Skip to content
Frontiers
Search

Original Research

Frontiers in Nuclear Medicine ·

Volume 6 - 2026 | https://doi.org/10.3389/fnume.2026.1777541

Prognostic impact of discordant lesions on [ 18 F]FDG and [ 68 Ga]Ga-FAPI-04 PET/CT compared to histological FAP expression in neuroendocrine neoplasms

  • Tim Jedamzik1*corresponding author
  • Aleksander Kosmala1
  • Stefan Kircher2
  • Marieke Heinrich1
  • Wiebke Schlötelburg1
  • Andreas K. Buck1
  • Alexander Meining3,4
  • Rudolf A. Werner3,5,6
  • Alexander Weich3,4
  • Kerstin Michalski1
  1. 1Department of Nuclear Medicine, University Hospital Würzburg, Würzburg, Germany
  2. 2Institute of Pathology, University of Würzburg, Würzburg, Germany
  3. 3NET-Zentrum Würzburg, European Neuroendocrine Tumor Society Center of Excellence (ENETS CoE), University Hospital Würzburg, Würzburg, Germany
  4. 4Internal Medicine II, Department of Gastroenterology, University Hospital Würzburg, Würzburg, Germany
  5. 5Department of Nuclear Medicine, LMU Hospital, Ludwig-Maximilians-University of Munich, Munich, Germany
  6. 6The Russell H Morgan Department of Radiology and Radiological Sciences, Johns Hopkins School of Medicine, Baltimore, MD, United States
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

On dual-tracer positron emission tomography/computed tomography (PET/CT) with 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) and fibroblast activation protein inhibitor ([⁶⁸Ga]Ga-FAPI-04), discordant lesions (FDG+/FAPI-) in aggressive neuroendocrine neoplasms (NENs) are linked to shorter progression-free survival (PFS). This study evaluated the prognostic value of such lesions in comparison to histological fibroblast activation protein (FAP) expression from a clinically obtained biopsy and their impact on PFS. 23 patients with aggressive NENs underwent both [¹⁸F]FDG and [⁶⁸Ga]Ga-FAPI-04 PET/CT as well as biopsy within a short period of time. PET parameters [standardized uptake values (SUV): SUVmax, SUVmean, SUVpeak] were measured, as were tumor volume (TV), and total lesion uptake (TLU = TV × SUVmean). FDG+/FAPI- lesions were identified. FAP expression was assessed immunohistochemically using the immunoreactive score (IRS-FAP). Correlations between PET metrics, IRS-FAP, and FDG+/FAPI- lesions were analyzed. Cox regression and log-rank test were used to evaluate associations with PFS. IRS-FAP correlated significantly with TV ([¹⁸F]FDG: p=0.0165; [⁶⁸Ga]Ga-FAPI-04: p = 0.0181) and TLU ([¹⁸F]FDG: p = 0.0170; [⁶⁸Ga]Ga-FAPI-04: p = 0.0253). There was no significant correlation for IRS-FAP with SUV parameters. FDG+/FAPI- lesions were found in 9/23 patients and associated with significantly shorter PFS (4 vs. 10 months, HR: 3.383, p = 0.0015). [¹⁸F]FDG-TV and [¹⁸F]FDG-TLU correlated with PFS, but IRS-FAP showed no significant association with either PFS or FDG+/FAPI- lesions. FAP expression on immunohistochemistry obtained from a single biopsy site does not predict discordant PET/CT findings. In contrast, the presence of FDG+/FAPI- lesions is a strong prognostic factor for reduced PFS. Thus, dual-tracer PET/CT may offer superior risk stratification compared to single-lesion histological FAP assessment in aggressive NENs.

Keywords

Introduction

Neuroendocrine neoplasms (NENs) refer to a heterogeneous group of tumors with rising incidence. Although they can develop in various parts of the body, they most commonly arise in the gastrointestinal tract and lungs (1). Their behaviour is primarily influenced by tumor grade and differentiation. They are classified into two main categories: highly aggressive, undifferentiated neuroendocrine carcinomas (NECs) and differentiated neuroendocrine tumors (NETs). NETs are further subdivided based on their proliferation index (Ki-67) into low-grade (G1; Ki-67: 1%–2%), intermediate-grade (G2; Ki-67: 3%–20%), and high-grade (G3; Ki-67: >20%). In rare instances, mixed neuroendocrine-non-neuroendocrine neoplasms (MiNENs) may develop, containing both NEN and non-NEN (typically adenoid) components. In most cases, the neuroendocrine component of MiNENs consists of NEC, making these tumors particularly aggressive (2).

Well-differentiated NENs often express high levels of somatostatin receptors (SSTRs), making them targetable with cold somatostatin analogues and suitable for theranostic applications using radiolabeled DOTA-peptides such as [68Ga]Ga/[177Lu]Lu-DOTATATE (3, 4). Conversely, 2-deoxy-2-[¹⁸F]fluoro-D-glucose [(¹⁸F)FDG] positron emission tomography/computed tomography (PET/CT) is primarily used in patients with poorly differentiated NENs that have lost tissue-specific receptor expression. It is also employed in differentiated high-grade NETs to identify dedifferentiated SSTR-negative lesions or aggressive subclones, which are critical for prognosis. Identifying these lesions can help guide re-biopsy or localized treatment strategies as well as prompt adjustments to systemic therapy as [18F]FDG PET has been shown to predict early disease progression and poor prognosis (5). Another potential PET imaging target in NENs is cancer-associated fibroblasts in the tumor stroma, which play a role in tumor invasion, proliferation, metastasis, and angiogenesis in many digestive system malignancies (6–8). Predominantly in fibroblasts of epithelial carcinomas, but also, albeit at lower levels, in tumors with neuroendocrine differentiation the fibroblast activation protein (FAP) is overexpressed and can be visualized with FAP-inhibitor (FAPI) PET imaging (9) (e.g. [68Ga]Ga-FAPI-04 PET) and can also be used for the staging of highly proliferative NENs (10).

