Post-HIFU surveillance of localized prostate cancer: role of PSMA-PET imaging versus the standard PSA–mpMRI protocol Authors: Jérémy Mercier, Lucas Bento, Thibaut Cassou-Mounat, Thomas Prudhomme, Séverine Lagarde, Xavier Game, Michel Soulie, Matthieu Thoulouzan, Anne-Sophie Bajeot, Mathieu Roumiguie Published: 2025 ABSTRACT AbstractIntroductionIntermediate-risk prostate cancer (PCa) is a heterogeneous subgroup for which the optimal treatment strategy remains a subject of debate. Of all the emerging options, high-intensity focused ultrasound (HIFU) therapy aims to treat the tumor while preserving healthy prostate structures. However, the post-treatment surveillance protocol remains poorly defined, particularly due to the limitations of PSA testing and conventional imaging. This narrative review evaluates the contribution of PSMA-PET metabolic imaging in post-HIFU follow-up compared to standard techniques (PSA, multiparametric MRI).MethodsA narrative review of the literature was conducted in the Medline, PubMed, and Cochrane (2008–2025) databases. Studies evaluating post-HIFU follow-up by PSA, mpMRI, and/or PSMA-PET were included. Data were analyzed descriptively and critically by two independent reviewers.ResultsEight studies were included out of the 1245 identified articles. Although mpMRI is currently standard, its sensitivity and specificity in the treated area is limited due to post-HIFU anatomical changes. The recently developed PI-FAB score improves the standardization of interpretation but remains operator-dependent. On the other hand, PSMA-PET offers excellent specificity (100%) and satisfactory sensitivity in detecting occult recurrences, even when PSA levels are low and including after mpMRI and systematic biopsies have failed.ConclusionPSMA-PET is a molecular imaging modality that appears promising for post-HIFU surveillance of favorable intermediate-risk prostate cancer. As a complement to PSA testing and multiparametric MRI, it can refine the detection and localization of recurrences and, in doing so, influence the selection and sequencing of salvage therapies in routine practice. Prospective studies are still needed to establish its impact on treatment decision-making and long-term oncologic outcomes.Keywords: Prostate cancer, Focal therapy, HIFU, Surveillance, PSMA, PET/CT, PET/MRI, Multiparametric MRI, PSA, Diagnostic accuracy. Introduction Prostate cancer (PCa) is the most common solid cancer in men in France, with nearly 60,000 new cases per year according to Insee (French National Institute of Statistics and Economic Studies) [1]. Thanks to improved screening techniques and the development of new therapies, survival rates at 5 and 10 years are 93% and 80%, respectively [2]. This high prevalence coupled with increasing longevity means that how to best balance the oncological benefit against the morbidity of treatments continues to be a topic of debate. Localized PCa is classified according to the risk of biological recurrence at 5 years [3]. While active surveillance is currently standard for low-risk forms, there are still differing views regarding the optimal strategy for intermediate-risk PCa. This heterogeneous risk group includes patients with a favorable prognosis, similar to that of low-risk patients in terms of occurrence of metastasis at 10 years, and for whom radical treatments can be avoided [4]. It is in this context that focal therapies, including HIFU (high-intensity focused ultrasound), have emerged. These therapies aim to destroy the tumor while preserving the surrounding healthy prostate tissue, thus minimizing side effects. Since 2023, the French National Authority for Health (Haute Autorité de Santé [HAS]) has authorized HIFU as a first-line curative treatment for patients aged 70 years and up who have localized intermediate-risk PCa and are ineligible for conventional treatments [5]. Outside of this context, HIFU should be offered as part of a clinical research protocol or register. While it seems as though this focal strategy is becoming part of the therapeutic arsenal, post-treatment surveillance of patients remains poorly codified. Surveillance based on clinical examination and PSA (prostate-specific antigen) testing is recommended [2]. Biological recurrence is defined as a PSA rise of + 2 ng/ml above the nadir according to the Phoenix criteria [6]. In practice, surveillance is complex because these criteria are tailored to whole-gland treatment and therefore seem less relevant for focal treatments. As is the case for diagnosis, multiparametric magnetic resonance imaging (mpMRI) of the prostate can be used in this context to identify targets where recurrence is suspected and which require biopsies. However, the use of