Clinical investigation: central nervous system
Phase II, two-arm RTOG trial (94-11) of bischloroethyl-nitrosourea plus accelerated hyperfractionated radiotherapy (64.0 or 70.4 Gy) based on tumor volume (> 20 or ≤ 20 cm2, respectively) in the treatment of newly-diagnosed radiosurgery-ineligible glioblastoma multiforme patients

https://doi.org/10.1016/S0360-3016(00)01412-7Get rights and content

Abstract

Purpose: To compare survivorship, and acute and delayed toxicities following radiation therapy (RT) of radiosurgery-ineligible glioblastoma multiforme (GBM) patients treated with tumor volume-influenced, high-dose accelerated, hyperfractionated RT plus bischloroethyl-nitrosourea (BCNU), using prior RTOG malignant glioblastoma patients as historical controls.

Methods and Materials: One hundred four of 108 patients accrued from June 1994 through May 1995 from 26 institutions were analyzable. Patients were histologically confirmed with GBM, and previously untreated. Treatment assignment (52 patients/arm) was based on tumor mass (TM), defined as the product of the maximum diameter and greatest perpendicular dimension of the titanium-gadolinium-enhanced postoperative MRI: Arm A, 64 Gy, TM > 20 cm2; or Arm B, 70.4 Gy, TM ≤ 20 cm2. Both Arms A and B received BCNU (80 mg/m2, under hyperhydration) days 1–3, 56–58, then 4 cycles, each 8 weeks, for a total of 6 treatment series.

Results: During the 24 months immediately post-treatment, the overall median survival was 9.1 months in Arm A (64 Gy) and 11.0 months in Arm B (70.4 Gy). Median survival in recursive partitioning analysis (RPA) Class III/IV was 10.4 months in Arm A and 12.2 months in Arm B, while RPA Class V/VI was 7.6 months in Arm A and 6.1 months in Arm B. There were no grade 4 neurological toxicities in Arm A; 2 grade 4 neurological toxicities were observed in Arm B (1 motor deficit, 1 necrosis at 157 days post-treatment).

Conclusion: This strategy of high-dose, accelerated hyperfractionated radiotherapy shortens overall RT treatment times while allowing dose escalation, and it provides the potential for combination with currently available, as well as newer, chemotherapy agents. Survival is comparable with previously published RTOG data, and toxicities are within acceptable limits.

Introduction

Malignant gliomas, specifically glioblastoma multiforme (GBM), have a well-established propensity for local invasiveness and tumor recurrence at, or adjacent to, their original location. While reports from randomized trials of the Brain Tumor Study Group/Brain Tumor Cooperative Group (BTSG/BTCG) 1, 2 and the Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group (RTOG/ECOG) (3) demonstrated the significant survival benefit of fractionated radiation therapy (RT) to total doses of 50–60 Gy, these clinical trials and others also demonstrated the therapeutic limits of conventional RT 4, 5. Lower RT doses are less effective, and higher doses appear to increase complications without improving tumor control. Because there is compelling radiobiologic evidence that an RT dose-response exists above 60 Gy for malignant glioma (6), several strategies to circumvent normal brain tolerance to RT have been investigated, including temporary and permanent interstitial brachytherapy 7, 8, stereotactic radiosurgery (SRS) 9, 10, and three-dimensional (3D) conformal treatment (11). Each of these approaches has produced encouraging pilot results in Phase II trials, and there are active Phase III cooperative group trials currently evaluating interstitial RT (BTCG 87-01) and stereotactic radiosurgery (RTOG 93-05).

Unfortunately, most malignant glioma patients do not meet eligibility criteria for the trials evaluating either interstitial brachytherapy or stereotactic radiosurgery (SRS). In a review of the 746 analyzable patients of a randomized dose-escalation trial (RTOG 83-02) of twice-daily RT with BCNU, only 89 (12%) met the eligibility criteria established in the first institutional report regarding SRS as boost therapy for newly diagnosed patients (12). Most patients were found to be ineligible because their maximal tumor diameter was greater than the 4.0-cm limit. One hundred ninety-two of these patients (26%) were identified as meeting the interstitial RT eligibility criteria of several institutional trials; most of these patients were also considered SRS-eligible (13). For the majority of malignant glioma patients, RT dose escalation is likely only through innovative modifications of external beam techniques.

Hyperfractionation in radiation oncology has come to mean the delivery of multiple daily RT fractions in smaller than standard dose increments, thus resulting in a larger total RT dose delivered over a standard period of time. The theoretical benefits of hyperfractionated RT are: (a) the ability to increase total RT dosage without increasing late tissue-effects; (b) an increased tumor killing for equivalent, late tissue-effects; (c) a lower oxygen enhancement ratio for lower RT fraction sizes, thereby reducing the contribution of hypoxia to radioresistance; and (d) fewer late tissue complications for equivalent, acute tissue complications.

Accelerated hyperfractionated RT (AHRT) is defined as a regimen of multiple daily RT fractions in slightly smaller than standard dose increments delivering an RT course over a shorter than standard time period. The total RT dose delivered is generally less than or equal to the dose delivered using standard fractionation. An advantage of this approach is the shortened elapsed-time during which the RT is delivered, thereby decreasing the opportunity for tumor cell repopulation during treatment 14, 15, 16, 17, 18, 19. A strong relationship appears to exist in central nervous system (CNS) irradiation between RT fraction size and the risk of late CNS injury, particularly radiation necrosis (20). It has been convincingly demonstrated that fraction size, and conversely the number of RT fractions, influences the risk of late RT neural tissue injury in a way that is more pronounced than for other organ systems (21). There are both clinical and laboratory data indicating a greater sparing of late CNS RT effects for the same total RT dose when smaller fraction sizes are utilized. For these reasons, CNS tumors, especially relatively radioresistent lesions such as malignant gliomas, are well-suited to an experimental approach using HRT or AHRT.

