Earlier Diagnosis of Progressive Disease during Bevacizumab Treatment Using O-(2- 18 F-Fluorethyl)-L-Tyrosine Positron Emission Tomography in Comparison with Magnetic Resonance Imaging

Abstract

Antiangiogenic treatment using bevacizumab in brain tumor patients may cause difficulties in the diagnosis of tumor progression (ie, nonenhancing tumor progression). Newly defined criteria for treatment assessment and diagnosis of tumor progression (ie, RANO [Response Assessment in Neuro-Oncology] criteria) have implemented signal alterations on T₂/fluid-attenuated inversion recovery (FLAIR) sequences to changes in contrast enhancement. However, T₂/FLAIR hyperintensity may be influenced by other causes (eg, radiation-induced leukoencephalopathy, peritumoral edema, gliosis). Positron emission tomography using the radiolabeled amino acid O-(2-[¹⁸F]fluoroethyl)-L-tyrosine (¹⁸F-FET-PET) may help detect the metabolically active tumor extent. We present ¹⁸F-FET-PET imaging findings in a glioblastoma patient during bevacizumab treatment suggesting an earlier diagnosis of tumor progression than magnetic resonance imaging changes, which are based on the RANO criteria.

CONVENTIONAL MEASUREMENTS of tumor response or progression using magnetic resonance imaging (MRI) rely predominantly on changes in contrast-enhancing tumor parts (eg, Macdonald criteria). Antiangiogenic treatment targeting the vascular endothelial growth factor (VEGF) pathway using bevacizumab cause a rapid decrease in contrast-enhancing tumor parts, with high radiographic response rates ranging between 30 and 60%, ¹ and, furthermore, has antitumoral activity. ² However, in one-third of all patients, glioblastomas are more prone to progress as nonenhancing tumors after bevacizumab treatment. ³ Recently defined RANO (Response Assessment in Neuro-Oncology) criteria for glioma progression recommend a significant change in fluid-attenuated inversion recovery (FLAIR)/T₂ hyperintensity on MRI as a surrogate for nonenhancing tumor. ⁴ However, it may be difficult to differentiate nonenhancing tumor from other causes of FLAIR/T₂ hyperintensity (eg, radiation-induced leukoencephalopathy, peritumoral edema, gliosis). Positron emission tomography using the radiolabeled amino acid O-(2-[¹⁸F]fluoroethyl)-L-tyrosine (¹⁸F-FET-PET) may offer an improved assessment of the extent of metabolically active tumor. ⁵

Case Report

A 66-year-old male patient with a secondary glioblastoma (malignant transformation of an anaplastic astrocytoma 9 months after the initial diagnosis) was treated after malignant transformation initally according to the European Organisation for Research and Treatment of Cancer (EORTC) 26981 study. ⁶ Ten months later, the first tumor progression occurred, and the patient underwent further surgery (partial resection) and photodynamic therapy. ⁷ For treatment of the second tumor progression, biweekly therapy with bevacizumab (Avastin, Genentech/Roche, Basel, Switzerland) (10 mg/kg body weight) and irinotecan (Campto, Aventis Pharma, Frankfurt, Germany) (125 mg/m² body surface) (BEV/IR) was initiated 22 months later. ⁸ To monitor BEV/IR treatment effects longitudinally,^9,10 contrast-enhanced standard MRI and ¹⁸ F-FET-PET scans were obtained at baseline before initiation of treatment, early after 2 weeks, and during follow-up every 3 months (Figure 1).

Figure 1.

Serial contrast-enhanced MRI (upper row), T₂-weighted (middle row), and ¹⁸F-FET-PET imaging (lower row). Twelve months after BEV/IR treatment, tumor progression is detected by ¹⁸F-FET-PET (increase in tumor to brain ratio from 2.5 to 4.1). In contrast, MRI after 12 months shows no significant changes in contrast enhancement or hyperintensity on the T₂-weighted image in comparison with the previous follow-up MRIs.

Multimodal imaging before initiation of BEV/IR showed multiple nodular contrast-enhancing lesions (contrast-enhanced T₁-weighted MRI, top row) adjacent to the resection cavity in the right temporal lobe and an inhomogeneous hyperintense signal alteration on the T₂-weighted MRI (middle row). The ¹⁸F-FET-PET image (bottom row) showed metabolically active tumor predominantly in spatial correspondence with the contrast-enhancing lesion. The maximal tumor to brain ratio (TBR) before BEV/IR treatment was 5.5 (uptake within a region of interest [ROI] of the tumor region normalized against the uptake within a larger reference ROI placed on the contralateral hemisphere in an area of normal-appearing brain tissue including white and gray matter^5,11).