Earlier studies have also been able to demonstrate relevant correlations between histological FAP expression and in vivo FAP expression, as represented in [68Ga]Ga-FAPI-04 PET/CT, for example in a pancancer microarray (11) or in non-small cell lung cancer (12). Our previous research additionally demonstrated that the presence of discordant (FDG+/FAPI-) lesions on dual-tracer PET/CT of aggressive NENs is associated with a significantly shorter progression-free survival (PFS) (13).

However, the relationship between histologic FAP expression and in vivo FAP expression in patients with NENs, as well as its connection to the presence of discordant lesions, is still unclear. Therefore, the aim of this study is to investigate the correlation of histological FAP expression compared to visual and quantitative parameters on dual-tracer [68Ga]Ga-FAPI-04 and [18F]FDG PET/CT as well as their prognostic value for aggressive NENs.

Material and methods

Patient cohort

This is a retrospective single-center study. Patients with aggressive NENs with intermediate or high proliferative activity, reflecting tumors with increased biological aggressiveness, including G2 NET, G3 NET, NEC, and MiNEN, who had received a [68Ga]Ga-FAPI-04 PET/CT, a [18F]FDG PET/CT, and a biopsy for histological examination were included. The time interval between the two PET scans had to be less than 3 months, and the time between the scan and the biopsy was allowed to be a maximum of 12 months, with no intermittent tumor-specific therapy. The order of the PET scans and biopsy was of no consequence for inclusion. If no follow-up data was available, the patient was excluded. Parts of this cohort were already included in previous studies (13, 14), however, four additional patients with G2-NETs were included in the present analysis to expand the cohort beyond the previously studied high-grade subgroup. Moreover, histological FAP expression using immunohistochemistry was assessed for the first time. Dual-tracer PET/CT was performed with clinically justified indication to diagnose possible tumor heterogeneity and aggressive lesions. Written informed consent was obtained from each patient for both PET scans. The need for further approval was waived by the local institutional review board because of the retrospective nature of the study (study 20230721 01).

Image acquisition and analysis

The synthesis and labelling of [18F]FDG and [68Ga]Ga-FAPI-04 were carried out as described in previous studies (15, 16). The average injected activity was 234 ± 38 MBq (range: 154–316 MBq) for [18F]FDG and 145 ± 12 MBq (range: 114–156 MBq) for [68Ga]Ga-FAPI-04. One hour after the injection of the respective tracer, patients underwent whole-body PET scans (from the vertex to the mid-thigh) with a scan duration of 2 min per bed position. A Biograph mCT device (64 or 128; Siemens Healthineers) was used, either with or without intravenous contrast material [with activated automatic tube current modulation; reference mAs set at 35 mAs for low-dose scans and 160 mAs for full-dose scans; tube voltage, 120 keV/100 keV [mCT 64/mCT 128]; pitch, 1.4/0.8 [mCT 64/mCT 128]; collimation, 64/128 × 0.6 mm; rotation time, 0.5 s; reconstructed axial slice thickness, 3.0–5.0 mm). PET images were reconstructed using standard parameters [3-dimensional mode; matrix, 200 × 200; iterations, 3; subsets, 24 [mCT 64]/21 [mCT 128]; gaussian filtering, 2.0 mm. Time-of-flight for PET reconstruction was used in scans performed with mCT 128, whereas this was not available on the Biograph 64].

All PET/CT scans were quantitatively analyzed by one reader using the open-source software Fiji (17) with the Beth Israel plugin (18), applying a 40% lesion-specific threshold as previously published by our group (16, 19–21). The analysis included standardized uptake values (SUV): SUVmax, SUVpeak, and SUVmean as well as tumor volume (TV) and total lesion uptake (TLU = TV × SUVmean) for both radiotracers. Given the high number of lesions in some patients, initial visual screening for potential discordant FDG-positive, FAPI-negative (FDG+/FAPI-) lesions was performed by identifying focal lesions showing uptake higher than liver background on [18F]FDG PET and absent or negligible uptake compared to surrounding tissue on [68Ga]Ga-FAPI-04 PET. To support the visual classification, semi-quantitative analysis was subsequently performed. FDG+/FAPI- lesions were required to demonstrate increased [18F]FDG uptake relative to blood pool activity, defined as a tumor-to-background ratio (TBR), calculated as SUVpeak of the lesion divided by SUVmean of the descending aorta, greater than 1.0, while showing no corresponding focal uptake on [68Ga]Ga-FAPI-04 PET. This threshold was chosen to ensure that [18F]FDG uptake exceeded physiologic blood pool activity, thereby providing a pragmatic semi-quantitative confirmation of visually increased metabolic activity. All FDG+/FAPI- discordant lesions were confirmed by consensus of a board-certified radiologist and a board-certified nuclear medicine specialist (AK and KM), both with over 5 years of experience in PET/CT interpretation. Both PET readers were blinded to the histological evaluation. Clinical follow-up data were used for PFS determination independent of the imaging reads.

Histological FAP-expression

The biopsy for histology was indicated and performed according to clinical practice guidelines. The sample obtained was examined by a board-certified pathologist who specialises on NENs and was blinded to the image evaluation. The histological assessment of FAP expression was performed using immunohistochemistry. Therefore, formalin-fixed, paraffin-embedded tissue sections were cut, mounted on slides, and incubated with a specific monoclonal antibody against FAP. Detection was carried out using standard immunohistochemical detection systems, allowing for microscopic visualization. The immunoreactive score (IRS) was used for reproducible evaluation and quantification of immunohistochemical FAP expression (22–24). The IRS consists of two subscores: the proportion of stained cells: 0 for 0%, 1 for 1%–10%, 2 for 10%–50%, 3 for 50%–80% and 4 for more than 80%. The second subscore is the intensity of the staining: 1 for weak intensity, 2 for medium intensity and 3 for strong intensity. The IRS is calculated by multiplying the two subscores and therefore has a range of 0–12.