To determine whether the previously observed RT dose response for malignant gliomas could be further exploited through the use of HRT or AHRT, a randomized Phase I/II RT dose escalation trial with the nitrosourea BCNU was conducted from 1983 to 1989 by the RTOG (RTOG 83-02) (22). The hyperfractionated component of this study utilized 1.2 Gy, RT fractions twice daily, to a total of four RT doses between 64.8 and 81.0 Gy. From these data, 72.0 Gy was identified as the total dose worthy of Phase III testing. The 72.0-Gy arm had the best survival time (12.8–14.0 months), with a 34% survival rate at 18 months. Delivered doses > 72.0 Gy were associated with an inferior survival. (23) An RT dose-blinded analysis of white matter changes on follow-up magnetic resonance (MR) and computed tomography (CT) scans confirmed a significantly higher rate (p = 0.051) of moderate and severe treatment-related changes at doses in excess of 72.0 Gy (24). A survival analysis, which included patient quality of life in its methodology, also identified the 72.0-Gy arm as the superior arm (25). The Phase III trial comparing this 72.0-Gy arm to standard RT in once-daily fractionation with BCNU (RTOG 90-06) was completed in 1994 and showed no survival advantage for hyperfractionation (26).

The final randomization in RTOG 83-02 was between the total AHRT doses of 48.0 and 54.4 Gy, in 1.6-Gy twice-daily fractions. The observed grade 3 and 4 RT-related toxicities at 18 months were 1% and 4%, respectively, considerably less than the rates observed ≥ 72.0 Gy HRT (22). In the radiologic evaluation of treatment-related white matter changes, the incidence of moderate or severe changes was significantly less in the 48.0- and 54.4-Gy arms than any of the HRT arms tested. It appeared that further dose escalation of AHRT in twice-daily, 1.6-Gy fractionation would be possible and might provide a means of RT dose escalation in implant- and radiosurgery-ineligible glioma patients. RTOG 94-11 was therefore initiated to evaluate AHRT at doses of 64 and 70.4 Gy.

Section snippets

Patient accrual

One hundred eight patients were accrued to the two arms of this treatment protocol (Arm A, 55 patients; Arm B, 53 patients). Eligibility criteria are listed in Table 1. Pretreatment evaluation included a complete history and physical examination; laboratory values including CBC, differential, platelet count, SMA-12 including renal and liver function tests; a chest X-ray; pre and postoperative tumor imaging (CT or MRI); some indication of the extent of surgical resection (biopsy, subtotal

Results

Between June 1994 and May 1995, 108 patients were accrued to the two arms of this treatment protocol (Arm A, 55 patients; Arm B, 53 patients) in a multi-institutional cooperative group environment (26 institutions). One hundred four of the 108 patients were analyzed: 52/55 Arm A patients (1 eligibility under review; 1 ineligible; 1 patient withdrew consent); and 52/53 Arm B patients (1 patient died before any treatment).

Age, gender, race, prior surgery, Karnofsky performance score, neurologic

Discussion

In RT dose escalation trials, the significance of such treatment factors as total RT dose, fraction size, interfraction interval, and dose homogeneity regarding complication rates has been examined and reported (23). Less attention has been devoted to the influence of the volume of normal tissue receiving high-dose RT on the incidence and severity of treatment-related complications. While it appears intuitive that higher complication rates would be associated with higher RT dose volumes, this

Conclusion

One hundred four evaluable patients were treated with an accelerated hyperfractionation schedule of 1.6 Gy, twice daily (Arm A, 64 Gy; Arm B, 70.4 Gy), with the total tumor dose determined by tumor volume (Arm A, > 20 cm2; Arm B ≤ 20 cm2) in a multi-institutional cooperative group environment (26 institutions). All patients received BCNU chemotherapy during the first 3 days of radiation, and continued every 8 weeks thereafter, for a total of 6 cycles in those who did not have progressive tumor.

Acknowledgements

The authors gratefully acknowledge the editorial assistance of Luanna R. Bartholomew, Ph.D.

References (37)

  • D.F Nelson et al.

    Hyperfractionation radiation therapy and bis-chlorethyl nitrosourea in the treatment of malignant glioma. Possible advantage observed at 72.0 Gy in 1.2 Gy b.i.d. fractionsReport of RTOG protocol 83-02

    Int J Radiat Oncol Biol Phys

    (1993)
  • D Yousem et al.

    White matter changes are significantly correlated with RT doseObservations from a randomized dose escalation trial for malignant glioma (RTOG 83-02)

    Int J Radiat Oncol Biol Phys

    (1993)
  • K.J Murray et al.

    Quality-adjusted survival analysis of malignant glioma patients treated with twice-daily radiation and carmustineA report of RTOG 83-02

    Int J Radiat Oncol Biol Phys

    (1993)
  • J.R Simpson et al.

    Influence of location and extent of surgical resection on survival of patients with glioblastoma multiformeResults of three consecutive RTOG clinical trials

    Int J Radiat Oncol Biol Phys

    (1993)
  • F Ellis

    Dose time and fractionationA clinical hypothesis

    Clin Radiol

    (1969)
  • H.D Thames et al.

    Changes in early and late radiation responses with altered dose fractionationImplications for dose-survival relationships

    Int J Radiat Oncol Biol Phys

    (1982)
  • M.D Walker et al.

    Randomized comparison of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery

    N Engl J Med

    (1980)
  • C.H Chang et al.

    Comparison of postoperative radiotherapy and combined postoperative radiotherapy and chemotherapy in the multidisciplinary management of malignant gliomas

    Cancer

    (1983)
  • Cited by (0)

    View full text