Prediction of benzimidazole therapy duration with PET/CT in inoperable patients with alveolar echinococcosis

Alveolar echinococcosis is a rare parasitic disease, most frequently affecting the liver, as a slow-growing tumor-like lesion. If inoperable, long-term benzimidazole therapy is required, which is associated with high healthcare costs and occasionally with increased morbidity. The aim of our study was to determine the role 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) in staging of patients with alveolar echinococcosis and to identify quantitative imaging parameters related to patient outcome and/or duration of benzimidazole therapy. In this single-center retrospective cohort study, 47 PET/CT performed for staging in patients with confirmed alveolar echinococcosis were analysed. In 43 patients (91%) benzimidazole therapy was initiated and was successfully stopped after a median of 870 days (766–2517) in 14/43 patients (33%). In inoperable patients, tests for trend of survivor functions displayed clear trends for longer benzimidazole therapy duration (p = 0.05; n = 25), and for longer time intervals to reach non-detectable serum concentration of Em-18 antibodies (p = 0.01, n = 15) across tertiles of SUVratio (maximum standardized uptake value in the echinococcus manifestation compared to normal liver tissue). Hence, in inoperable patients with alveolar echinococcosis, PET/CT performed for staging may predict the duration of benzimidazole therapy.

Imaging data acquisition. All imaging examinations followed basic study protocols. For PET/CT, patients fasted for at least four hours, FDG dosage was body-weight adjusted, the uptake time was standardized to 60 min in supine position, a non-enhanced CT scan was performed and used for attenuation correction, and data was acquired with arms overhead whenever possible. Body weight, height, and blood glucose level were measured prior to imaging, and blood glucose levels < 12 mmol/l were accepted 13 . Five different types of PET/CT scanners were used throughout the study period, i.e. Discovery STE, Discovery LS, Discovery RX, Discovery MI, and Discovery 690 (all GE Healthcare, Waukesha, WI). To compensate for differences in the sensitivity of the different PET/CT scanner generations, we measured the metabolic activity in normal/non-infected liver tissue and in the mediastinal blood pool for reference.
Contrast-enhanced CT was performed of the chest and/or abdomen after intravenous injection of 80 ml iodinated contrast material (Visipaque ® 320, GE Healthcare), timed for imaging at the portal venous phase with a tube voltage of 120 kV and a tube current-time product of 100-320 mAs.
MRI examinations were performed following a standard liver MRI protocol, which included at least T2-weighted sequences with and without fat saturation in axial and/or coronal plane and T1-weighted sequences with fat saturation before and after i.v. contrast administration in arterial (typically 30 s), portal venous (typically 60-90 s) and delayed phases (typically 120-240 s). Extracellular contrast agents were used from different vendors. Imaging was either performed at 1.5 or 3.0 T (Aera, Avanto or Skyra, Siemens Healthineers, Siemens, Erlangen, Germany; GE Signa HDxt or GE MR750w, GE Healthcare, Waukesha, WI; Ingenia or Achieva, Philips Healthcare, Best, the Netherlands) using dedicated phased array channel coils. Image analysis and definitions. All PET/CT and contrast-enhanced CT data sets were retrospectively reanalysed in consensus by two experienced and doubly board certified nuclear medicine physicians and radiologist on a dedicated workstation (Advantage Workstation, Version 4.6; GE Healthcare Biosciences, Pittsburgh, PA). All MRI data were reanalysed by a radiologist and a doubly board certified nuclear medicine physician and radiologist in consensus. Readers were blinded to all clinical patient outcome data, and collected the following data: For PET/CT, contrast-enhanced CT, and MRI, readers quantified the number of detectable hepatic and extrahepatic lesions, and the size of the largest lesion. With PET/CT and contrast-enhanced CT, the extent of disease was staged (i.e. PNM stage (P = parasitic mass in the liver; N = involvement of neighboring organs; M = metastasis) as defined by the WHO Informal Working Group on Echinococcosis 2,14-16 , the CT findings were classified according to the EMUC-CT classification 17,18 , and the presence and pattern of lesion calcifications was determined.
MRI findings were classified, according to the recommendations by Kodoma et al. 19 (i.e. type 1: multiple small cysts without a solid component; type 2: multiple small cysts with a solid component; type 3: a solid component surrounding a large and/or irregular cysts with multiple small cysts; type 4: a solid component without cysts; type 5: a large cyst without a solid component).
For contrast-enhanced CT and MRI, the presence of contrast enhancement was rated, using a four-point score (score 1: no enhancement; score 2: no clear enhancement; score 3; suspicion of enhancement; score 4: clear contrast enhancement).
Furthermore, quantitative imaging parameters were measured in PET/CT, i.e. maximum and peak standardized uptake value (SUVmax and SUVpeak) of FDG, in the largest and/or most FDG-avid manifestation as well as in non-infected liver tissue (the latter for reference). SUVmax is defined as the maximum intensity voxel, SUVpeak as the average activity concentration within a 1 cm 3 spherical volume of interest centered on the hottest voxel, both within a defined subspace of the PET image matrix. www.nature.com/scientificreports/ Finally, serum samples were tested at the Institute of Parasitology, University of Zurich. EMII/3-10 or its derivative EM-18 were used, which are encoded by part of the EM-10 gene sequence, and the EM-VF Western blot for serological confirmation of alveolar echinococcosis 10 .
Patient follow-up. Clinical follow-up of all patients was performed by reviewing electronic patient charts.
Patient data were recorded at the time of staging and at the last recorded clinical visit (latest retrospective followup in December 2020). Patient demographics, laboratory data, and clinical information were assessed for all patients, including data on patient survival, duration of benzimidazole therapy and time to reach no detectable levels of Em-18 antibodies.
Statistical analyses. Statistical analyses were performed using commercially available software (Stata/SE, Version 16.1, StataCorp, College Station, TX). Variables were expressed as median and IQR (25th, 75th percentiles) or percentages. We used tests for trend of survivor functions to assess the association of tertiles of SUVmax, SUVpeak, and SUVratio, the number and maximum size of echinococcus manifestations, the five categories of the KODOMA score, and the four categories of the EMUC CT classification with the duration of benzimidazole therapy and the duration to reach no detectable levels of Em-18. We used the log-rank test to assess the association of curative versus non-curative surgery with the duration of benzimidazole therapy and the duration to reach no detectable levels of Em-18. Kaplan-Meier estimates were used to describe survival from date of diagnosis to last clinical follow-up or death at 1 and 5 years. A p-value of ≤ 0.05 was considered to indicate statistical significance.