Statistical analysis

Statistical analysis was done using GraphPad Prism version 10.4.1 (GraphPad Software). Descriptive data are expressed using mean and corresponding standard deviation. Survival data were analyzed by Kaplan–Meier curves and log-rank comparison as well as univariate Cox proportional-hazards regression and a limited multivariate Cox proportional-hazards regression using the presence of FDG+/FAPI- lesions and age. Spearman correlations were used between IRS-FAP and image parameters. Given the exploratory nature of this study, no formal correction for multiple comparisons was applied. PFS was defined both radiographically by a board-certified radiologist using RECIST 1.1 (25) and clinically at evidence of disease progression (by the treating physicians) or death by any cause. The day of [68Ga]Ga-FAPI-04 PET was used as the reference point for calculation of PFS. To assess the robustness of the findings, a sensitivity analysis was performed excluding patients with biopsy-to-imaging intervals exceeding 100 days.

Results

Patient cohort

We included 23 patients (13 male, 10 female) with a mean age of 61 ± 10.0 years (range: 38–81 years) who underwent dual-tracer PET/CT with [18F]FDG and [68Ga]Ga-FAPI-04 between July 2020 and March 2023. The mean interval between the two PET scans was 10 ± 13 days (range: 1–63 days). The mean time interval between the first PET scan and biopsy was 44 ± 69 days (range: 1–343 days), and in 18/23 cases the biopsy was performed prior to imaging. Only two patients had a time interval between imaging and biopsy exceeding 60 days.

Histologically, most cases were high-grade NENs, including five patients with G3-NETs (22%) and nine patients with NECs (39%). Five patients had MiNEN (22%). Biopsies were most frequently obtained from the liver (n = 11; 48%) or the primary tumor site (n = 6; 26%), and less frequently from a lymph node (n = 2; 9%) or another metastatic site (n = 4; 17%).

The primary tumor was most commonly located in the gastrointestinal tract (n = 14; 61%) or remained unknown [Cancer of unknown primary (CUP); n = 7; 30%]. Detailed patient characteristics as well as the biopsy locations are provided in Table 1.

Table 1. Patients characteristics.

Patient no.SexAgeHistologyPrimary tumorKi-67Site of biopsyDays between biopsy and first PET scan
1F59MiNENColon80Liver13
2F63MiNENColon85Primary tumor1
3M81NECCUP80Bone10
4M56G3Pancreas70Liver24
5F64G2Small intestine7Liver29
6F38G3Pancreas60Liver36
7F75G3Small intestine20Liver25
8M57NECCUP90Esophagus343
9M42G3Pancreas40Primary tumor14
10M53MiNENStomach90Primary tumor59
11F49NECBreast90Primary tumor18
12M64NECBladder90Primary tumor29
13F58G2CUP10Lymphnode31
14F66NECCUP70Pancreas56
15F60MiNENColon90Liver3
16M56NECPancreas14Liver28
17M53G3Pancreas40Liver13
18M68NECStomach90Primary tumor11
19M66G2Small intestine5Liver39
20M68MiNENPancreas80Liver41
21M69NECCUP80Bone19
22F67NECCUP80Lymphnode40
23M63G2CUP7Liver129

MiNEN, mixed neuroendocrine-non-neuroendocrine neoplasm; NEC, neuroendocrine carcinoma; CUP, cancer of unknown primary.

FDG+/FAPI- discordant lesions

Metastases were found in most patients (n = 21; 91%), most frequently in the liver (n = 15; 65%) and in lymph nodes (n = 15; 65%), followed by bone metastases (n = 8; 35%), less frequently in the lungs (n = 3, 13%) or in the peritoneum (n = 2; 9%). FDG+/FAPI- discordant lesions were found in 9/23 patients, most frequently in patients with NEC (n = 5), followed by patients with G3-NET (n = 2), patients with G2-NET (n = 1) and patients with MiNEN (n = 1). These were located in the liver (n = 4), in lymph nodes (n = 4) and in the skeleton (n = 3). Among the nine patients with FDG+/FAPI- lesions, biopsies were obtained from the same organ in two cases (both liver). However, in one of the two cases, it could be determined that the discordant lesion and the biopsied lesion were not the same, in the other case, this was not possible. In the remaining seven patients, biopsies were obtained from anatomically distinct sites and therefore did not spatially correspond to the discordant lesions.

Correlation with FAP expression

The cohort showed a median IRS-FAP of 4, including 4 patients with IRS-FAP of 0, 2 patients with IRS-FAP of 1, 2 patients with IRS-FAP of 2, 6 patients with IRS-FAP of 4, 6 patients with IRS-FAP of 6, 2 patients with IRS-FAP of 9 and 1 patient with IRS-FAP of 12. Figure 1 depicts an exemplary patient with FDG+/FAPI- liver lesions and the corresponding histological FAP expression. For illustrative purposes, a case with a high FAP expression (IRS-FAP = 12) was selected. FAP expression appears as a light brown staining, while cell nuclei are counterstained in blue with hematoxylin. Notably, FAP expression is predominantly cytoplasmatic, with at most only minimal membranous staining.

Maximum-intensity projection (MIP), PET and fused images of [<sup>18</sup>F]FDG and [<sup>68</sup>Ga]Ga-FAPI-04 PET/CT of a patient with a G3 pancreatic NET with multiple metastases in the liver. The time interval between the scans was 14 days, and 13 days between [<sup>68</sup>Ga]Ga-FAPI-04 PET and biopsy. An exemplary FDG+/FAPI- lesion in the liver is marked with a green arrow. In addition, the histological fibroblast activation protein (FAP) expression is shown in the lower part of the image. The biopsy site is marked with blue arrows.
Figure 1. Maximum-intensity projection (MIP), PET and fused images of [18F]FDG and [68Ga]Ga-FAPI-04 PET/CT of a patient with a G3 pancreatic NET with multiple metastases in the liver. The time interval between the scans was 14 days, and 13 days between [68Ga]Ga-FAPI-04 PET and biopsy. An exemplary FDG+/FAPI- lesion in the liver is marked with a green arrow. In addition, the histological fibroblast activation protein (FAP) expression is shown in the lower part of the image. The biopsy site is marked with blue arrows.

