Challenging Axillary Lymph-Nodes on F-18-FDG, F-18-Choline and Ga-68-DOTATOC PET/CT in Cancer Patients throughout Covid-19 Vaccination Era

Objectives: unexpected detection of axillary lymphadenopathy (AxL) in cancer patients (Pts) represents a real concern during COVID-19 vaccination era. Benign reactions may take place after vaccine inoculation, confounding image interpretation in patients undergoing F-18-FDG, F-18-Choline and Ga-68-DOTATOC PET/CT. It may simulate loco-regional metastases/disease. To assess PET/CT ndings after COVID-19 vaccination in cancer patients and the consequent impact on their management. Methods: we evaluated 333 patients undergoing PET/CT (257 F-18-FDG, 54 F-18-Choline and 23 Ga-68 DOTATOC) scans after rst vaccination with mRNA vaccine (Pzer-BioNTech). Uptake index (SUVmax) of suspected AxL was dened signicant when the ratio >1.5 as compared to the contralateral inoculation site. Besides, co-registered CT (Co-CT) features of target lymph-nodes were evaluated. Nodes with aggregate positivity were further investigated. Results: overall, the prevalence of apparently positive lymph-nodes on PET scans was 17.1% during vaccination era. 107 Pts had undergone PET/CT before COVID-19 pandemic and only 3 shown reactive lymph-nodes with a prevalence of 2.8% (p<0.001 as compared to vaccination era). 84.2% exhibited benign characteristics on Co-CT images and only 9 Pts needed additional appraisal. Conclusions: the correct interpretation of images is crucial to avoid unnecessary management changes and to rule out invasive procedures in cancer Pts undergoing diagnostics. An accurate anamnestic interview and the precise assessment of nodes’ Co-CT characteristics when performing PET/CT may help to address the diagnostic hypothesis. number of patients exhibited benign characteristics on Co-registered-CT images and only few needed additional appraisals.


Introduction
The coronavirus disease 2019 (Covid-19) pandemic has caused millions of deaths with great commitment by world health services [1]. The scienti c endeavor has led to the development of vaccines useful to reduce infections, transmissibility, and mortality [2]. The P zer/BioNTech vaccine was the rst SARS-CoV-2 (Covid- 19) vaccine to be approved by Regulatory Agencies [3][4], followed by the Oxford/AstraZeneca and Moderna vaccines [5]. The administration of vaccines has been a priority for health workers and the most fragile patients such as elderly and immunosuppressed [6].
Mild systemic symptoms like fatigue, myalgia, and headache along with locoregional adverse reactions, such as shoulder pain and erythema have been reported after COVID-19 vaccination, as for other types of vaccines [7][8]. However, recent data have suggested a higher incidence of local reactive lymphadenopathy in the axillary and neck regions at the side of injection as compared to other vaccines [9]. With the increasing use of vaccines, the presence of lymphadenopathies has become a diagnostic dilemma for cancer patients. In this setting the differential diagnosis concerning adenopathy is mandatory to exclude metastatic or recurrent cancers [10]. Accordingly, the United States Society of Breast Imaging, the Canadian Society of Breast Imaging, the Canadian Association of Radiologists, and a multidisciplinary team (MDT) of experts from three leading cancer centers in USA [11] have released recommendations about the management of lymphadenopathy due to Covid-19 vaccination. Cases of ipsilateral adenopathy in the axillary and supraclavicular regions following deltoid Covid-19 vaccinations are quickly emerging in the international literature and are being increasingly referred to healthcare services for advice and management [12].
The PET/CT (Positron Emission Tomography -Computed Tomography) holds a widely validated and recognized role in cancer diagnostics [13][14][15]. On the other hand, the most common radiotracers used in PET/CT imaging are not cancer speci c at all [13]. Among the others the radio-labeled glucose analogue F-18-2-uoro-2-deoxyglucose (F-18-FDG) is the most employed tracer in PET imaging. It accumulates both in cancer and non-cancer cells having high rate of glycolysis, such as activated leukocytes in in ammatory conditions [16]. As a result, the clinical use of (F-18-FDG) PET/CT for assessing in ammatory and infectious diseases is increasing worldwide [17]. Moreover, the tracer F-18-Choline is used for the appraisal of prostate cancer. It is a radiolabeled precursor of phosphatidylcholine, an essential phospholipid of cell membranes. Both in cancer cells and reactive lymph-nodes the enhanced choline uptake is explained by the increased mitogenic activity, which requires more membrane precursors [18]. Finally, radiolabeled somatostatin analogues, such as Ga-68-DOTATOC, -DOTANOC and -DOTATATE are utilized to assess neuroendocrine tumors expressing somatostatin receptors.
We aimed to study PET/CT ndings after COVID-19 vaccination in cancer patients and the consequent impact on their management.