Results
Patient population. PET/CT for staging was performed in 47 patients with serology-confirmed alveolar echinococcosis. An additional contrast-enhanced CT of the abdomen or chest and abdomen was performed in 35 patients, and an additional MRI of the liver was performed in 26 patients. Demographics of the patient population are displayed in Tables 1, 2 and 3.
At the time of the PET/CT examination, benzimidazole therapy was already started in 16 (34%) patients [for 23 days ]. In four (9%) patients, no benzimidazole therapy was initiated throughout the study period (all asymptomatic patients with no detectable levels of Em-18 antibodies and negative PET/CT). Thirteen patients (28%) were operated with a curative approach (Table 2), while 34 patients (72%) were deemed inoperable (Table 3).
Contrast-enhanced CT and MRI. The disease extent was classified according to the EMUC-CT classification 17,18 for CT and according to the recommendations by Kodoma et al. 19 for MRI, and data for each patient is given in Tables 2 and 3, if the respective examination was performed. Furthermore, data on contrastenhancement is given for both imaging modalities in Tables 2 and 3.
Outcome. All patients were clinically followed for a median of 5.3 years (IQR 3.3-7.0) after their initial diagnosis and for a median of 5.1 years (IQR 3.1-6.1) after their staging PET/CT examination. The overall mortality was 1.8 per 100 patient years (95% confidence interval [CI] 0.7-4.3). The one-year and five-year freedom from all-cause mortality was 100% (CI 100-100), and 90% (CI 76-96), respectively. Five events occurred in 277.6 years of follow-up time; i.e. three patients (6%) died because of reasons considered unrelated to alveolar echinococcosis (thyroid cancer, leukemia, sepsis), in two patients (4%) the cause of death remained unknown (Table 3).
Benzimidazole therapy. In 43/47 patients (91%) with alveolar echinococcosis, benzimidazole therapy was initiated. Therapy was prematurely stopped due to adverse effects in 4/43 (9%) patients (i.e. after 74, 241, 565 and 1052 days). Four patients died during ongoing therapy (Table 3). In 14/43 (33%) patients, benzimidazole In four of the patients (9%) with negative PET/CT findings at staging, no benzimidazole therapy was initiated, and no detectable levels of antibodies were recorded throughout the entire study period. Overall 18 of 47 patients (38%) had no detectable levels of Em-18 antibodies (or its predecessor EmII/3-10) at the time of initial diagnosis; in 8 patients (17%) the data for the initial level of antibodies was unavailable, and 21 had positive levels of antibodies (45%).

