1

Role of Ga-68 Prostate Specific Membrane Antigen PET/CT in recurrent Gliomas

BACKGROUND

Gliomas

Gliomas constitute the bulk of primary brain tumors. Histopathologically gliomas can be stratified into four grades (I, II, III, and IV), according to WHO. Grade I or II tumors constitute low-grade gliomas with better prognosis [1]. Whereas high-grade gliomas comprise of Grade III like anaplastic ones and Grade IV like glioblastoma [1]. Although the management plan and prognosis vary significantly among different tumor grades., most them recur after primary treatment [2] And it is well known that low-grade gliomas usually recur as high-grade ones, and high-grade gliomas recur as even more aggressive tumors than their primary counterparts. So, an accurate neuroimaging is needed for detection of recurrence. The primary modality used for this purpose is a multimodal MRI with different sequences. [3] The main challenge of the imaging is to differentiate viable recurrences from the post interventional changes like pseudo-progression and radiation necrosis. These post therapy changes can cause worsening of symptoms and radiological features masquerading as a recurrence. Although biopsy still remains the gold standard for diagnosis of recurrence, it may not be feasible in all the cases. Besides, sometimes limited sampling may lead to inconclusive finding.

As a part of molecular imaging, many PET tracers have been investigated in this regard. Those include 18F-FDG being glucose analogue, 18F-FLT representing nucleoside metabolism, and 18F-FDOPA, 18F-FET, 11C-MET as amino acids analogues. Among these, 18F-FDG is the most commonly used tracer due to its broader use and easy availability [4]. However, high physiological uptake in brain, is its significant limitation. The main limitation of other tracers is the need of onsite cyclotron for their production, making their availability difficult. So, the search for an ideal modality is still on and the latest addition to this search, is radio ligand labelled Prostate Specific Membrane Antigen (PSMA). It is a new but potentially promising radio-tracer, currently showing its utility in different malignancies.

2

Prostate Specific Membrane Antigen (PSMA)

It is a type II membrane glycoprotein and is typically overexpressed in primary and metastatic lesions of prostate malignancy as the name suggests. [5] However, this overexpression is not limited to prostate cancer only. It has been reported that, PSMA is overtly expressed in the vascular endothelium of various other malignancies where significant neovascularization is seen. [6] Gliomas being highly vascularized tumors, have shown significant PSMA expression in their vascular endothelium especially in high grade ones. If this PSMA expression can be assessed through noninvasive molecular imaging, this would further ease the management these tumors. This will not only open a new avenue in the diagnosis but also as therapeutics in theses tumors. Few published articles including case reports have shown feasibility of PSMA labelled imaging in brain lesions like gliomas, CNS lymphoma and metastases. Different lesions have shown different grades of PSMA uptake in the PET/CT with high grade gliomas showing high uptake. Most of the articles published till date are either case reports or studies comprising very few numbers of patients. We would like to use the PET tracer Ga -68 PSMA for the detection of recurrence in patients with glioma.

Research Question

Does Ga-68 PSMA PET-CT has better diagnostic accuracy in detection of recurrent gliomas than conventional imaging modalities?

REVIEW OF LITERATURE

Followings are the few studies, that were published, evaluating PSMA imaging in brain tumors.

Arun Sasikumar et al (2017) assessed in ten patients, including five recurrent GBM cases, the feasibility of PSMA imaging for brain tumors and compared the findings with 18F-FDG. [7] In the five suspected recurrent glioma cases, the former showed better visualization of the recurrent lesion than the later, due to significantly high TBR in the former. They inferred that 68Ga PSMA PET-CT could be useful as non-invasive modality in detection of brain lesions.

Arun Sasikumar et al (2018) included fifteen patients with glioma with ten suspected patients recurrences in glioblastoma. [8] PSMA PET/CT scan suggested recurrences in 9 out of the ten patients, and histopathology confirmed the same. The scan-negative patient showed no evidence

3

of abnormality on follow-up imaging studies. They observed the potential of Ga-PSMA PET/CT as an imaging modality in evaluation of recurrent glioma.