Patients
We assessed 333 oncological patients (overall; 143 women; 65± 13 years) who underwent F-18-FDG (132 women; 63± 13 years), or F18-Choline (all men; 74± 8 years) or Ga-68-DOTATOC (11 women; 60± 13 years) PET/CT scans from March to May 2021 for staging, re-staging or follow-up assessment. Aggregated individual data are presented in table 1. From the beginning of COVID-19 vaccination era, all patients referred either for F-18-FDG, F-18-Choline or Ga-68-DOTATOC PET/CT scans were invited to report about the COVID-19 vaccination type they received, as well as timing and anatomical site of inoculation. To make our observations as homogeneous as possible, only patients vaccinated with the rst dose of P zer-BioNTech, presenting age at entry of 18 years or older and negative pregnancy test were evaluated. Patients had undergone comprehensive reevaluation including clinical and oncological data. Insu cient information about the kind of vaccine and/or the anatomical region of injection constituted a reason for exclusion. Patients whose lymph-node sites were previously treated by recent external radiotherapy before PET/CT imaging were excluded from this study to avoid confounding items.
We also collected data from PET/CT scans in the same patients before COVID 19 pandemic, if any, as reference.
Procedures included in the study were submitted to our institutional review board for approval. All patients who underwent the study signed an informed consent form in accordance with the Declaration of Helsinki.

Imaging technique
The patients were well-hydrated before receiving F-18-FDG (370-555 MBq), F-18-Choline (230-300 MBq) or Ga-68-DOTATOC (250-300 MBq) via a peripheral venous catheter on the contralateral site of vaccination. Sixty minutes after the tracer injection, PET and modulated low dose CT were carried out with a commercial PET/CT scanner (GE Discovery VCT scanner) that combined a PET scanner and a Light Speed VCT sixty-four row MDCT system. MDCT (pitchx 1.5; 120 mAs; 120 kVp) was performed without contrast medium. The PET scanning was subsequently performed, acquiring 3 min per bed position and six to eight bed per patient depending on patient height encompassing the whole body. The raw CT data were reconstructed into transverse images with a 3.75-mm section thickness. Sagittal and coronal CT images were generated by reconstruction of the transverse data. Raw PET data were reconstructed with and without attenuation correction into transverse, sagittal, and coronal images.
Attenuation correction was based on CT attenuation coe cients, which were determined by iterative reconstruction. Patients were fasted and blood glucose levels were determined only before F-18 FDG administration (cut-off value of less than 8.3 mmol/L; 150 mg/dL).

Imaging evaluation
All hybrid images were reviewed by using PET/CT fusion software (Volumetrix for PET-CT and AW volume share 4.5, GE Healhcare). The PET studies were evaluated both visually and semi-quantitatively by SUVmax by two nuclear medicine physicians with 20 yrs of expertise. Then, the SUVmax body weight corrected values were determined by using the vendor-provided software. According to the method previously used by Thomassen et al. [21], only vaccine-draining lymph-nodes (target lesion) that had shown a SUVmax ratio ≥1.5 when compared to the contralateral site, were circumspectly studied on CT.
Bilaterality and multi-centricity were assessed, if any. All target lesions were visually assessed for shape and hilus characteristics and measured at CT in one maximum diameter at least, in short axis images by using the above-mentioned vendor-provided software. A target lesion was de ned as reactive lymph-node if two of the three abovementioned features indicated a benign nature (oval in shape; clearly evident hilus; <10mm diameter) with soft-tissue/mediastinal window settings, along with metabolic ndings.
Patients with lesions mismatching with normal anatomy or physiology and suspected of disease recurrence/relapse were referred for ultrasound and biopsy, if any. Study algorithm is reported in gure 1.

Follow-up assessment
Patients for whom a short follow-up PET/CT imaging was scheduled according to clinical needs, were reevaluated for both metabolic and morphological ndings.