morphological imaging following treatment that has resulted in structural changes to the organ is questionable. The hypothesis of this study is that current metabolic imaging (PSMA-PET) could rectify this issue. The objective of our review was to shed light on the techniques and results of the current follow-up protocol for patients who have undergone focal HIFU prostate treatment (focal, hemiablation, subtotal) and summarize the results of studies evaluating the use of PSMA-PET metabolic imaging. Materials and methods Research strategy A systematic search of the Medline, PubMed, and Cochrane Library databases from January 2008 to January 2025 was carried out. The keywords used included: “HIFU/focal therapy/ablation; prostate cancer; perioperative/postoperative/surveillance; PSA; mpMRI; PET/CT/PSMA.” Study selection and quality assessment Original publications in English or French that were published between January 1, 2008, and January 31, 2025 and evaluated follow-up after focal HIFU treatment of PCa (focal, hemiablation, or subtotal) were included. Studies had to explicitly describe a surveillance protocol based on PSA testing, mpMRI, and/or PSMA metabolic imaging (PET/CT or PET/MRI), with measurable results in terms of diagnostic performance or clinical impact. Articles that did not meet these criteria were excluded: reviews, editorials, letters to the editor, conference abstracts without complete data, publications in other languages, or studies without a recurrence evaluation protocol or that did not report relevant clinical indicators. Each study was examined by two independent reviewers (ASB and JM) to reduce selection bias. Following an initial selection based on titles and abstracts, a full-text analysis was carried out to confirm the article’s selection. Finally, the references of the selected articles were reviewed in order to identify other possible sources of data. Any disagreements regarding the selection of studies were resolved by a third reviewer (MR). The methodological and clinical diversity of the included studies meant that a meta-analysis would not be appropriate. A narrative summary of the data was therefore undertaken. Results The literature search identified 1,245 records. After removal of duplicates and screening of titles and abstracts, 1,112 records were excluded because they did not meet the predefined eligibility criteria (e.g. not involving HIFU treatment, no post-HIFU surveillance imaging, non-original articles, or insufficient outcome data). Of the 133 full-text articles assessed for eligibility, a further 125 were excluded for the same reasons, leaving eight studies that fulfilled all inclusion criteria and were included in this narrative review (Fig. 1). PRISMA flow diagram 1/Contribution of mpMRI in post-HIFU follow-up mpMRI protocol mpMRI protocol mpMRI is the standard imaging tool for perioperative follow-up of patients who have received focal therapy for PCa. It enables morphological (T2-WI), functional (diffusion, ADC), and vascular (DCE) assessment, which is essential for both planning treatment and detecting recurrences. Tissue changes induced by the thermal destruction of the gland often complicate the interpretation of anatomical and vascular images, thereby making it more difficult to detect recurrence in the treated areas [7]. Schaudinn et al. described several characteristics of post-HIFU mpMRI in patients with no recurrence at 12 months, including artifacts associated with fibrosis, gland asymmetries or residual hematomas [8]. These non-specific enhancement patterns make distinguishing between benign post-treatment changes and tumor recurrence a more complicated task. The acquisition protocol included T2-weighted (T2w) in three planes (axial, sagittal, coronal), as well as an axial T1-weighted with injection of a contrast agent (ce-T1w) following bolus administration of 15 to 20 ml of gadobutrol (Gadovist®, Bayer Pharmaceuticals). Axial diffusion-weighted imaging (DWI) was included in the protocol for 27 of the 30 patients at two of the three centers. Postoperative assessment by mpMRI following focal treatment of PCa is currently undertaken using standard protocols for the early detection of PCa in untreated glands. b.Diagnostic performance Diagnostic performance A recent meta-analysis evaluated the diagnostic performance of mpMRI imaging in detecting recurrence of PCa following HIFU treatment [9]. A total of 19 studies (of which 17 were publications and 2 conference abstracts) and 703 patients were included. The studies were methodologically heterogeneous in terms of MRI protocols (1.5 T vs. 3 T, multiparametric or non-multiparametric), HIFU treatment methods (partial or subtotal ablation) and radiological interpretation