Probability to stop benzimidazole therapy.
To determine the probability to stop benzimidazole therapy, we excluded all patients never treated with benzimidazole within the study period (n = 4), and all patients who received benzimidazole therapy for longer than 30 days prior to the PET/CT examination (n = 6). Hence, 37 patients were included in this analysis, 12 were operated, while 25 were not. The probability to stop benzimidazole therapy was significantly higher (p < 0.001) in operated patients (Fig. 3).
To determine whether the number, the maximum size and/or the metabolic activity in the echinococcosis manifestations measured in the staging PET/CT examination were related to the probability of stopping benzimidazole therapy, we additionally excluded all operated patients. Hence, 25 patients were included in this subanalysis, showing a clear trend for longer benzimidazole therapy duration across tertiles of SUVratio (p = 0.05; Fig. 4). Respective p-values for SUVmax, SUVpeak, SUVpeak ratio, and the number and the maximum size of the echinococcosis manifestations were not significant (0.56, 0.24, 0.08, 0.38, and 0.21). Meaningful analyses of the KODOMA and EMUC CT classification were not feasible because of very skewed distributions.

Probability to reach no detectable levels of Em-18 antibodies.
To determine the probability of reaching no detectable levels of Em-18 antibodies, we excluded all patients never treated with benzimidazole within the study period (n = 4), all patients who received benzimidazole therapy for longer than 30 days prior to the PET/CT examination (n = 6), and all patients with no available data (n = 8) or no detectable levels of Em-18 antibodies at the time of initial diagnosis (n = 10). Hence, 19 patients were included in this analysis, 4 were operated, while 15 were not. The probability of reaching no detectable levels of Em-18 was significantly higher (p < 0.001) in operated patients.   www.nature.com/scientificreports/ To determine whether the number, the maximum size and/or the metabolic activity in the echinococcosis manifestations measured in the staging PET/CT examination were related to the probability of reaching no detectable levels of Em-18, we additionally excluded all operated patients. Hence, 15 patients were included in this subanalysis, showing a significant trend for longer time intervals for reaching no detectable levels of Em-18 across tertiles of SUVratio (p = 0.011); the median time to reach no detectable levels of Em-18 was 3.8 years (95% CI 2.4-n.a.). The respective p-values for trend for the other parameters were: 0.047 for SUVpeak ratio, 0.089 for SUVmax, 0.30 for SUVpeak, 0.84 for the number, and n.s. for the maximum size of the echinococcosis  Table 3) showed intense metabolic activity [in maximum intensity reconstructions of PET (A) and fused PET/ CT images (B)] in a hilar lesion infiltrating the pancreas (dilated pancreatic duct on axial contrast-enhanced CT images in (C). SUVmax was 11.1 in the echinococcosis manifestation, and 3.1 in non-infected liver tissue; SUVratio was 3.6. After 1458 days of benzimidazole therapy, levels of Em-18 antibodies became undetectable, but therapy was not yet stopped at the end of the present study period (in total after 1633 days of therapy).  Table 3) showed mild metabolic activity [in maximum intensity reconstructions of PET (A) and fused PET/CT images (B)] at the margins of a round and hypodense lesions in the right liver lobe [as shown on axial contrastenhanced CT images in (C)]. SUVmax was 4.7 in the echinococcosis manifestations, and 3.2 in non-infected liver tissue; SUVratio was 1.5. After 1000 days of benzimidazole therapy, levels of Em-18 antibodies became undetectable, but therapy was not yet stopped at the end of the present study period (in total after 1131 days of therapy).