Statistical Analysis
Statistical analysis was performed using MedCalc Software version 20.014. Numeric variables and frequencies of categorical variables were calculated. Fisher's exact test was used to evaluate the proportions of incidental ndings observed before and after vaccinations. A p value < 0.05 was considered statistically signi cant.
Among the 333 patients 107 had undergone PET/CT before COVID-19 pandemic and only 3 shown reactive lymph-nodes with a prevalence of 2.8% (p < 0.001 as compared to vaccination era).  Fifty-seven patients had an apparently positive lymph-node on PET/CT scan (hypermetabolic/receptor overexpression; Fig. 2), none bilateral, two patients shown multiple ipsilateral nodes.
Among them 48 (84.2%) showed benign characteristics on co-recorded CT images and were considered disease-free avoiding further investigations. Nine patients (15.8%) presented suspected CT features (round shape, max diameter > 10mm and/or not evident hilus), 7 patients underwent ultrasound appraisal that excluded adenopathy disease, two had a further mdc CT; only for one of them a metastatic disease was evoked.
Twenty-six of 57 (45.6%) patients who apparently presented positive lymph-node on PET scan throughout the vaccination era have had a PET/CT also before COVID-19 pandemic; none showed suspected axillary nodes at that time.
No muscular and subcutaneous uptake on the deltoid region were observed. The shortest time range from vaccination was 6 days, when normally the non-speci c deltoid's muscular uptake has already disappeared.
Also, twenty-eight patients of 57 (49.1%) with previous target lesion received a negative short-term PET/CT follow-up within three months. Of note, 23/47 (49%) patients who underwent F-18-FDG PET/CT and exhibited suspected axillary nodes suffered from cancers such as breast, melanoma and lymphoma, where the presence or the absence of lymphatic spread would have really changed their management. None had disease con rmed.

Discussion
To our knowledge this is the rst study assessing the PET/CT ndings with different tracers after COVID-19 vaccination in cancer patients and the consequent impact on their management.
The prevalence of locoregional PET positive lymph-nodes after SARS-CoV-2 vaccine inoculation was signi cantly higher as compared to pre-COVID-19 era.
After recently vaccine administration reactive lymphadenopathy in the ipsilateral axillary and/or supraclavicular area is most likely to occur. Unfortunately, it represents a serious challenge in cancer patients who undergo F-18-FDG, since it might mimic disease onset, relapse, or recurrence [22][23][24][25] whereas for F-18-Choline and Ga-68 DOTATOC PET/CT scans it may constitute only an unusual nding. Our patients were referred for staging, re-staging or follow-up to PET/CT examinations with different radioligands according to their diagnostic inquiry. Incidentally, the rate of apparently positive lymphnodes was signi cant in this setting. Although our ndings may appear not astonishing, they deserve caution, particularly for cancer patients as recently reported by some other authors [23][24][25]. The correct interpretation of this secondary effect of vaccine is of primary importance to avoid unnecessary changes in patients' management and to rule out undue biopsies as well as to differentiate them from malignancies, recurrence and/or metastases.
We focused on the importance of a correct anamnestic interview concerning recent vaccination in oncological patients undergoing diagnostics and particularly on the precise appraisal of co-recorded CT features when they go through PET/CT scans. In this study patients were invited to report about vaccine inoculation from the beginning of COVID-19 vaccination era. In addition, suspected lymph-nodes on the PET scan were reviewed on co-registered CT for malignancy [25]. According to daily practice, nodes with aggregate positivity were further investigated by US or mdc CT, at least [26]. This approach avoided undue further investigations for most of patients and unnecessary invasive procedures [27] for those supposed positive. In fact, 84% of patients suspected positive on PET scan resulted to have benign features on co-recorded CT images. In only one case imaging con rmed malignancy. Recently, some authors [23][24][25] have reported similar results but, in smaller cohorts, sometimes in form of case report, using one tracer and after adenovirus-based vaccine which is known to elicit in ammatory response [29]. Our study was conducted collecting data from 333 consecutive patients, irrespective of the tracer used, who underwent mRNA SARS-CoV-2 vaccination and reporting on ndings both in pre-COVID-19 period and thereafter in a subset of patients. Almost 50% of patients had PET/CT scan negative in this two moments con rming the incidental nature of the observations during vaccination age.
Moreover, we renew the importance of the correct interpretation of these incidental ndings during metabolic imaging in cancer patients to prevent unjusti ed misapprehension and the temporal enlargement of diagnostic path, as well as unnecessary changes in therapy and management.
When PET positive, the uptake index (SUVmax) was signi cant in all the three radiotracers satisfying the method of Thomassen [21] and justifying the need for further caution.
Concerning the nodes' metabolic positivity duration few information are still available. It seems that reactive processes to vaccination may last up to 70 days [30][31]. Indeed, the longer the time elapsed the lower the tracers' uptake intensity once scans performed newly. Accordingly, in our patients the difference in SUVmax values between the radiotracers could be explained by the time elapsed from vaccination and the different pharmacodynamic characteristics as showed in table 2.
Finally, we suggest careful interpretation of PET/CT data in cancer patients who recently undergo SARS-CoV-2 vaccine (and others vaccine as well) [32][33][34][35][36] considering the vaccine site of injection and addressing for subsequent imaging appraisal for the ambiguous lymph-nodes to con rm or reject the diagnostic hypothesis. In doubtful cases, the biopsy remains the last useful tool.

Perspectives
We aim to collect more data, especially from patients vaccinated with different vaccines, to evaluate any differences in terms of uptake's intensity and duration as well as from patients after the second and/or the booster dose.