The quantitative analysis showed a statistically significant correlation between IRS-FAP with TV ([18F]FDG: p = 0.0165; [68Ga]Ga-FAPI-04: p = 0.0181) and TLU ([18F]FDG: p = 0.0170; [68Ga]Ga-FAPI-04: p = 0.0253) for both radiotracers. The correlation plots are shown in Figure 2.

Correlation plots of immunoreactive score (IRS) with tumor volume (TV) and total lesion uptake (TLU) for both [<sup>18</sup>F]FDG and [<sup>68</sup>Ga]Ga-FAPI-04 PET/CT. For better visualization, TV and TLU were log-transformed in this figure. The correlation coefficients and respective *p*-values are depicted in Table 2.
Figure 2. Correlation plots of immunoreactive score (IRS) with tumor volume (TV) and total lesion uptake (TLU) for both [18F]FDG and [68Ga]Ga-FAPI-04 PET/CT. For better visualization, TV and TLU were log-transformed in this figure. The correlation coefficients and respective p-values are depicted in Table 2.

Table 2. Correlation of immunoreactive score of fibroblast activation protein (FAP) expression with imaging and other histological parameters.

ParameterSpearman r95% CIp (two-tailed)
[<sup>18</sup>F]FDG SUVmax0.1903−0.2530–0.56750.3846
[<sup>18</sup>F]FDG SUVpeak0.168−0.2744–0.55170.4435
[<sup>18</sup>F]FDG SUVmean0.1688−0.2736–0.55230.4412
[<sup>18</sup>F]FDG TV0.49440.09045–0.75870.0165
[<sup>18</sup>F]FDG TLU0.49240.08780–0.75760.017
[<sup>68</sup>Ga]Ga-FAPI-04 SUVmax0.1998−0.2437–0.57420.3607
[<sup>68</sup>Ga]Ga-FAPI-04 SUVpeak0.1831−0.2599–0.56250.4029
[<sup>68</sup>Ga]Ga-FAPI-04 SUVmean0.4047−0.02192–0.70670.0554
[<sup>68</sup>Ga]Ga-FAPI-04 TV0.48840.08252–0.75530.0181
[<sup>68</sup>Ga]Ga-FAPI-04 TLU0.46520.05262–0.74210.0253
Histology (G2, G3, NEC, MiNEN)0.02056−0.4059–0.43960.9258
FDG+/FAPI- discordance0.1646−0.2776–0.54920.4530

TV, tumor volume; TLU, total lesion uptake (SUVmean × TV); MiNEN, mixed neuroendocrine-non-neuroendocrine neoplasm; NEC, neuroendocrine carcinoma.

There was no statistically significant correlation between IRS-FAP and SUVmax, SUVpeak or SUVmean for either [18F]FDG or [68Ga]Ga-FAPI-04. However, the correlation of IRS-FAP and SUVmean of [68Ga]Ga-FAPI-04 showed a clear trend that only slightly failed to be significant (p = 0.0554). The correlation parameters of IRS-FAP are also shown in detail in Table 2. There was no statistically significant correlation between IRS-FAP and the presence of FDG+/FAPI- discordant lesions (p = 0.4530) as also depicted in Figure 3.

Immunoreactive score of fibroblast activation protein (FAP) expression depending on the presence of FDG+/FAPI-discordant lesions. There is no statistically significant difference (*p* = 0.4681).
Figure 3. Immunoreactive score of fibroblast activation protein (FAP) expression depending on the presence of FDG+/FAPI-discordant lesions. There is no statistically significant difference (p = 0.4681).

Prognostic value for PFS

The median PFS of this patient cohort was 6 months. PFS was significantly shorter in patients with FDG+/FAPI- discordant lesions (4 months) compared with those without the presence of FDG+/FAPI- lesions (10 months; HR: 3.383, 95% CI: 1.041–10.99, p = 0.0015) in univariate Cox regression as shown in Figure 4.

Kaplan–Meier curves for progression-free survival (PFS) with and without FDG+/FAPI- lesions. Dashed lines represent the respective confidence intervals. The difference in the log-rank test is statistically significant (*p* = 0.0015) and is confirmed by univariate Cox regression (HR: 3.383, 95% CI: 1.041-10.99, *p* = 0.0015).
Figure 4. Kaplan–Meier curves for progression-free survival (PFS) with and without FDG+/FAPI- lesions. Dashed lines represent the respective confidence intervals. The difference in the log-rank test is statistically significant (p = 0.0015) and is confirmed by univariate Cox regression (HR: 3.383, 95% CI: 1.041-10.99, p = 0.0015).

In multivariate Cox regression including age and FDG+/FAPI- discordance, the presence of discordant lesions remained independently associated with shorter PFS (HR: 8.269, 95% CI: 2.227–40.70, p = 0.0014), whereas age also remained significant (HR: 1.073, 95% CI: 1.016–1.141, p = 0.0105).

The IRS-FAP showed no statistically significant influence on PFS (HR: 1.06, 95% CI: 0.9221–1.206, p = 0.3875). Of the PET parameters, only [18F]FDG -TV (HR: 1.003, 95% CI: 1.001–1.006, p = 0.0119) and [18F]FDG -TLU (HR: 1.000, 95% CI: 0.000–1.001, p = 0.008) had a statistically significant influence on PFS, whereas this was not significant for [68Ga]Ga-FAPI-04-TV (HR: 1.002, 95% CI: 0.9993–1.003, p = 0.1235) and [68Ga]Ga-FAPI-04-TLU (HR: 1.000, 95% CI: 0.9998–1.000, p = 0.6013) as well as SUVmax, SUVpeak and SUVmean for both radiotracers (Table 3).

Table 3. Impact of imaging-derived and clinical parameters on progression-free-survival.