Priyanka Verma et al (2019) included ten treatment naive patients with initially MRI suspected gliomas in their study to demonstrate PSMA expression in them. [9] Post surgery histopathology of brain lesions, were compared with PSMA and FDG PET findings. They found that 68Ga- PSMA PET/CT imaging is potent enough to characterize the grade of gliomas with high grade ones showing higher PSMA expression than the low-grade ones.

Elife Akgun et al (2020) tried in 35 brain tumor patients to assess if there is any correlation between pathological tumor grade and PSMA expression using PSMA PET/MR [10]. A moderate to high positive correlation between the tumor grade and PSMA expression was found. So, they concluded that 68Ga PSMA PET imaging has the potential in differentiating different grades of gliomas.

Jolanta Kunikowska et al (2020) tried to assess the potential of 68Ga-PSMA PET/CT in fifteen patients of glioblastoma who had suspicious recurrences in the MRI imaging. [11] They found PSMA PET/CT to be highly efficient in detecting recurrent glioblastoma in the studied patients.

Vallejo-Armenta et al (2021) included forty-one patients in their study with suspected brain tumors. [12] Before surgery, all the patients underwent 99mTc-PSMA SPECT brain imaging to evaluate the PSMA expression in the tumors. They found that PSMA was highly expressed in the glioblastomas and metastatic brain lesions, whereas low grade expression was seen in grade I-II gliomas. The sensitivity, specificity, PPV, NPV and accuracy of 99mTc-PSMA SPECT were 100%, 94%, 100%, 77% and 95%, respectively.

Lacunae in the literature

The above published studies assessed the feasibility of PSMA-ligand based imaging in the management of various primary and secondary brain lesion. Only one study among these, has assessed sensitivity and specificity of PSMA-ligand based imaging in detection of primary brain lesion. However, none of the studies have assessed them in recurrent brain tumors using 68Ga PSMA PET/CT or compared them with the routinely used modalities MRI and FDG PET/CT.

4

AIM & OBJECTIVES

Aim

To evaluate the efficiency of 68 Ga-PSMA-11 PET-CT imaging in detection of recurrence among patients with histologically proven gliomas and to compare it with the conventional investigations CEMRI and 18F- FDG PET/CT.

Objectives

Primary objective

Detection of sensitivity and specificity of 68Ga PSMA-11 PET-CT to diagnose disease recurrence in gliomas

Secondary objectives

1. Comparison of tumor SUVmax between 68Ga-PSMA and 18F-FDG PET-CT images

2. Comparison of sensitivity and specificity of Ga-68 PSMA-11 PET-CT with that of CEMRI in diagnosing disease recurrence in gliomas

3. Correlation between WHO tumor grade in histology and SUVmax of the recurrent lesion on 68Ga- PSMA PET-CT scan

4. Correlation between PSMA expression in histopathology specimen (using IHC) and SUVmax of recurrent lesion on 68Ga- PSMA PET-CT scan.

5. Any adverse events of Ga-68 PSMA-11 will be recorded according to CTCAE [13]

METHODOLOGY

Study design

This is a prospective, single center, observational study evaluating the accuracy of 68Ga-PSMA PET-CT in detection of recurrent in gliomas. The study will be performed in the Department of Nuclear Medicine, AIIMS, Bhubaneswar in collaboration with department of neurosurgery, radiology and pathology. The study will be performed after getting permission from the Institute Ethical Committee (IEC).

Patient selection

Inclusion Criteria:

Patients with age more than18 years.

Patients having past history of histologically proven glioma.

Patients who have undergone prior treatment with tumor resection and/or radiation therapy and/or chemotherapy.

Clinically suspected cases of recurrence.

Patient should be willing to and able to give written informed consent

Exclusion Criteria:

Pregnant lady.

Breastfeeding mother.

Patients who will deny to give consent.

Any brain primary other than high grade glioma

Patient diagnosed with other primary malignancy

Patient with life threating neurological emergency

Sample size calculation

After initial therapy of surgery and/or radiotherapy, recurrences occur in most of patients with gliomas. Currently, there is significant paucity of published studies regarding the usefulness of PSMA imaging to detect of recurrent disease in patients with known gliomas. None of the published studies have reported sensitivity or specificity of PSMA imaging in this setting.

