*2.5. Image Analysis*

Two Nuclear Medicine Physicians with over two decades experience reporting PET/CT images reviewed the hypoxia 68Ga-Nitroimidazole PET/CT images. Reconstructed images were displayed as maximum intensity projection image, PET, CT, and fused PET/CT in the axial, coronal and sagittal planes on a dedicated workstation equipped with syngo software (Siemens Medical Solutions, Lincolnshire, IL, USA).

### *2.6. Qualitative Analysis*

The 60-min whole body images were analysed for bio-distribution as well as additional tumor related information. We performed qualitative assessment of the images and recorded our findings using a grading scale of 0–3 with 0 = no uptake/uptake less than normal background tissues, 1 = uptake similar to background activity, 2 = focal uptake above background activity and 3 = uptake markedly above background activity [27]. For this visual analysis, the background was considered as the blood pool.

#### *2.7. Semi-Quantitative Analysis*

Semi-quantitative analyses were also used with calculation of SUVmax, SUVmean, HTV and Tumor to muscle ratio (TMR) as well as Tumor to blood ratio (TBR). The muscle tissue for the calculation of the TMR was the gluteus muscle and the aorta and left ventricle were used for the tumor to blood ratio. The tumor to muscle or blood ratio was calculated as the tumor SUVmax uptake divided by the SUVmean uptake in the gluteus muscle or the blood (aorta). Hypoxic tumors were identified based on the semi-quantitative analysis as those with at least one voxel with a 68Ga-Nitroimidazole signal greater than the calculated threshold. A semi-automatic spherical volume of interest (VOI) was drawn encircling the primary lesion in the cervix using an SUV threshold of 1.4 and a 3D isocontour of 41%. The VOI was manually adjusted to exclude areas of physiological uptake adjacent to the primary lesion. The maximum and mean standardized uptake values (SUVmax and SUVmean) and the hypoxic tumor volume (HTV) of the primary lesion in each patient were recorded. The hypoxic tumor volume (HTV) was defined as the volume in the 68Ga-Nitroimidazole PET dataset with an intensity greater than the determined threshold. This analysis was based on similar principles as the MTV determination. We divided the HTV by the MTV of the primary tumor to calculate a fractional HTV.

18F-FDG PET/CT images were analyzed for the presence of nodal and distant metastases. Nodal metastasis was differentiated from inflammation based on the experience of the interpreting nuclear physicians using the following criteria as indicative of metastatic nodes: the presence of asymmetry, size, architecture or consistency (absence of fatty hilum/center, necrosis), the degree of uptake and the location. Disagreements were resolved by consensus.

### *2.8. Immunohistochemical Assessment*

The slides obtained from endocervical biopsies of the cervix for initial diagnosis were retrieved for immunohistochemical analysis. The slides were evaluated by an experienced pathologist from the department of Anatomical Pathology, National Health Laboratory Services (NHLS) at our institute. Hypoxia inducible factor 1 alpha (HIF-1α) monoclonal rabbit anti-human antibodies (clone EP118) were used to evaluate the HIF-1α expression. The tissue samples had all been fixed in 10% buffered formalin, processed and embedded in paraffin routinely. Sections were cut at 3 μm using a Leica TP1020 microtome and dried overnight at 60 ◦C. After deparaffinization in xylene, the sections were rehydrated in decreasing ethanol solutions and incubated in 0.3% hydrogen peroxide for 10 min, to block endogenous peroxidase. DAKO auto-stainer with PT link method of antigen retrieval was done. Citrate buffer solution (pH 6) for 20 min was done. Peroxidase was incubated for 5 min. Then, the slides were washed in PBS and put in the autostainer. Then the antibody was put in for 30 min and washed again in PBS. Envision fluid (polymer-peroxidase method, EnVision+/HRP, DAKO, Glostrup Kommune, Denmark) was added, followed by incubation for 30 min. Bound antibodies were visualized by using 0.05% 3,3 -diaminobenzidine solution (DAB solution, DAKO). Finally, sections were counterstained with Meyer's hematoxylin and mounted in Entellan (Merck, Darmstadt, Germany). The grade of staining was defined via a light microscope. HIF-1α protein expression, (nuclear unless otherwise specified), was graded in intensity form 0 (negative) to 3 (strong), with 1 for weak intensity, 2 for moderate intensity. The distribution of staining was assessed as 25%, 50%; 75% and 100% of the tumor cells and allocated scores 1, 2, 3 and 4. A value of the sum of intensity and distribution out of 6 was used to score the positivity of the staining. HIF score = [intensity (I) + distribution (D)].

#### *2.9. Statistical Analysis*

Baseline clinical and demographic information of the patients was analyzed with descriptive statistics. Categorical data are presented as frequencies while continuous variables are presented as mean ± standard deviation (SD) or as median (interquartile range, IQR). Statistical analysis was performed using the commercially available software package SPSS 28.0. Normalcy was assessed by means of the Kolmogorov–Smirnov test. For non-normal distributed data, the Mann–Whitney test, Kruskal–Wallis test and Spearmanrank test were used when appropriate. For normal distributed data, ANOVA with posthoc Bonferroni correction, the student *t*-tests and the Pearson correlation test were used when appropriate.

Multivariate analysis was performed using Cox-regression including in sequential order of statistical significance those variables that were found to be significant in univariate analysis. Statistical significance was defined as a *p*-value ≤ 0.05

#### **3. Results**

All 20 women included in the study underwent both 68Ga-Nitroimidazole PET/CT and 18F-FDG PET/CT on different days with a median interval of 6 days (range: 1–32) between the two scans. The mean age of the study population was 44.65 ± 11.43 years (range: 26–72). The majority (n = 16/20; 80%) of the patients had squamous cell carcinoma while the rest had various other histological subtypes of cervical cancer. Most patients were referred for a scan with an initial clinical impression of locally advanced disease, however, some of the patients were upstaged by the 18F-FDG PET/CT. Seventeen patients had a FIGO clinical stage of between IIB and IIIC2 while two patients had stage IVA and one stage IVB. Thirteen patients demonstrated pelvic lymph node metastases on 18F-FDG PET/CT. The median hemoglobin level was 10 g/dL (IQR: 9–11.75 g/dL). Demographic characteristics of the patients are displayed in Table 1 below.

**Table 1.** Demographic characteristics of the patient cohort.


SD: Standard Deviation; FIGO: International Federation of Gynecology and Obstetrics.