Early follow-up MRI 2 weeks after initiation of BEV/IR treatment revealed a slight decrease in both the intensity of contrast enhancement and the contrast-enhancing tumor volume, as well as a decrease in T₂ signal hyperintensity. According to the RANO criteria, ⁴ a partial response could not be diagnosed (ie, at least both a decrease in contrast enhancement ≥ 50% and an improved nonenhancing lesion on T₂/FLAIR MRIs). In contrast, the corresponding ¹⁸F-FET-PET image showed a clear decrease in the TBR from 5.5 to 3.6 (relative change −34%) 2 weeks after initiation of BEV/IR treatment.

During a further course of 9 months of BEV/IR treatment, MRI showed no relevant changes in both contrast enhancement and the extent of hyperintensity on the T₂-weighted images. According to the RANO criteria, ⁴ MRI findings were consistent with stable disease. Follow-up ¹⁸F-FET-PET imaging demonstrated a subsequent decrease in TBRs (reduction in TBR, 3.3, 2.9, and 2.5, respectively).

However, 12 months after BEV/IR treatment, tumor progression was detected by means of ¹⁸F-FET-PET only (increase in TBR from 2.5 to 4.1). In contrast, MRI after 12 months showed no significant changes in contrast enhancement or hyperintensity on the T₂-weighted image in comparison with the previous follow-up MRIs. Progressive disease was diagnosed with MRI according to the RANO criteria only 16 months after initiation of BEV/IR treatement. ⁴ The diagnosis of progressive disease was based on a significant increase in the nonenhancing lesion as assessed by the T₂-weighted image.^3,4

Discussion

In line with previously reported results,^{9,10,12 18}F-FET-PET may add significant information for treatment monitoring of BEV/IR in patients with recurrent high-grade gliomas (ie, earlier diagnosis of tumor progression). It should be emphasized that this diagnostic information goes far beyond what is possible by assessing treatment response using standard MRI and RANO criteria. Furthermore, the findings of this case report suggest that the amino acid PET tracer ¹⁸F-FET is able to monitor early effects of antiangiogenic treatment response, especially within a few weeks after treatment initiation (ie, as assessed by a significant decrease in TBRs). The rationale for using radiolabeled amino acids such as ¹⁸F-FET to evaluate treatment response in glioma patients is based on the increased uptake of the radiotracer in these tumors and reflects increased expression of amino acid transporters, ¹³ for example, in tumor vessels. ¹⁴ Thus, treatment-induced effects of BEV on tumor angiogenesis may change amino acid uptake.

Amino acid transport appears to be a very sensitive parameter to monitor treatment effects in patients with glioma, and results similar to those in this report could also be observed with the amino acid PET tracer L-[methyl-¹¹C]methionine (¹¹C-MET). ¹¹C-MET has been succesfully used to monitor and predict treatment effects in patients treated with standard chemotherapy, that is, adjuvant temozolomide (5 of 28) or PCV (procarbazine, CCNU [lomustine], and vincristine) chemotherapy,^15,16 dose-intensified chemotherapy with temozolomide, ¹⁷ and experimental treatment (eg, use of a tyrosine kinase inhibitor in combination with hydroxyurea). ¹⁸

Furthermore, the results of studies in patients with brain tumors using other amino acid PET tracers, for example, alpha-[¹¹C]methyl-L-tryptophan (¹¹C-AMT) or 3,4-dihydroxy-6-[¹⁸F]fluoro-l-phenylalanine (¹⁸F-FDOPA), are promising regarding the delineation of brain tumor extent in comparison with routine MRI.^19,20 Thus, these tracers could be helpful to monitor neuro-oncologic treatment. It should be noted that the use of ¹¹C-labeled tracers remains restricted to centers with an on-site cyclotron because of the short half-life of the ¹¹C isotope (20 minutes). In contrast, amino acids labeled with ¹⁸F (half-life, 109 minutes) allow more widespread use.

For the PET tracer 3′-deoxy-3′-¹⁸F-fluorothymidine (¹⁸F-FLT), a thymidine analogue reflecting deoxyribonucleic acid (DNA) metabolism, it has been previously demonstrated that a prediction of treatment response is possible within the first 2 weeks after BEV/IR treatment initiation and is superior to standard MRI.^21,22 Given that the use of ¹⁸F-FLT-PET is especially limited in tumors without disruption of the blood-brain barrier, ²³ the use of the amino acid PET tracer ¹⁸F-FET could therefore be a valuable alternative for assessment of early treatment response, and it should be evaluated in prospective studies.

PET using 2-[¹⁸F]fluoro-2-deoxy-D-glucose (¹⁸F-FDG) has been successfully used for monitoring treatment effects in patients with primary central nervous system lymphoma. ²⁴ However, the high physiologic glucose consumption of the brain, the variable glucose uptake of brain tumors, and ¹⁸F-FDG uptake in inflammatory processes may limit its use.

Conclusion

A prospective clinical trial cohort should be performed to further assess the diagnostic potential of ¹⁸F-FET uptake in assessing the effects of antiangiogenic treatment.