criteria, which were often derived from PI-RADS and unsuitable for the post-treatment setting. The overall sensitivity and specificity of mpMRI for detecting local recurrences of PCa were 0.81 [95% CI: 0.72–0.90] and 0.91 [95% CI: 0.86–0.96], respectively, with the area under the ROC curve being 0.81. However, subgroup analyses revealed poorer performance in large studies of > 50 patients, reducing sensitivity to 68% and specificity to 75%, which would suggest that the diagnostic performance may have been overestimated in small studies. In addition, the heterogeneous definitions of suspicious lesions on MRI and the lack of standardization in post-HIFU surveillance protocols mean that the comparability of results is limited. These findings highlight the need for a specific interpretation system for post-partial prostate ablation MRI. They also illustrate the specific challenges associated with imaging following focal prostate treatment, particularly in differentiating between recurrences in treated target areas and those occurring outside of the target areas. c.PI-FAB score development PI-FAB score development In 2019, Hötker et al. confirmed the significant change in the appearance of mpMRI following focal therapy, highlighting the importance of standardizing interpretation [10]. With this in mind, Giganti et al. proposed the PI-FAB (Prostate Imaging after Focal Ablation) score in 2023, which is a visual assessment system specifically for post-focal therapy MRI [11]. PI-FAB uses a 3-point scale to assess the three MRI sequences in sequential order: [1] dynamic contrast-enhanced (DCE) sequences; [2] DWI (diffusion-weighted imaging), divided into assessment of the high b-value sequence first, followed by the ADC (apparent diffusion coefficient) map; and [3] T2-WI (T2-weighted imaging). Key information such as the date of treatment, ablation method, tumor burden (including pre-treatment Gleason grade group, maximum cancer core length and PSA), and post-treatment PSA kinetics should always be provided by the medical consultant. In a recent retrospective single-center study, Pausch et al. evaluated the PI-FAB score’s relevance for post-treatment follow-up of patients who had undergone focused HIFU treatment for localized PCa [12]. The main objective was to analyze the inter-reader reproducibility and diagnostic performance of the PI-FAB when applied to mpMRI at different points in time during the follow-up. Seventy-three patients who had undergone focal HIFU treatment and follow-up combining mpMRI and targeted prostate biopsies were included. Two radiologists with different levels of experience reviewed the mpMRIs carried out at 6, 12 and 36 months post-HIFU using the PI-FAB score. Diagnostic performance was compared against the histopathological biopsy results. The PI-FAB score showed significant to near-perfect inter-reader agreement (AC1 coefficient: 0.80–0.95), as well as a high specificity (87–98%) and negative predictive value (88–100%) for excluding intra-target recurrence. However, sensitivity (reader 1: 43%, reader 2: 14%) and positive predictive value (14–33%) were more modest, with notable differences between readers, suggesting that this is influenced by radiological expertise. At 36 months, only two cases of intra-target recurrence were reported, meaning that the sensitivities (100%/50%) and specificities (90%/87%) should be interpreted with caution. The study did not provide any information on post-treatment PSA kinetics. Thus, despite its significant diagnostic value, mpMRI remains limited by post-treatment glandular asymmetry and the insufficient specificity of certain radiological signs. The PI-FAB score represents a significant step forward, but it needs to be integrated into expert reading and possibly supplemented by more sensitive forms of imaging. 2/Specific studies evaluating PSMA-PET in post-HIFU follow-up Positron emission tomography (PET) imaging targeting prostate-specific membrane antigen (PSMA) is a major step forward in the evaluation of PCa, particularly in the context of recurrence or clinical suspicion despite negative conventional examinations. The ccAFU recommendations confirm that PSMA-PET does not have any place as a first-line test for diagnosing localized PCa [2]. However, it may be considered in specific situations, particularly in cases where there is a strong clinical suspicion accompanied by negative biopsies or contraindications to mpMRI, as well as to guide a second series of biopsies. Two recent studies have demonstrated the value of this method, based primarily on use of the tracer [68Ga]-PSMA-11, in post-focal HIFU treatment follow-up for localized PCa. Technique Technique In the study by Burger et al., the authors used