6

However, there is one study, which used Tc-PSMA SPECT imaging for detecting different primary brain tumors and showed a sensitivity and specificity of 100% and 94% respectively. So, we expect the sensitivity and specificity of 68Ga PSMA PET/CT to be at least 90%. With the above available data, the calculated sample size is 35 using the formula n= Za/22p(1-p)/d2, where: n= sample size Z= normal distribution percentile, p= presumed sensitivity/specificity, d= maximum marginal error of estimate. For a= 0.05, the value of Za/2 is 1.96 and the precision of 10% inserted in place of d [14].

MRI imaging and interpretation

All patients with clinical suspicion of recurrent glioma will undergo gadolinium contrast enhanced MRI examination. The sequences include T1, T2, FLAIR (fluid-attenuated inversion recovery), SWI (susceptibility weighted imaging) and DWI (diffusion weighted imaging) in axial plane followed by gadolinium-enhanced fat saturated T1-weighted images in all planes. Perfusion and magnetic resonance spectroscopy (MRS) will also be executed. MRI images will be analyzed and interpreted by an experienced radiologist as per the RANO (Response Assessment in Neuro-oncology) criteria [15].

PET-CT acquisition and image analysis

18F-FDG PET-CT acquisition

Patients will be instructed to come for 18F FDG PET-CT scanning with about 6 hours of fasting. The blood glucose level will be checked in all patients before tracer injection and a value of 180 mg/dl will be taken as upper limit. About 370 MBq 18F FDG will be injected intravenously to the patients. After a gap of 45–60 min, the patients will undergo imaging on a dedicated PET-CT scanner (GE Discovery IQ GEN 2). First a diagnostic CT will be performed followed by PET study in the PET-C scanner. PET will be acquired with one or two bed positions and each bed position for 5 minutes. Attenuation correction will be applied using the CT-based transmission images. PET images will be then reconstructed using the iterative reconstruction method and then will then be evaluated in the adw software-based system.

7

68 Ga-PSMA PET-CT acquisition

68Ga-PSMA will be synthesized in-house using 68Ga-Ge generator. The entire labelling processes

will be carried in the HOT LAB of Department of Nuclear Medicine, AIIMS Bhubaneswar. After, production, about 3 mCi of 68 Ga-PSMA will be injected intravenously. One hour after the injection, PET-CT brain will be performed on a dedicated PET-CT scanner (GE Discovery IQ GEN 2) system with acquisition parameters same as that 18F-FDG PET/CT. All the imagings i.e. CEMRI, and both PET-CTs will be done within a time period of 4 weeks.

18F-FDG PET-CT image analysis

PET-CT studies will be analyzed by two NM physicians independently, without any prior knowledge about the clinical and anatomical imaging findings. CT and fused PET-CT images will be correlated during the analysis. The PET-CT imaging finding will be considered as positive when (1) there is visibly increased 18F-FDG uptake compared to the background, or (2) if there is anatomical lesion on CT with normal 18F-FDG uptake. For quantitative data, a volume of interest (VOI) will be drawn around the suspected visualized and/or previously known site of lesion and in the normal areas of contralateral brain parenchyma (background). From these VOIs, SUVmax of the tumor tissue and of the contralateral normal brain tissue will be obtained, so that tumor to background ratio can be calculated.

68Ga-PSMA PET-CT image analysis

68Ga-PSMA PET-CT images will be considered as positive for recurrence when the lesion will show obvious tracer uptake compared to the normal brain parenchyma on contralateral side. Tumor-to-background ratio will be calculated from the SUVmax of the visualized lesion/previously known site and of the contralateral normal brain parenchyma.

Histopathology examination

Histological Diagnosis and Grading:

In all the patients who will undergo re-surgery, the surgical sample will be processed before staining with Hematoxylin and Eosin. The Hematoxylin and Eosin-stained histology slides then will be examined by an experienced Neuropathologist for diagnosis and histological grading. The classification and grading will be done as per recent WHO classification of CNS tumors (2016).