Discussion
We analysed the role of PET/CT in staging of patients with alveolar echinococcosis. Our study results are: (i) time to reach no detectable levels of Em-18 antibodies and the duration of benzimidazole therapy is significantly shorter in patients operated with a curative intent. (ii) in staging of inoperable patients, SUVratio (a quantitative imaging parameter derived from PET/CT) is associated with the time to reach no detectable levels of Em-18 antibodies and with the duration of benzimidazole therapy. (iii) clinical outcome was excellent, i.e. the 1-year and 5-year freedom from all-cause mortality was 100%, and 90%, respectively (five events occurred in 278 years of follow-up time; i.e. in three patients the cause of death was considered unrelated to alveolar echinococcosis, in two patients it remained unknown).
With the introduction of benzimidazole therapy, survival rates of patients with alveolar echinococcosis have become comparable to normal patient populations 20 . In our study population, five events occurred in 278 years of follow-up, and none could be clearly attributed to alveolar echinococcosis as the cause of death. Similarly, a previous publication 21 with a more selected patient population (i.e. patients were only included, if PET/CT was negative and levels of Em-18 antibodies were not detectable at the end of benzimidazole treatment) described two events, occurring in 292 patient years of follow-up, and these events also could not be clearly related to alveolar echinococcosis.
As an alternative measure of patient outcome, we analyzed the time to reach no detectable levels of Em-18 antibodies and the duration of benzimidazole therapy and its relation to curative surgery and quantitative PET/ CT parameters, acquired at initial staging of the disease.
Curative surgery of metabolically active alveolar echinococcosis is generally combined with two years of benzimidazole treatment 22 . Accordingly, in all operated patients of the present study who were followed for at  www.nature.com/scientificreports/ least two years, benzimidazole treatment was successfully stopped and Em-18 antibodies dropped to undetectable levels.
In metabolically inactive alveolar echinococcosis a watch and wait strategy may be suggested 22 . In line with this strategy, four patients of the present study cohort were not treated with benzimidazole and did not develop detectable levels of Em-18 antibodies throughout the whole study period.
In inoperable, metabolically active alveolar echinococcosis long-term benzimidazole treatment is recommended, and may be discontinued two years after echinococcus manifestations become metabolically inactive on PET/CT and Em-18 antibodies become not detectable 21,22 . To the best of our knowledge, the relation of the time interval to reach undetectable levels of EM-18 and the duration of benzimidazole therapy with the metabolic activity in PET/CT (i.e. SUVratio of the most FDG-avid liver lesion) has not been evaluated previously. Our results underline the hypothesis, that in inoperable patients, the metabolic activity in manifestations of alveolar echinococcosis in PET/CT performed for staging, is associated with the time to reach no detectable levels of Em-18 antibodies and with the duration of benzimidazole therapy. Hence, PET/CT may be used to predict the duration of benzimidazole therapy in patients with alveolar echinococcosis in the future.
Limitations of the study. A rather small number of patients was examined with all three imaging modalities (i.e. PET/CT, contrast-enhanced CT and MRI) in the present retrospective study of a rare disease. Furthermore, the study population was heterogeneous, including inoperable patients, patients undergoing curative surgery and patients who were not treated at all. However, after defining several exclusion criteria to preclude bias, we were able to perform statistically reliable subanalyses on patient outcome in inoperable patients with alveolar echinococcosis, in which PET/CT was performed for staging.

Conclusion
In inoperable patients with alveolar echinococcosis, PET/CT performed for staging may predict outcome, as the quantitative imaging parameter SUVratio is associated with the time to reach no detectable levels of Em-18 antibodies and with the duration of benzimidazole therapy. Furthermore, in selective patients with undetectable levels of Em-18 antibodies, negative PET/CT findings at staging may allow for a watch and wait strategy.