a hybrid PET/MRI system (Signa PET/MRI, GE Healthcare) with an 85 MBq injection of [68Ga]-PSMA-11 combined with the administration of furosemide to limit bladder activity [13]. A 15-minute targeted acquisition of the prostate revealed occult recurring foci on mpMRI. In a clinical case reported by Topoozian et al., a [68Ga]-PSMA-11 PET/CT scan detected an intraprostatic lesion that could not be seen on MRI or systematic biopsies, guided the focal ablation by HIFU, then confirmed the response by the disappearance of the PSMA focus [14]. From a technical perspective, these studies illustrate the versatility of the [68Ga]-PSMA-11 radiotracer, which is used in various different protocols depending on the type of imaging (PET/MRI or PET/CT), injected doses (from 85 to 176 MBq) and acquisition times (Table 1). The use of synchronous multiparametric MRI sequences enables good anatomical localization, while quantitative analysis by SUVmax or SUVpeak improves lesion characterization. Table 1Technical characteristics of studies using PSMA-PET following HIFUStudyRadiolabelMethodEquipmentDoseTime after injectionSpecific feature Burger et al. (2019) [13][68Ga]-PSMA-11PET/MRISigna PET/MRI85 MBq60 min15 min targeted pelvic imaging Topoozian et al. (2023) [14][68Ga]-PSMA-11PET/CTNot statedNot specifiedNot specifiedPET/CT-US fusion for biopsy targeting and post-HIFU follow-upPET Positron Emission Tomography, PET/CT Positron Emission Tomography combined with Computed Tomography, PET/MRI Positron Emission Tomography combined with Magnetic Resonance Imaging, PSMA Prostate-Specific Membrane Antigen, HIFU High-Intensity Focused Ultrasound, mpMRI Multiparametric Magnetic Resonance Imaging, PI-FAB Prostate Imaging after Focal Ablation, AUC area under the ROC curve, n.r. not reported Technical characteristics of studies using PSMA-PET following HIFU PET Positron Emission Tomography, PET/CT Positron Emission Tomography combined with Computed Tomography, PET/MRI Positron Emission Tomography combined with Magnetic Resonance Imaging, PSMA Prostate-Specific Membrane Antigen, HIFU High-Intensity Focused Ultrasound, mpMRI Multiparametric Magnetic Resonance Imaging, PI-FAB Prostate Imaging after Focal Ablation, AUC area under the ROC curve, n.r. not reported b.Diagnostic performance Diagnostic performance In 2019, Burger et al. evaluated [68Ga]-PSMA-11 PET/MRI in 10 patients with histologically proven recurrence following HIFU but negative MRI [13]. All recurrences were in the area initially treated using HIFU. PET/MRI detected recurrence in 6 patients with a sensitivity of 55%, specificity of 100%, PPV of 100% and NPV of 85%. All clinically significant tumors (Gleason ≥ 3 + 4 or ≥ ISUP 2) were identified, highlighting the potential of this method for detecting occult recurrences. In 2023, a case report described a 75-year-old man with persistent PSA elevation who underwent several mpMRIs and systematic biopsies, all of which were negative [14]. PSMA imaging revealed a focal lesion in the left middle lobe of the prostate with intense avidity (SUV = 5). A PSMA-PET/CT-ultrasound fusion guided biopsy confirmed a Gleason 7 (3 + 4) adenocarcinoma. The patient was treated using focused ultrasound ablation. Post-treatment follow-up at seven months showed complete disappearance of the PSMA-avid focus and no suspicious lesions on mpMRI. A targeted biopsy confirmed the absence of residual cancer, revealing only scar tissue. At one year, PSA had fallen to 0.7 ng/mL and no urinary or sexual function impairment was observed. Tables 2 and 3 summarize the results of mpMRI and PSMA-PET/CT performance in the post-treatment follow-up of patients who have received HIFU treatment. Table 2Summary of clinical studies evaluating PSMA-PET molecular imaging in post-HIFU follow-upReferenceStudy typePopulationMain objectiveImaging techniqueKey resultsSpecific contribution of PSMA-PET Burger et al. (2019) [13]Interim analysis of a prospective single-center study10 patients with post-HIFU biochemical recurrence, positive biopsy and negative mpMRITo evaluate the ability of 68Ga-PSMA-11 PET/MRI to detect local recurrences that cannot be seen on mpMRI[68Ga]-PSMA-11 PET/MRI carried out within 3 months following biopsyQuandrant-by-quadrant sensitivity: 55%; Specificity: 100%; all lesions with an ISUP score of ≥ 2 were detectedDetection of clinically significant local recurrences not identified by mpMRI following HIFU Topoozian et al. (2023) [14]Case reportPatient with systematically negative MRI and biopsies but persistent clinical suspicionTo evaluate the usefulness of [68Ga]-PSMA-11 PET/CT in detecting clinically significant prostate cancer[68Ga]-PSMA-11 PET/CTIdentification of a focus of intense PSMA uptake in the prostate; guided