Mapping Therapeutic Response in a Patient with Glioblastoma Multiforme through Short Interval Scanning

2000 Annual Meeting of the Society for Neuro-Oncology, Chicago, IL, Nov. 9-12, 2000

To gain insight into imaging markers and their response to therapy we mapped dynamic changes in a contrast enhancing lesion in a patient with histopathologically confirmed diagnosis of glioblastoma multiforme (GBM). The patient was scanned 15 times in 104 days on a 3T MR unit. The following imaging data were serially acquired: 3D T2-weighted, and T1 pre-and-post contrast (256x192 matrix, 24 cm field of view, inter-slice gap 0). Volumetric data for tumor, as measured from contrast enhancing tissue, necrosis and peritumoral edema were determined through the use of two 3-dimensional imaging analysis algorithms. The algorithms, a nearest neighbor tissue segmentation approach and a surface modeling algorithm were applied independently to the images. Rates of change for tumor, edema and necrosis were determined from the volumetric data. Volumetric measures clearly indicated growth. Treatment with temozolomide was initiated and volumetric data and growth rates determined.

Results. Nearest neighbor based measurements of contrast enhancing tissue initially indicated tumor growth at an approximate rate of 1.5cm³/month (0.8 cm³ on day 10 to 2.5 cm³ on day 42). Therapy was initiated on day 45 of the study. For the 11 days following the initiation of therapy an approximate growth rate of 1.4cm³/month was observed. During the remainder of the study, volumes stabilized at 3.1 cm³; a clear cessation of growth had been obtained. Change in peritumoral edema followed a similar pattern. Following the initiation of therapy, an attenuation and then a cessation of growth was observed. The alteration in growth rates may serve as a surrogate marker for therapy response.

Conclusion. Short-interval scanning provided clear evidence of an alteration in the growth rates upon initiation of treatment. Growth rate attenuation may serve as a marker for, and help characterize the dynamics of, therapeutic response. A better understanding of imaging markers and how they respond to therapy may be gained through short interval imaging. In addition, the surface modeling algorithm may be used to determine differential growth and guide biopsy to these regions.

Grant Support: (to P.T. and A.W.T.): NIMH/NIDA (P20 MH/DA52716), P41 NCRR (RR13642); NLM (LM/MH05633), NSF (BIR 93-22434), NCRR (RR05956) and NINDS/NIMH (NS38753); (to T.F.C.) NCI CA 76524, and ACS EDT-119.

Related Publications