Clinical Investigation
Early Expansion of the Intracranial CSF Volume After Palliative Whole-Brain Radiotherapy: Results of a Longitudinal CT Segmentation Analysis

https://doi.org/10.1016/j.ijrobp.2009.03.016Get rights and content

Purpose

To assess cerebral atrophy after radiotherapy, we measured intracranial cerebrospinal fluid volume (ICSFV) over time after whole-brain radiotherapy (WBRT) and compared it with published normal-population data.

Methods and Materials

We identified 9 patients receiving a single course of WBRT (30 Gy in 10 fractions over 2 weeks) for ipsilateral brain metastases with at least 3 years of computed tomography follow-up. Segmentation analysis was confined to the tumor-free hemi-cranium. The technique was semiautomated by use of thresholds based on scanned image intensity. The ICSFV percentage (ratio of ICSFV to brain volume) was used for modeling purposes. Published normal-population ICSFV percentages as a function of age were used as a control. A repeated-measures model with cross-sectional (between individuals) and longitudinal (within individuals) quadratic components was fitted to the collected data. The influence of clinical factors including the use of subependymal plate shielding was studied.

Results

The median imaging follow-up was 6.25 years. There was an immediate increase (p < 0.0001) in ICSFV percentage, which decelerated over time. The clinical factors studied had no significant effect on the model.

Conclusions

WBRT immediately accelerates the rate of brain atrophy. This longitudinal study in patients with brain metastases provides a baseline against which the potential benefits of more localized radiotherapeutic techniques such as radiosurgery may be compared.

Introduction

Palliative whole-brain radiotherapy (WBRT) represents the standard of care for many patients with brain metastases (1). For most patients, survival is short. Nevertheless, a subset of patients (recursive partitioning analysis Class 1) have been identified, some of whom can be expected to survive for years (2). Preventing decline in neurocognitive function, a possible consequence of cerebral atrophy, is an important goal for these long-term survivors, particularly in view of the correlation with quality of life (3). Concerns regarding radiation toxicity have prompted recommendations for conformal radiotherapeutic techniques, such as radiosurgery, and the avoidance of WBRT whenever possible 4, 5. This argument was supported by a recently presented randomized study from M. D. Anderson Cancer Center (Houston, TX) evaluating WBRT and radiosurgery vs. radiosurgery alone (6). Chang found a significant decline in neurocognitive function at 4 months within the combined arm. Nevertheless, WBRT can improve intracranial disease control (7) and may prevent tumor-related early loss of neurocognitive function 8, 9, 10.

To date, research into the neurocognitive decline in patients with brain tumors has focused on the serial administration of rating scales for cognitive function 8, 9, 10, 11. Though very informative, the interpretation of results can be confounded by recurrent tumor and the administration of additional treatments such as chemotherapy and reirradiation. Our study has adopted a complementary approach by focusing on morphologic changes in the brain after a commonly prescribed palliative radiation dose prescription. The method selected was a segmentation technique previously developed to segment images of patients with dementia from other causes. Published normal-population data were used as a control (12).

Section snippets

Methods and Materials

Patients with a single brain metastasis completing WBRT (30 Gy in 10 fractions over a period of 2 weeks) with at least 3 years of computed tomography (CT) follow-up, within a single institution, were identified from a database of 799 consecutive consultations for patients referred with brain metastases between July 1990 and September 2003. Whole-brain irradiation was delivered by use of a parallel opposed pair of lateral whole-brain ports with the radiation dose prescribed at the midplane on

Results

Nine patients (five women and four men) met the inclusion criteria for the study. A total of 103 CT scans were performed on these patients, but 9 were unsuitable for segmentation analysis and excluded. Exclusions were because of missing axial slices or poor-quality images, which did not permit accurate discrimination between tissues. The median number of CT scans analyzed per patient was 9 (range, 5–16). The median number of CT slices segmented per scan was 18, and a total of 1,782 segmented

Discussion

This study has shown that WBRT immediately accelerates the rate at which the ICSFV percentage increases over time. It should be noted, however, that ICSFV is a relatively small intracranial compartment and the corresponding percentage reductions in the volume of brain parenchyma (cerebral atrophy) are proportionately less.

Age-associated brain atrophy was initially confirmed through autopsies (14). Segmentation studies subsequently quantified the rate of age-related atrophy, confirming an

Acknowledgments

We are grateful to Anne Martel (Imaging Research Scientist, Sunnybrook Health Sciences Centre) for her guidance on the image segmentation methodology used in this study and for comments on the final manuscript. In addition, we thank Mike Bronskill (Imaging Research, Sunnybrook Health Sciences Centre), who helped to make the segmentation software available.

References (32)

  • F. Doetsch et al.

    Subventricular zone astrocytes are neural stem cells in the adult mammalian brain

    Cell

    (1999)
  • A.A. Brandes et al.

    Radiotherapy of the brain in elderly patients. Contra

    Eur J Cancer

    (2000)
  • M.N. Tsao et al.

    Whole brain radiotherapy for the treatment of multiple brain metastases

    Cochrane Database Syst Rev

    (2006)
  • Chang E. Phase III randomized clinical trial of radiosurgery with or without whole brain irradiation in patients newly...
  • R.A. Patchell et al.

    Postoperative radiotherapy in the treatment of single metastases to the brain: A randomized trial

    JAMA

    (1998)
  • J. Li et al.

    Regression after whole-brain radiation therapy for brain metastases correlates with survival and improved neurocognitive function

    J Clin Oncol

    (2007)
  • Cited by (9)

    • Two-staged gamma knife radiosurgery for treatment of numerous (&gt;10) brain metastases

      2020, Clinical Neurology and Neurosurgery
      Citation Excerpt :

      Perilesional normal tissue irradiation increases when too many lesions are irradiated, which should result in more treatment-related complications. However, applying WBRT instead of GKS is also debatable because of its side effects such as decline in neurocognitive functions, adverse effects in quality of life measures, or increase in cerebral atrophy [27,28], and time-consuming treatment course. There are several limitations to this study.

    • Neurocognitive Effects Following Cranial Irradiation for Brain Metastases

      2015, Clinical Oncology
      Citation Excerpt :

      Data describing neurocognitive effects more than 6 months after WBRT for brain metastases are scant and limited by high dropout rates and confounding factors. An imaging study of nine long-term survivors with a median survival of 6.25 years showed acceleration in the rate of cerebral atrophy after WBRT compared with normal aging [28] but correlation with NCF was not reported. Some evidence suggests early detrimental effects may improve at later time points.

    View all citing articles on Scopus

    P.S. was supported by The Peters Fellowship in Brain Tumour Clinical Care and Research. S.L.G. and P.S. were supported by donations from W. Ren and D. Molson to the CNS Site Group at the Odette Cancer Centre.

    Conflict of interest: none.

    View full text