l

Physics

Volume

45, Number

3 Supplement

1999

accelerator using a pair of opposed lateral fields. Radiotherapy was given in conventional fractionation (50 Gy/25 fx) (n = 597) or hypofractionated (30 Gy/lO fx) (n = 50). Patients were grouped according to the RTOG-RPA prognostic classification system. Class 1 (n = 60 patients): 5 65 years: KPS 2 70, controlled primary disease and no extracranial metastases; class 3 (n = 324 patients): KPS < 70; class 2 (n = 263): all other patients. Actuarial survival was calculated from the first day of radiotherapy using the product limit method of Kaplan and Meier. The log-rank test was employed to test for statistically significant differences. For multivariate analysis we used a proportional hazards model. Results: Median survival was 3.5 months for all patients. 13 patients were still alive and censored at the date of last follow-up. Median survival was 8.7 months (class l), 5.2 months (class 2) and 2.0 months (class 3), respectively (p < 0.0001). Classes (1 vs. 3 and 2 vs. 3), number of brain metastases (1 vs. > 1) and neurosurgical resection (yes vs. no) were analyzed simultaneously in a multivariate analysis. A better prognosis was found in class 1 patients compared with class 3 patients (RR 0.67. 95% C.I. 0.55-0.8), in class 2 patients patients compared with class 3 patients (RR 0.8, 95% C.I. 0.71.0.92), in patients with solitary brain metastasis compared to patients with multiple brain metastases (RR 0.86, 95% C.I. 0.78-0.96) and patients with postoperative radiotherapy compared to patients with primary radiotherapy (RR 0.76, C.I. 0.68-0.84). Conclusion: corroborates on survival. prognosis.

i 0 i 7 Lesser

Our analysis of patients with brain metastases treated the RTOG-RPA prognostic classification system. Patient In addition patients with solitary lesions and patients

INITIAL RESULTS OF A PHASE II TRIAL DIAGNOSED GLIOBLASTOMA (GBM).

G’: Kleinberg

La, Grossman

Sa, Piantadosi

with postoperative or primary whole brain radiotherapy and disease characteristics had a highly significant impact with postoperative radiotherapy had a more favourable

OF RSR13,

S’, O’Neill

A NEW

A3, Pearlman

RADIOENHANCER,

J’, Phillips

P5. Herman

IN NEWLY T6, Gerber

M7

Wake Forest University School qf Medicine, Winston-Salem, NC, USA’; The Johns Hopkins Oncology Center, Baltimore, MD, USA’; Universit)~ of Alabama at Birmingham, Birmingham, AL, USA’; H. Lee Mofjtt Cancer Center, Tampa, FL, USA4; University of Pennsylvania, Philadelphia, PA, USA’; University of Texas Health Science Center, San Antonio, TX, USA6; Allos Therapeutics, Inc., Denver; CO, USA7 Purpose: To determine survival, safety, pharmacodynamics and pharmacokinetics of RSR13 100 mg/kg per day along with cranial radiotherapy in the treatment of glioblastoma. RSR13, an allosteric modifier of hemoglobin (Hgb), is a novel radioenhancing agent which noncovalently binds to Hgb, thereby reducing oxygen binding affinity and increasing tissue oxygen release in capillaries. Materials and Methods: In this 50 patient multi-institution phase II trial, patients with newly diagnosed GBM (Karnofsky Performance Status ~60) received daily RSRI 3 100 mg/kg IV over l/2 hour along with standard cranial radiotherapy (60 Gy/30 fractions). Supplemental oxygen by nasal cannula (4 L/mm) was given during RSR13 infusion and continued until after the radiotherapy (RT) treatment. RT was given within 30 minutes of the end of the RSR13 infusion. Although accrual is complete. only 32 patients have been followed for at least a month after treatment and are currently evaluable for toxicity, pharmacodynamic and pharmacokinetic outcome. All patients are evaluable for survival. The pharmacodynamic endpoint was ~50. the partial pressure of oxygen at which hemoglobin is 50% saturated. Pharmacokinetic assessment included plasma and red blood cell RSR13 concentrations. Results: 25132 (78%) patients completed the planned RSR13 dosing. Four patients discontinued the study drug for the following reasons: allergic reaction to RSR13, hypoxemia, pulmonary embolism and sepsis. Three other patients had 5 to 8 RSR13 doses held during treatment for a bacterial infection, azotemia with nausea/vomiting, and a deep venous thrombosis but then resumed RSRl3 dosing and completed the study. 4/32 patients developed transient self-limited nonoliguric renal dysfunction without any signs of tubular injury. This generally occurred only when patients were on concurrent medications that affect renal blood flow such as antihypertensives working through the renin-angiotensin system or chronic NSAIDs. When the concurrent medications were discontinued and/or creatmme normalized, RSR13 dosing was safely resumed. The mean RSRI 3 concentration at end infusion was 571 i67 ugiml in plasma and 559% 116 ug/ml in red blood cells where RSRl3 is bound to hemoglobin. The mean peak shift in p50 was 11.5022.54 mmHg; a shift of 43% from baseline indicating an increased tendency towards oxygen unloading. With median follow-up of 5.5 months, the 6 month survival (195% confidence interval) by lifetable analysis is 842 12%. Conclusion: These preliminary results confirm that daily RSR13 administration is safe and that the desired pharmacodynamic effect can be reliably achieved. Further survival follow-up is needed to determine whether a phase III study of RSR13 with cranial radiotherapy for GBM should proceed.

10

i 8

SINGLE DOSE MENINGIOMAS

Lo SS. Cho KH. University

Hall WA,

of Minnesota

VERSUS Gerbi

Hospital

FRACTIONATED

BJ, Higgins and Clinic,

PD, Unger Minneapolis,

STEREOTACTIC J, Levitt MN,

RADIOTHERAPY

FOR

SH USA

Purpose: To evaluate the efficacy of stereotactic radiotherapy (SRT) in patients with meningiomas by comparing two different regimens, single dose and fractionated radiotherapy. Method and Materials: Between 1992 and 1998, 59 patients (16 male and 43 female) were treated with SRT. Thirty nine patients were treated with single dose radiosurgery (SRS). The remaining twenty patients with tumors adjacent to vital stmctures or of large sizes were treated with fractionated stereotactic radiotherapy (FSRT). 48 and 11 had benign and atypical or malignant meningioma respectively. The median dose for the SRS group was 1400 cGy (800-4500 cGy), and the median dose for the FSRT group was 5400 cGy (4000-6000 cGy) delivered in a median of 30 fractions (I 80.250 cGy/fraction). The median