ParameterHazard ratio95% CIp
IRS-FAP1.060.9221–1.2060.3875
Ki67%1.0050.9905–1.0210.5035
Histology G2, G3, NEC, MiNEN1.1150.7106–1.7230.6254
Age1.0681.012–1.1380.0255
[<sup>18</sup>F]FDG SUVmax1.0260.9851–1.0640.1938
[<sup>18</sup>F]FDG SUVpeak1.050.9732–1.1250.1783
[<sup>18</sup>F]FDG SUVmean1.030.9766–1.0790.2387
[<sup>18</sup>F]FDG TV1.0031.001–1.0060.0119
[<sup>18</sup>F]FDG TLU1.0001.000–1.0010.008
[<sup>68</sup>Ga]Ga-FAPI-04 SUVmax1.0050.9619–1.0430.8227
[<sup>68</sup>Ga]Ga-FAPI-04 SUVpeak1.0130.9125–1.1030.7919
[<sup>68</sup>Ga]Ga-FAPI-04 SUVmean1.0080.9438–1.0680.7866
[<sup>68</sup>Ga]Ga-FAPI-04 TV1.0020.9993–1.0030.1235
[<sup>68</sup>Ga]Ga-FAPI-04 TLU1.0000.9998–1.0000.6013

IRS-FAP, immunoreactive score of fibroblast activation protein (FAP) expression; TV, tumor volume; TLU, total lesion uptake (SUVmean * TV); MiNEN, mixed neuroendocrine-non-neuroendocrine neoplasm; NEC, neuroendocrine carcinoma.

Sensitivity analysis excluding the two patients with biopsy-to-imaging intervals exceeding 100 days yielded consistent results, with unchanged statistical significance for the association of FDG+/FAPI- discordant lesions, age, [18F]FDG-TV, and [18F]FDG-TLU with PFS, as well as for the correlations between IRS-FAP and PET-derived tumor burden parameters. The detailed results of the sensitivity analysis can be found in the supplement.

Discussion

This study was—to the best of our knowledge—the first to examine the relationship between histological FAP expression and in vivo PET imaging findings using dual-tracer [68Ga]Ga-FAPI-04 and [18F]FDG PET/CT in patients with aggressive NENs. In addition, we evaluated the prognostic value of FAP expression and the presence of discordant FDG+/FAPI- lesions with respect to PFS. Our findings provide further insights into tumor biology and the utility of PET-based biomarkers in this highly heterogeneous tumor entity.

Consistent with our previous work by Michalski et al. (13), we observed that FDG+/FAPI- discordant lesions were present in a substantial proportion of patients with aggressive NENs, particularly in NEC. This may be an indication of increasing metabolic heterogeneity and stromal remodeling with higher grade disease (26). The presence of FDG+/FAPI- lesions was associated with a significantly shorter PFS (4 vs. 10 months) and remained an independent prognostic factor after adjustment for age, highlighting its potential as a negative prognostic imaging biomarker. This has already been shown for dual-tracer PET with [18F]FDG and somatostatin receptor (SSTR)-directed ligands in patients with NENs by Chan et al. (27, 28) as well as in patients with prostate cancer using dual-tracer PET with [18F]FDG and prostate-specific membrane antigen (PSMA) ligands (29–31). This reinforces the clinical relevance of identifying FDG+/FAPI- lesions, which may reflect tumor subclones with a more aggressive and dedifferentiated phenotype and thus low FAP expression, but high glucose metabolism. However, this interpretation remains hypothesis-generating and requires confirmation in larger prospective cohorts.

Interestingly, in contrast to earlier studies in other tumor entities (11, 23), semi-quantitative [68Ga]Ga-FAPI-04 PET parameters did not correlate significantly with histological FAP expression (IRS-FAP) in NENs. This may be because voxel-level readouts such as SUVmax and SUVpeak capture only the hottest voxels, which may reflect heterogeneous outlier clusters unrepresentative of true stromal density, whereas the global assessment of tracer uptake (TV, TLU) averages out these outliers and better parallels the overall FAP expression. The near-significant trend for [68Ga]Ga-FAPI-04 SUVmean (p = 0.0554) further supports this. However, the presence of FDG+/FAPI- lesions did not correlate with IRS-FAP either. This could mean that although [68Ga]Ga-FAPI-04 PET reflects stromal activity at a global level, discordant lesions may be driven more by clonal heterogeneity or temporal evolution of the tumor than by the average stromal activation assessed histologically in a single biopsy. However, in addition to possible sampling limitations inherent to immunohistochemistry (32), the spatial mismatch between biopsy sites and imaging-defined discordant lesions represents an important methodological limitation. Among patients with FDG+/FAPI− discordant lesions, biopsies were obtained from the same organ in only a minority of cases, and even in those cases, lesion-level correspondence could not be confirmed. In most patients, biopsies originated from anatomically distinct tumor sites, precluding direct lesion-by-lesion comparison between histological FAP expression and PET imaging findings. Given the known intrapatient heterogeneity of aggressive NENs, this mismatch may substantially limit the biological interpretability of correlations between immunohistochemistry and imaging-derived parameters. Consequently, histological FAP expression from a single biopsy site may not adequately reflect the spatial heterogeneity of stromal activation across the whole-body tumor burden assessed by PET/CT. Future prospective studies with spatially matched biopsies are therefore essential to validate the tissue–imaging relationship at the lesion level, especially in view of the increasing interest in stroma-targeted therapies and theranostic approaches with FAP-directed radioligands (10, 33).