biopsy confirming clinically significant cancerDetection of lesions that cannot be seen on MRI, enabling accurate diagnosis despite negative conventional examinations Summary of clinical studies evaluating PSMA-PET molecular imaging in post-HIFU follow-up Table 3Comparison of the diagnostic performance of multiparametric MRI and PSMA-PET in post-HIFU follow-upStudyMethodPopulationResultsMain comments Pausch et al. (2024) [12]mpMRI (PI-FAB reading)55 patients post-HIFUAUC = 0.87Inter-reader agreement k ≥ 0.70Initial validation of the PI-FAB score, good reproducibility and satisfactory performance Ahn et al. (2023) [9]mpMRI (meta-analysis)812 patients (12 studies)Sensibility 81%Specificity 73%Inter-study heterogeneity; satisfactory but variable performance depending on the criteria Giganti et al. (2023) [11]mpMRI (PI-FAB proposal)Expert protocoln.r.Standardization proposal to improve post-treatment interpretation Burger et al. (2019) [13][68Ga]-PSMA-11 PET/MRI10 patients with a positive biopsy and negative mpMRISensitivity 55% (quadrant-based)Specificity 100%Detection of recurrences that cannot be seen on MRI in 100% of patients with a Gleason score of ≥ 4 + 3 Topoozian et al. (2023) [14][68Ga]-PSMA-11 PET/CT1 patient with negative conventional imagingn.r.Identification of a suspicious lesion that cannot be seen on MRI, confirmed by biopsyPET Positron Emission Tomography, PET/CT Positron Emission Tomography combined with Computed Tomography, PET/MRI Positron Emission Tomography combined with Magnetic Resonance Imaging, PSMA Prostate-Specific Membrane Antigen, HIFU High-Intensity Focused Ultrasound, mpMRI Multiparametric Magnetic Resonance Imaging, PI-FAB Prostate Imaging after Focal Ablation, AUC area under the ROC curve, n.r. not reported Comparison of the diagnostic performance of multiparametric MRI and PSMA-PET in post-HIFU follow-up AUC = 0.87 Inter-reader agreement k ≥ 0.70 Sensibility 81% Specificity 73% Sensitivity 55% (quadrant-based) Specificity 100% PET Positron Emission Tomography, PET/CT Positron Emission Tomography combined with Computed Tomography, PET/MRI Positron Emission Tomography combined with Magnetic Resonance Imaging, PSMA Prostate-Specific Membrane Antigen, HIFU High-Intensity Focused Ultrasound, mpMRI Multiparametric Magnetic Resonance Imaging, PI-FAB Prostate Imaging after Focal Ablation, AUC area under the ROC curve, n.r. not reported Discussion Optimizing follow-up for patients with favorable intermediate-risk PCa that have been treated using HIFU focal therapy is an issue highlighted by the HAS [5]. While focal therapy is a promising alternative to radical treatments, it also requires post-treatment follow-up to be codified. This narrative review underscores the limitations of MRI-PSA and highlights the growing potential of molecular imaging, particularly PSMA-PET, as a complementary strategy to address the issue of patient follow-up. mpMRI is currently the standard in post-HIFU follow-up in cases of PSA increase with a historically established Phoenix threshold for whole-gland treatment. It provides an anatomical assessment of the prostate gland, enabling recurrences in the residual healthy gland to be identified, but it remains limited in the treated area, as reported by Kaufmann [15]. Post-treatment changes such as fibrosis, hematomas, edema, and chronic inflammation induce signal changes that are often difficult to distinguish from recurring lesions. Furthermore, repeated use of gadolinium-based contrast agents for mpMRIs gives rise to safety concerns, particularly for patients requiring extended follow-up. Gadolinium deposits have been observed in the brain, particularly in the dentate nuclei and globus pallidus, even in patients without impaired renal function [16]. Although the clinical significance of these deposits remains debated, their accumulation following repeated injections has been reported in several publications [16, 17]. In addition, high-frequency (29 MHz) prostate micro-ultrasound is an emerging means of assessing PCa and offers a higher spatial resolution than mpMRI [18]. It would enable real-time detection of clinically significant lesions and could eventually be integrated into selection or follow-up protocols following focal therapy. A prospective study by Klotz et al. (2020) showed that the sensitivity and negative predictive value of micro-ultrasound (94%) was greater than that of mpMRI (89–90%) in the detection of clinically significant PCa (≥ ISUP 2). Its modest specificity (22%) was also equivalent to that of mpMRI. It could therefore guide more accurate targeted biopsies in treated areas, particularly when mpMRI is affected by HIFU sequelae and when recurrence is identified through molecular imaging. It