PET imaging pitfall in recurrent oligodendroglioma with drug-resistant epilepsy: illustrative case

Abstract

BACKGROUND

MRI and PET imaging using radiolabeled amino acids offer a valuable adjunct for defining the extent of tumor growth in gliomas, assessing treatment response, and distinguishing true progression from pseudoprogression.

OBSERVATIONS

The authors report the case of a 37-year-old patient with intractable epilepsy and suspected recurrence of an oligodendroglioma. MRI revealed progressive gadolinium enhancement in the right frontal lobe, and both fluorethylthyrosine PET and translocator protein imaging demonstrated high tracer uptake in the same region, raising concern for tumor recurrence and progression. The patient underwent extensive resection of the right frontal lobe beyond contrast enhancement. Postoperatively, seizures were well controlled. Interestingly, histopathological analysis revealed predominantly normal brain tissue next to a limited area of tumor recurrence, demonstrating that the observed MRI and PET abnormalities were seizure-induced changes rather than tumor recurrence.

LESSONS

In patients with epileptic seizures, MRI and PET abnormalities may overestimate tumor recurrence and should be interpreted with caution.

https://thejns.org/doi/10.3171/CASE25589

Postictal MRI may display a broad spectrum of morphological abnormalities. Vasogenic edema visualized on FLAIR and T2-weighted images and restriction in diffusion-weighted imaging have been previously reported in patients after status epilepticus.¹ Interpretation of MRI abnormalities can become more challenging if a patient is suspected to have tumor recurrence, because deciding whether new contrast enhancement is the cause or a consequence of epileptic activity is crucial for treatment. PET imaging using radiolabeled amino acids offers a valuable adjunct for defining the extent of tumor growth in gliomas, assessing treatment response, and distinguishing true progression from pseudoprogression.^2,3 Amino acid transporters of the L-type (LAT) are overexpressed in neoplastic brain tissue, leading to an increased F-18-fluorethylthyrosine ([¹⁸F]FET) uptake in gliomas.² The third-generation mitochondrial translocator protein (TSPO) ligand [¹⁸F]GE-180, although primarily considered as a marker of microglia activation and neuroinflammation, also provides a high tumor-to-background contrast, due to TSPO overexpression by tumor cells.^4,5 We present the case of a patient with suspected tumor recurrence and medically intractable seizures and discuss different imaging modalities and their value for treatment planning in this context.

Illustrative Case

A 37-year-old patient with a history of right frontal oligodendroglioma, WHO CNS grade 3, was admitted to our hospital due to recurrent seizures. Up to this point, epilepsy had been well controlled with levetiracetam monotherapy. Previous therapies have included re-resection followed by 5 cycles of procarbazine and lomustine (PC) chemotherapy. Posttreatment MRI was performed 8 weeks after the final cycle of PC chemotherapy and revealed new contrast enhancement adjacent to the resection cavity, suggesting tumor recurrence (Fig. 1A and B). Resection of the contrast-enhancing lesion was proposed.

FIG. 1.

A and B: Axial T1-weighted postcontrast MR images acquired 8 weeks after the last chemotherapy cycle and prior to seizure onset. C and D:Postictal axial T1-weighted MR images with contrast agent. E and F:Corresponding axial FLAIR MR sequences demonstrating extensive signal alterations in the right frontal lobe.

At admission to the hospital, clinical examination revealed a latent paresis of the left arm with intact sensory function and cranial nerves. A mild cognitive impairment regarding attention and executive function was observed. Seizure semiology included versive seizures with head rotation to the left, tonic seizures of the left arm, and dialeptic episodes. All semiological findings were consistent with a symptomatogenic zone within the right frontal lobe.⁶ The patient exhibited no signs of meningitis; bloodwork, including electrolytes and inflammatory markers, was within normal limits. A CT scan ruled out intracranial bleeding but demonstrated progressive cerebral edema.