Regarding prognosis, our data showed that, in addition to the presence of discordant lesions and the expected patient age, [18F]FDG-derived tumor burden metrics ([18F]FDG-TV and [18F]FDG-TLU) were significantly associated with shorter PFS, whereas histological FAP expression (IRS-FAP) was not. These results are consistent with previous reports highlighting the prognostic value of [18F]FDG uptake in aggressive NENs (5) and suggest that tumor metabolism remains a more reliable predictor of outcome than stromal activity in this patient population. It also indicates that (dual-tracer) PET/CT with [18F]FDG and [68Ga]Ga-FAPI-04 as a whole-body diagnostic tool provides deeper insights into tumor phenotype in all lesions and may be superior for estimating prognosis in patients with aggressive NENs. Based on this, individualized treatment decisions may be guided by the identification of high-risk lesions, whereas immunohistochemistry reflects only local tumor biology. However, the prediction of [68Ga]Ga-FAPI-04-TV for PFS was not significant in this study (p = 0.1235), whereas it was still significant in the study by Michalski et al., which used a similar cohort. This highlights the need for further studies in larger cohorts, ideally in a prospective setting, to enable more definitive conclusions.

Our findings must be interpreted in light of several limitations. First, the retrospective, single-center design and modest sample size limit the generalizability and statistical power of our results. Nevertheless, aggressive NENs represent a rare tumor entity, and the combination of histological FAP assessment with both [18F]FDG and [68Ga]Ga-FAPI-04 PET/CT within a clinically narrow time window is logistically demanding, thus, a cohort of 23 patients still represents a comparatively robust sample for such a multimodal analysis. Second, the temporal intervals between diagnostic procedures warrant consideration. The interval between the two PET scans was short for nearly all patients, while the time between PET imaging and biopsy was longer in a few cases. Importantly, these extended intervals were isolated exceptions, with only two cases exceeding 60 days. Sensitivity analysis excluding these patients yielded consistent results, supporting the robustness of the findings and confirming that inclusion of these patients did not materially influence the study conclusions. Moreover, even in aggressive NENs, substantial biological changes or major shifts in tumor phenotype are not typically expected within short time periods of days to a few weeks, further mitigating the potential impact of these intervals. Overall, the temporal alignment achieved in this study still reflects tightly controlled conditions for this rare disease.

Furthermore, although discordant lesions were assessed by two experienced readers in consensus, formal interobserver variability was not evaluated. Therefore, some degree of observer dependence cannot be excluded. However, the additional use of semi-quantitative criteria reduces subjectivity and supports the robustness of lesion classification. It should also be noted that no formal correction for multiple comparisons was applied. Given the exploratory nature of this study and the limited sample size inherent to this rare tumor entity, correction methods such as Bonferroni may be overly conservative and increase the risk of type II error, potentially obscuring biologically relevant associations. Nevertheless, the risk of type I error cannot be excluded, and therefore the observed associations, particularly those with borderline statistical significance, should be interpreted with caution and considered exploratory. Importantly, the association between FDG+/FAPI- discordant lesions and shorter PFS remains highly significant and supports its role as a prognostically relevant imaging biomarker.

Another limitation relates to the lack of full formal harmonization of the PET/CT acquisition protocols. Two different PET/CT systems were used, and examinations were performed with or without intravenous CT contrast depending on clinical indication. However, the reconstruction parameters were nearly identical across scanners, with the only systematic difference being the absence of time-of-flight capability in Biograph 64. Given that all quantitative PET metrics were derived from harmonized reconstruction settings, and that the use of CT contrast does not substantially affect PET tracer uptake, lesion conspicuity, or the identification of discordant lesions, the impact of these protocol differences on the study results is expected to be minimal.

Finally, our cohort partially overlapped with previous work, however, several patients from the earlier studies did not meet the specific criteria for the present analysis, and additional patients were included. Furthermore, the addition of histopathological analysis in this study allowed novel insights into FAP biology in aggressive NENs and therefore does not diminish the clinical relevance of the findings presented.

In summary, histologic FAP expression, as measured by IRS-FAP, correlates significantly with metabolic TV and TLU for both radiotracers. However, it shows neither a significant association with the presence of FDG+/FAPI- lesions nor a significant prognostic impact on PFS. Conversely, our findings confirm that the presence of FDG+/FAPI- discordant lesions on [18F]FDG and [68Ga]Ga-FAPI-04 PET/CT is associated with a significantly shorter PFS in patients with aggressive NENs. Dual-tracer PET/CT as a whole-body diagnostic could represent a superior method for risk stratification and therapeutic decision-making in these patients than immunohistochemical examination of FAP expression from a single tumor site. However, prospective studies with larger sample sizes and spatially matched biopsies are needed to consolidate the results and guide the use of dual-tracer PET/CT in precision oncology.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnume.2026.1777541/full#supplementary-material

Supplementary material Supplementary material at the source

Supplementary material Supplementary material at the source

Footnotes

Edited by: Nikolaos Papathanasiou, General University Hospital of Patras, Greece

Reviewed by: Paulina Cegla, Greater Poland Cancer Center (GPCC), Poland

Zhehao Lyu, First Affiliated Hospital of Harbin Medical University, China

Statements

Conflict of interest

AB and KM have received speaker honoraria from Novartis. RW has received speaker honoraria from Novartis and reports advisory board work for Novartis and Bayer. All other authors declare they have no financial interests.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Interdisciplinary Center of Clinical Research (IZKF), University Hospital of Wuerzburg (grant Z-3BC/08 to KM).

Data availability

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics approval

The studies involving humans were approved by Ethics Committee of Wuerzburg (waiver number 20230721 01). The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants' legal guardians/next of kin because of the retrospective character of the study. Written informed consent was not obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article because no conclusions about the patient can be drawn from the published PET/CT images. Furthermore written informed consent for diagnostic procedures was obtained. For research purposes, we acted in accordance with Bavarian Hospital Law [Art. 27 Abs. 4 Bayerisches Krankenhausgesetz (BayKrG)], which waives the need for additional informed consent in terms of retrospective studies.

