The participation of the immune system in therapy response is a key element in successful cancer treatment. The preclinical GL261 glioblastoma (GB) is an immunocompetent model that can provide tools for assessing immune system participation in therapy response. Our group has developed a noninvasive method for therapy response follow-up in GL261 GB based in Magnetic resonance spectrocopy imaging (MRSI) acquisitions and generation of nosological images able to identify tumour responding to temozolomide therapy or actively proliferating, in a volumetric approach, establishing a numerical index for tumour response stratification (tumour responding index, TRI). An oscillatory pattern (6-7 day, compatible with immune cycle in brain) of TRI values was shown in longitudinal studies while no apparent changes were observed in tumour volume. In addition, lymphocyte-like cells were observed in responding cases.
The purpose of this work was to analyze retrospective cases of GL261 GB (control and TMZ-treated) and characterize the immune system population for correlation with nosological imaging data.
n=6 cases were used, being n=2 high response, n=2 intermediate response and n=2 control cases. Immunostainings for CD3 and Iba-1 for T cells and microglia/macrophage characterization, respectively, were performed. Analysis of immunostained cells was done with NDPView and Image J software and responding (R) and unresponsive/control (U/C) in vivo assessed regions were compared.
Significant (p<0.05) differences were found for CD3 positive cell counting (n=147) for R and U/C zones in all cases (4.8±2.9 vs 3.3±2.5 positive cells/field respectively). Iba-1 also presented significant (p<0.05) differences between R and U/C zones (n=148, 21.9±11.4 vs 16.8±9.7% of positive immunostained cellular areas, respectively). Responding zones could achieve values up to 42% and U/C zones could reach values as low as 1.4% for Iba-1 positive immunostaining. Although it could be contributing to the MRSI-detected spectral pattern changes, it is worth noting that large variability is seen and differences between regions within the same tumour are not so clear cut in certain individual cases. Additional immunostainings such as CD206 (M2 activated microglia/macrophage marker) and FoxP3 (regulatory T cells marker) are currently being performed.
Our results suggest that spectral pattern changes sampled in vivo by MRSI upon GB treatment could be closely related to immune population quantitative changes within tumour tissue. If fully confirmed, the nosological imaging information could allow early assessment of the immune system elicitation by a given therapy, with clear translational potential.