Despite rapid escalation to a combination of three antiepileptic drugs (AEDs), seizures remained medically intractable prior to surgery. Serial EEG demonstrated interictal epileptiform discharges originating from the right frontal lobe and transmitting to the left frontal lobe. A new MRI study performed after 8 weeks suggested further tumor progression, with T2 hyperintesity extending posteriorly to the precentral gyrus and laterally to the sylvian fissure. Cortical contrast enhancement was visible not only adjacent to the resection cavity but also extending dorsally into the prefrontal cortex (Fig. 1C and D). Corresponding FLAIR sequences demonstrated extensive signal alterations (Fig. 1E and F).

In order to further investigate the etiology of the progressive contrast enhancement, [¹⁸F]FET and TSPO PET were performed. Both scans demonstrated a highly increased tracer uptake (target-to-background ratio 5.3 and 5.1, respectively) matching the region of contrast enhancement, raising suspicion of rapid tumor growth (Fig. 2).

FIG. 2.

A: Postictal T1-weighted axial, sagittal, and coronal MR images with contrast agent. B: Axial, sagittal, and coronal FET PET images. C: Axial, sagittal, and coronal TSPO PET images showing FET and TSPO uptake matching the contrast-enhancing regions. SUV = standardized uptake value.

Given the medically intractable epilepsy and radiological evidence of progression (Fig. 1C–F), the patient underwent radical resection beyond the contrast-enhancing lesion, including the orbitofrontal cortex, the frontal pole, and gadolinium-enhancing parts of the prefrontal cortex. Intraoperatively, the cortex appeared diffusely altered with a rather firm consistency. Postoperatively, the patient developed a transient left hemiplegia, which fully resolved within 48 hours. MRI confirmed complete resection of the contrast-enhancing lesion (Fig. 3A and B).

FIG. 3.

A and B:Axial postoperative MR images with contrast agent, showing extensive resection of the right frontal lobe.

Histopathological examination of the en bloc specimen confirmed tumor recurrence; however, it was limited to an area along the border of the previous resection cavity (Fig. 1A and B) on the frontal lobe. The rest of the corresponding contrast-enhancing lesion was histologically normal brain tissue without signs of tumor infiltration.

In the postoperative period, there were no further seizures, except for a single episode in the early postoperative phase. Because the high-dose AED combination therapy resulted in sedation, AEDs were stepwise reduced and levetiracetam was exchanged with brivaracetam. The patient’s alertness improved, but he displayed signs of frontal lobe syndrome with worsening of attention and diminished self-awareness.

A final EEG study showed no epileptic activity. After a period of clinical observation without further seizures, the patient was discharged from the hospital.

During long-term follow-up over 30 months, the patient continued to exhibit signs of frontal lobe syndrome, including impulse control disorder and reduced drive. No further epileptic seizures occurred during this period. However, the triple antiepileptic therapy was maintained in consultation with his treating neurologist.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

We presented the case of a 37-year-old patient with recurrent oligodendroglioma presenting with drug-resistant epilepsy.⁷ MRI and PET imaging were suggestive of tumor recurrence.⁸ Although subtotal frontal lobe resection effectively controlled epilepsy, histopathological analysis revealed that tumor recurrence was restricted to the orbitofrontal cortex adjacent to the previous resection cavity (Fig. 1A and B); the remaining gadolinium-enhancing cortical tissue demonstrated no evidence of tumor infiltration.

Oligodendrogliomas typically exhibit both cortical and subcortical involvement, with a predilection for the frontal lobes.⁹ Contrast enhancement and calcification are seen in both WHO CNS grade 2 and 3 oligodendrogliomas.⁹

It has been previously described that seizures can cause MRI changes.¹⁰ These may include signal alterations on T2-weighted, FLAIR, and diffusion-weighted imaging, often with cortical swelling.¹⁰ In addition, transient cortical contrast enhancement has been observed as a typical seizure-associated MRI change, and occasionally leptomeningeal contrast enhancement can also be seen. ^10,11–15