is an interesting alternative: faster, less expensive and potentially more accessible. In this context, the use of PSMA-PET as a perioperative examination for focal treatment using HIFU is conceivable. Unlike MRI, this molecular imaging technique is not greatly affected by post-ablation tissue changes and is based on the metabolic detection of tumor cells via overexpression of PSMA. Its sensitivity for detecting recurrence of PCa is greater, particularly at very low PSA levels [2]. Nevertheless, PSMA-PET also has important limitations that need to be acknowledged. The access to PSMA radiotracers remains uneven, and the higher cost of this technique compared with standard PSA–mpMRI surveillance may restrict its use to selected centres or indications. Radiolabelled PSMA enables precise delineation of intraprostatic cancer lesions, with excellent image contrast [19]. PSMA PET has shown particular value in the detection of high-grade lesions, as there is a correlation between uptake intensity and tumour grade. High intraprostatic PSMA uptake is observed in cancer foci with higher Gleason scores, namely 7 (4 + 3), 8 and 9, and tracer hyperuptake (SUV) increases progressively with the Gleason score, becoming very intense for primary Gleason pattern 5 (SUVmax > 12) [20–22]. Thus, PSMA PET may be more effective in detecting aggressive tumours and may help differentiate low-grade from high-grade lesions. Moreover, no study has reported any alteration in PSMA uptake related to prior biopsy, and PSMA PET even appears to detect more bilateral lesions than MRI alone [23, 24]. This review reinforces the idea that PSMA-PET can play a key role in the early stages of post-HIFU surveillance. Its diagnostic performance, which remains high despite low PSA levels, and its impact on treatment strategy make it particularly appealing. However, its integration into clinical practice remains dependent on indications and protocols becoming standardized and accessibility being improved. Robin et al. confirm that PSMA-PET can guide decisions regarding treatment in the majority of cases with a PSA level of < 0.5 ng/mL [25]. It is essential to determine whether these early changes to treatment actually improve the long-term oncological outcomes when using focal HIFU to treat PCa. Prospective, controlled, large-scale trials are required to establish the exact place of PSMA-PET in standardized follow-up after focal therapy. Conclusion Follow-up after focal treatment of primary PCa is currently centered around PSA-based surveillance, mpMRI and confirmatory biopsies. These methods have limitations, particularly due to a PSA threshold that is defined based on the Phoenix criteria for whole-gland treatment, post-treatment tissue changes and difficulties in interpreting mpMRI. PSMA-PET appears to be a promising form of imaging for post-HIFU surveillance. Its integration into clinical practice could reduce the need for invasive biopsies and improve detection of recurrences. Prospective studies are required to confirm its impact on long-term oncological outcomes. Abbreviations Apparent diffusion coefficient Prostate cancer Contrast-enhanced T1-weighted Dynamic contrast-enhanced Diffusion-weighted imaging Haute Autorité de Santé—French National Health Authority High-intensity focused ultrasound Multiparametric Magnetic Resonance Imaging (mpMRI) Prostate imaging after focal ablation Prostate-specific antigen Prostate-specific membrane antigen Maximum standardized uptake value Peak standardized uptake value T2-weighted imaging Computed Tomography (CT) Positron Emission Tomography (PET) Positron Emission Tomography/Magnetic Resonance Imaging (PET/MRI) Positron Emission Tomography/Computed Tomography (PET/CT) Positive predictive value Negative predictive value Author contributions Jérémy Mercier: Conceptualization, Methodology, Writing - Original DraftLucas Benton : Visualization, Manuscript editingThomas Prudhomme : Writing - Review & Editing, Methodology, Manuscript editingSeverine Lagarde : Visualization, Manuscript editingThibault Cassou-Mounat : Writing - Review & Editing, Methodology, Manuscript editingXavier Game : Writing - Review & Editing, Methodology, Manuscript editingMichel Soulie : Writing - Review & Editing, Methodology, Manuscript editingMatthieu Thoulouzan : Visualization, Manuscript editingAnne Sophie Bajeot : Writing - Review & Editing, Methodology, Manuscript editingMathieu Roumiguie : Conceptualization, Writing - Review & Editing, Methodology, Manuscript editing. Funding This research received no external funding. Data availability No datasets were generated or analysed during the current study. Declarations Conflict of interest The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.