Author contributions

TJ: Data curation, Investigation, Methodology, Software, Validation, Visualization, Writing – original draft, Writing – review & editing. AK: Investigation, Software, Writing – original draft, Writing – review & editing. SK: Data curation, Formal analysis, Writing – original draft, Writing – review & editing. MH: Writing – original draft, Writing – review & editing. WS: Writing – original draft, Writing – review & editing. AB: Funding acquisition, Resources, Writing – original draft, Writing – review & editing. AM: Writing – original draft, Writing – review & editing. RW: Writing – original draft, Writing – review & editing. AW: Conceptualization, Formal analysis, Project administration, Supervision, Writing – original draft, Writing – review & editing. KM: Conceptualization, Data curation, Funding acquisition, Methodology, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review & editing.

References

  1. 1. Sultana Q, Kar J, Verma A, Sanghvi S, Kaka N, Patel N A comprehensive review on neuroendocrine neoplasms: presentation, pathophysiology and management. J Clin Med. (2023) 12 (15): 5138. doi:10.3390/jcm12155138
  2. 2. Rindi G, Mete O, Uccella S, Basturk O, La Rosa S, Brosens LAA Overview of the 2022 WHO classification of neuroendocrine neoplasms. Endocr Pathol. (2022) 33 (1): 115 – 54. doi:10.1007/s12022-022-09708-2
  3. 3. Al-Nahhas A. Nuclear medicine imaging of neuroendocrine tumours. Clin Med (Lond). (2012) 12 (4): 377 – 80. doi:10.7861/clinmedicine.12-4-377
  4. 4. Fortunati E, Argalia G, Zanoni L, Fanti S, Ambrosini V. New PET radiotracers for the imaging of neuroendocrine neoplasms. Curr Treat Options Oncol. (2022) 23 (5): 703 – 20. doi:10.1007/s11864-022-00967-z
  5. 5. Binderup T, Knigge U, Johnbeck CB, Loft A, Berthelsen AK, Oturai P (18)F-FDG PET is superior to WHO grading as a prognostic tool in neuroendocrine neoplasms and useful in guiding PRRT: a prospective 10-year follow-up study. J Nucl Med. (2021) 62 (6): 808 – 15. doi:10.2967/jnumed.120.244798
  6. 6. Huang L, Xu A-M, Liu S, Liu W, Li T-J. Cancer-associated fibroblasts in digestive tumors. World J Gastroenterol. (2014) 20 (47): 17804 – 18. doi:10.3748/wjg.v20.i47.17804
  7. 7. Tao L, Huang G, Song H, Chen Y, Chen L. Cancer associated fibroblasts: an essential role in the tumor microenvironment. Oncol Lett. (2017) 14 (3): 2611 – 20. doi:10.3892/ol.2017.6497
  8. 8. Wang Y, Liang Y, Xu H, Zhang X, Mao T, Cui J Single-cell analysis of pancreatic ductal adenocarcinoma identifies a novel fibroblast subtype associated with poor prognosis but better immunotherapy response. Cell Discov. (2021) 7 (1): 36. doi:10.1038/s41421-021-00271-4
  9. 9. Siripongsatian D, Promteangtrong C, Kunawudhi A, Kiatkittikul P, Boonkawin N, Chinnanthachai C Comparisons of quantitative parameters of Ga-68-labelled fibroblast activating protein inhibitor (FAPI) PET/CT and [(18)F]F-FDG PET/CT in patients with liver malignancies. Mol Imaging Biol. (2022) 24 (5): 818 – 29. doi:10.1007/s11307-022-01732-2
  10. 10. Mori Y, Dendl K, Cardinale J, Kratochwil C, Giesel FL, Haberkorn U. FAPI PET: fibroblast activation protein inhibitor use in oncologic and nononcologic disease. Radiology. (2023) 306 (2): e220749. doi:10.1148/radiol.220749
  11. 11. Mona CE, Benz MR, Hikmat F, Grogan TR, Lueckerath K, Razmaria A Correlation of (68)Ga-FAPi-46 PET biodistribution with FAP expression by immunohistochemistry in patients with solid cancers: interim analysis of a prospective translational exploratory study. J Nucl Med. (2022) 63 (7): 1021 – 6. doi:10.2967/jnumed.121.262426
  12. 12. Li C, Chen Q, Tian Y, Chen J, Xu K, Xiao Z (68)Ga-FAPI-04 PET/CT in non-small cell lung cancer: accurate evaluation of lymph node metastasis and correlation with fibroblast activation protein expression. J Nucl Med. (2024) 65 (4): 527 – 32. doi:10.2967/jnumed.123.266806
  13. 13. Michalski K, Kosmala A, Hartrampf PE, Heinrich M, Serfling SE, Schlötelburg W [(18)F]FDG and [(68)Ga]Ga-FAPI-04 imaging for outcome prediction in patients with high-grade neuroendocrine neoplasms. J Nucl Med. (2024) 65 (12): 1899 – 903. doi:10.2967/jnumed.124.268288
  14. 14. Kosmala A, Serfling SE, Schlötelburg W, Lindner T, Michalski K, Schirbel A Impact of 68Ga-FAPI-04 PET/CT on staging and therapeutic management in patients with digestive system tumors. Clin Nucl Med. (2023) 48 (1): 35 – 42. doi:10.1097/RLU.0000000000004480
  15. 15. Lindner T, Loktev A, Altmann A, Giesel F, Kratochwil C, Debus J Development of quinoline-based theranostic ligands for the targeting of fibroblast activation protein. J Nucl Med. (2018) 59 (9): 1415 – 22. doi:10.2967/jnumed.118.210443
  16. 16. Kosmala A, Serfling SE, Dreher N, Lindner T, Schirbel A, Lapa C Associations between normal organs and tumor burden in patients imaged with fibroblast activation protein inhibitor-directed positron emission tomography. Cancers (Basel). (2022) 14 (11): 2609. doi:10.3390/cancers14112609