Increased ¹⁸FET uptake has also been previously described in patients in the postictal phase, and histopathological examinations have shown an increased expression of LAT1/LAT2/CD98 in seizure-affected brain tissue.¹⁶ The same holds true for ¹¹C-methionine uptake, which is also mediated by LAT1/LAT2.¹⁷ In our case, TSPO PET added to the complexity of the interpretation. TSPO is a highly sensitive marker of neuroinflammation and microglia activation.^4,5 An increased TSPO uptake distant from the epileptogenic zone after a seizure has suggested microglia activation following a seizure and induction of structural changes that may facilitate further seizures.^18,19 Postictal TSPO abnormalities may persist for several weeks, with slow recovery of those metabolic changes between 4 and 12 weeks following seizure onset and antiepileptic treatment.¹⁶ MRI cortical enhancement may resolve within 10 to 21 days after seizure control.¹¹ These findings were extensively discussed in an interdisciplinary exchange with colleagues from the neuroradiology, nuclear medicine, and neurology departments, taking into account the possibility of seizure-induced changes. From our perspective, the suspicion of a oligodendroglioma, WHO CNS grade 3, tumor recurrence precluded a prolonged “watch and wait” approach to allow resolution of potential seizure-induced changes. Retrospectively, seizure interruption (which in our case would have required anesthesia) followed by a short-interval MRI study could have been considered as an alternative strategy. However, it should be noted that seizure-related MRI abnormalities can persist for several weeks, and therefore the optimal timing for follow-up has not been established, and it may depend on the duration and severity of the seizures.^11,12

When considering surgical treatment for lesional epilepsy, the resection strategy should aim to include the structural lesion and the epileptogenic zone. This is particularly challenging in extratemporal lobe epilepsy, in which the epileptogenic zone is often less well defined compared with temporal lobe epilepsy.²⁰ In order to increase the chances of seizure freedom, resection of all abnormal cortex respecting eloquent areas as well as preservation of underlying white matter tracts and en passage vessels seems a reasonable strategy.²⁰

In summary, both FET and TSPO PET failed to reliably distinguish between recurrent glioma and seizure-affected but histologically normal brain tissue. This highlights an important pitfall of PET imaging in the context of suspected tumor recurrence in patients with concurrent seizure activity.

Lessons

Drug-resistant epilepsy in glioma patients presents a significant challenge for clinical decision-making. In the present case, resection of both the lesion and metabolically altered, seizure-affected cortex resulted in seizure control and allowed for a reduction in AED therapy. It is important to recognize that gadolinium enhancement on MRI, as well as ¹⁸FET uptake and TSPO uptake, is influenced by seizures, limiting their ability to differentiate glioma from seizure-affected brain tissue. Cortical enhancement on MRI and PET abnormalities should therefore be interpreted with caution in surgical decision-making.

Disclosures

Dr. Albert reported personal fees from Telix Pharmaceuticals, Novartis, Advanced Accelerator Applications, OncLive, Servier, Medsir, ABX, ITM, and Carthera; and grants from Thomas Kirch Foundation, Novocure, Telix Pharmaceuticals, and Servier, outside the submitted work.

Author Contributions

Conception and design: Schmutzer-Sondergeld, Niedermeyer, Thon. Acquisition of data: Schmutzer-Sondergeld, Niedermeyer, Kirchleitner, Albert, Thon. Analysis and interpretation of data: Schmutzer-Sondergeld, Niedermeyer, Albert, Thon. Drafting the article: Schmutzer-Sondergeld, Niedermeyer, Albert, Thon. Critically revising the article: all authors. Reviewed submitted version of manuscript: Schmutzer-Sondergeld, Niedermeyer, Quach, Thon. Approved the final version of the manuscript on behalf of all authors: Schmutzer-Sondergeld. Administrative/technical/material support: Schmutzer-Sondergeld, Niedermeyer, Kirchleitner. Study supervision: Schmutzer-Sondergeld, Niedermeyer.

Correspondence

Michael Schmutzer-Sondergeld: LMU University Hospital, LMU Munich, Germany. michael.schmutzer@med.uni-muenchen.de.