8

Immunohistochemistry

The expression of PSMA will be assessed by Immunohistochemistry. Paraffin sections will be deparaffinized using xylene and graded concentration of ethanol. Sections will be rehydrated and endogenous peroxidase activity will be blocked by incubating the sections with H2O2. After antigen retrieval using appropriate buffer, endogenous biotin blocking will be done. Slides will be incubated with PSMA monoclonal antibodies for appropriate duration (subjected to standardization). Mild counterstaining will be performed with hematoxylin followed by mounting the slides.

Quantification of immunohistochemistry

Immunohistochemical scoring will be done by the neuropathologists blinded to the histological diagnosis and the clinical/ radiological parameters. Both the staining intensity and proportion of positively stained cell and will be calculated. The labeling index (LI) will be calculated as a percentage of positively stained cell. Care will be taken to exclude the inflammatory cells in the counts. The stain intensity will be assessed in a scale of 0–3 (0, no staining; 1, light brown; 2, brown; 3, dark brown). The immunohistochemical expression will be graded by combining the LI and the intensity.

Reference standard

Biopsy/pathology will be considered as the gold standard for the diagnosis of recurrent viable tumor whenever, the patient undergoes re-surgery. The patients in whom, surgery is not feasible or not opted by the patient, clinical and/or imaging follow-up will be used as the reference standard. On follow up, the patients who will have disease-related adverse events, progression of disease on imaging, will be taken as positive for recurrence. Both pre & post-interventional imaging will also be studied to exclude residual lesions from that of recurrent tumors while assessing all the radiological images.

Statistical analysis

Clinical and demographic information of the patients will be represented in the form of mean + standard deviation or median with range. All the quantitative parameters like sensitivity, specificity will be represented with 95 % confidence intervals (95% CI). Receiver operating

9

characteristic (ROC) curve analysis will be done to find the diagnostic accuracy of the PET/CT imaging modality. A correlation between the SUVmax of tumor in 68Ga PSMA PET/CT imaging and pathological PSMA expression in the IHC will be evaluated. The statistical software STATA 16.0 and SPSS 25.0 will be used for all the statistical analyses.

Safety evaluation

Ga-68 PSMA-11 has been studied extensively in human, especially in the management of prostate cancer for over a decade. It has been found to be completely safe with no reported significant adverse event. Moreover, we will do the quality control check before injecting the radiotracer to the patients.

Regarding the radiation issue, 3 mCi Ga-68 delivers approximately 2 mSv of effective radiation dose which is about a fourth of radiation dose received from CT abdomen[16]. Regarding the permissible dose limit of radiation, as per the International Atomic Energy Agency (IAEA), if there is a medical or research justification of the radiation use in the patients or volunteer, radiation may be administered judiciously [17].

Study outcomes

1. Sensitivity, specificity and diagnostic accuracy of Ga-68 PSMA-11 PET-CT in diagnosis of recurrent glioma.

2. Comparison of diagnostic accuracy of Ga-68 PSMA PET-CT with that of F-18 FDG PET-CT and CEMRI.

3. Correlation between SUVmax of lesion on Ga-68 PSMA PET-CT with pathological PSMA expression on IHC.

Novelty/Innovation

There are very few studies already published, regarding the feasibility of 68Ga PSMA PET-CT in the management of gliomas. Besides, no study as of now, has calculated the sensitivity and specificity of 68Ga PSMA PET-CT in detection of recurrence in patients with gliomas or compared the findings with routinely used imaging modalities CEMRI and FDG PET-CT.

10

Relevance of the study

We believe that 68 Ga PSMA PET-CT has the potential to play an imperative role in noninvasively evaluating recurrent gliomas and can overcome the limitations of the currently used modalities. Besides, extrapolation of the PSMA expression, which is an indirect evidence of neovascularization, can also be used to assess the treatment options like VEGF inhibitors (Bevacizumab). These studies can also pave the way for further studies involving PSMA-based radio ligand therapy, which is currently being successfully applied in patients with metastatic prostate cancer. Last but not the least, PSMA imaging may give an idea about the vascularity of the tumor for preoperative planning by the surgeon with respect to the type of approach & anticipated blood loss during surgery if needed in some instances.