  17. 17. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T Fiji: an open-source platform for biological-image analysis. Nat Methods. (2012) 9 (7): 676 – 82. doi:10.1038/nmeth.2019
  18. 18. Kanoun S, Tal I, Vrigneaud J. Beth Israel plugin for FIJI: a free and open source software tool for PET/CT processing in research. Eur J Nucl Med Mol. (2016) I (43): S389 – 90.
  19. 19. Werner RA, Hänscheid H, Leal JP, Javadi MS, Higuchi T, Lodge MA Impact of tumor burden on quantitative [(68)Ga] DOTATOC biodistribution. Mol Imaging Biol. (2019) 21 (4): 790 – 8. doi:10.1007/s11307-018-1293-9
  20. 20. Werner RA, Bundschuh RA, Bundschuh L, Lapa C, Yin Y, Javadi MS Semiquantitative parameters in PSMA-targeted PET imaging with [(18)F]DCFPyL: impact of tumor burden on normal organ uptake. Mol Imaging Biol. (2020) 22 (1): 190 – 7. doi:10.1007/s11307-019-01375-w
  21. 21. Serfling SE, Lapa C, Dreher N, Hartrampf PE, Rowe SP, Higuchi T Impact of tumor burden on normal organ distribution in patients imaged with CXCR4-targeted [(68)Ga]Ga-PentixaFor PET/CT. Mol Imaging Biol. (2022) 24 (4): 659 – 65. doi:10.1007/s11307-022-01717-1
  22. 22. Remmele W, Stegner HE. Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Pathologe. (1987) 8 (3): 138 – 40.
  23. 23. Spektor A-M, Gutjahr E, Lang M, Glatting FM, Hackert T, Pausch T Immunohistochemical FAP expression reflects (68)Ga-FAPI PET imaging properties of low- and high-grade intraductal papillary mucinous neoplasms and pancreatic ductal adenocarcinoma. J Nucl Med. (2024) 65 (1): 52 – 8. doi:10.2967/jnumed.123.266393
  24. 24. Meyerholz DK, Beck AP. Principles and approaches for reproducible scoring of tissue stains in research. Lab Invest. (2018) 98 (7): 844 – 55. doi:10.1038/s41374-018-0057-0
  25. 25. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. (2009) 45 (2): 228 – 47. doi:10.1016/j.ejca.2008.10.026
  26. 26. Luciani MG, Seok J, Sayeed A, Champion S, Goodson WH, Jeffrey SS Distinctive responsiveness to stromal signaling accompanies histologic grade programming of cancer cells. PLoS One. (2011) 6 (5): e20016. doi:10.1371/journal.pone.0020016
  27. 27. Chan DL, Pavlakis N, Schembri GP, Bernard EJ, Hsiao E, Hayes A Dual somatostatin receptor/FDG PET/CT imaging in metastatic neuroendocrine tumours: proposal for a novel grading scheme with prognostic significance. Theranostics. (2017) 7 (5): 1149 – 58. doi:10.7150/thno.18068
  28. 28. Chan DL, Hayes AR, Karfis I, Conner A, Mileva M, Bernard E [(18)F]FDG PET/CT-avid discordant volume as a biomarker in patients with gastroenteropancreatic neuroendocrine neoplasms: a multicenter study. J Nucl Med. (2024) 65 (2): 185 – 91. doi:10.2967/jnumed.123.266346
  29. 29. Hartrampf PE, Lapa C, Serfling SE, Buck AK, Seitz AK, Meyer PT Development of discordant hypermetabolic prostate cancer lesions in the course of [(177)Lu]PSMA radioligand therapy and their possible influence on patient outcome. Cancers (Basel). (2021) 13 (17): 4270. doi:10.3390/cancers13174270
  30. 30. Michalski K, Ruf J, Goetz C, Seitz AK, Buck AK, Lapa C Prognostic implications of dual tracer PET/CT: PSMA ligand and [(18)F]FDG PET/CT in patients undergoing [(177)Lu]PSMA radioligand therapy. Eur J Nucl Med Mol Imaging. (2021) 48 (6): 2024 – 30. doi:10.1007/s00259-020-05160-8
  31. 31. Rosar F, Ribbat K, Ries M, Linxweiler J, Bartholomä M, Maus S Neuron-specific enolase has potential value as a biomarker for [(18)F]FDG/[(68)Ga]Ga-PSMA-11 PET mismatch findings in advanced mCRPC patients. EJNMMI Res. (2020) 10 (1): 52. doi:10.1186/s13550-020-00640-2
  32. 32. Mebratie DY, Dagnaw GG. Review of immunohistochemistry techniques: applications, current status, and future perspectives. Semin Diagn Pathol. (2024) 41 (3): 154 – 60. doi:10.1053/j.semdp.2024.05.001
  33. 33. Privé BM, Boussihmad MA, Timmermans B, van Gemert WA, Peters SMB, Derks YHW Fibroblast activation protein-targeted radionuclide therapy: background, opportunities, and challenges of first (pre)clinical studies. Eur J Nucl Med Mol Imaging. (2023) 50 (7): 1906 – 18. doi:10.1007/s00259-023-06144-0

Citation

Jedamzik, T., et al. (2026). Prognostic impact of discordant lesions on [ 18 F]FDG and [ 68 Ga]Ga-FAPI-04 PET/CT compared to histological FAP expression in neuroendocrine neoplasms. Frontiers in Nuclear Medicine, Volume 6. https://doi.org/
Received
Accepted
Published
Volume
6 - 2026

Copyright

© 2026 Jedamzik, Kosmala, Kircher, Heinrich, Schlötelburg, Buck, Meining, Werner, Weich and Michalski.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Tim Jedamzik

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

ReviewFrom task-specific AI to multimodal foundation models in PET/CT: a readiness framework for clinical translationAmgad Muneer, Maliazurina B. Saad, Hui Xu, Jia WuOriginal ResearchAbsolute stress myocardial blood flow is reduced in patients with prior cardiac disease despite normal perfusion and ejection fraction on rubidium-82 PET: an Arabian Gulf cohortAhmad Alenezi, Masoud Garashi, Layla Ghadenfer, Satish Panchadar, Gautam Biswas, Kawther AliMini ReviewQuantification in bone SPECT/CTJovana Zivanovic, Jules Zhang-Yin