Abstract
Objective
This study aimed to evaluate the prognostic value of [18F]ALF-NOTA-FAPI-04 PET/CT-derived parameters, including the maximum standardized uptake value (SUVmax), FAPI-avid tumor volume (FTV), and total lesion FAP expression (TLF), in patients with recurrent gastric cancer undergoing combined chemoimmunotherapy. We sought to establish a noninvasive imaging biomarker framework to optimize patient stratification and therapeutic decision-making.
Materials and Methods
This retrospective cohort study analyzed 51 patients with recurrent gastric cancer who received programmed cell death protein 1 (PD-1) inhibitors combined with chemotherapy after gastrectomy. All patients underwent [18F]ALF-NOTA-FAPI-04 PET/CT within 14 days of chemoimmunotherapy. Semi-quantitative parameters (SUVmax, FTV, and TLF) were derived using semi-automated tumor segmentation. The primary endpoint of this study was the assessment of the clinical efficacy of chemoimmunotherapy, categorized as a durable or nondurable clinical benefit. The secondary endpoints included progression-free survival (PFS) and overall survival (OS). In a subgroup of 16 patients who underwent sequential [18F]FDG PET/CT within seven days of [18F]ALF-NOTA-FAPI-04 PET/CT, tracer uptake values were compared between the two PET/CT examinations.
Results
51 patients were included. The median PFS and OS were 7 and 10 months, respectively. Durable clinical benefit (DCB) was observed in 30 patients and showed significantly lower SUVmax, FTV, and TLF values than non-DCB. TLF demonstrated the highest diagnostic accuracy for DCB (area under the receiver operating characteristic curve [AUC] = 0.80). Multivariable analysis identified TLF ≥ 188.88 SUVbw·cm3 as an independent factor associated with PFS (hazard ratio [HR] = 7.29, P = 0.001) and FTV ≥ 44.17 cm3 as an independent factor associated with OS (HR = 5.16, P = 0.010). In the subgroup analysis of 16 patients, semi-quantitative analysis demonstrated consistently higher values of [18F]ALF-NOTA-FAPI-04-derived parameters than [18F]FDG-derived parameters (all P < 0.001).
Conclusion
[18F]ALF-NOTA-FAPI-04 PET/CT may be a useful imaging tool for predicting clinical outcomes in patients with recurrent gastric cancer undergoing treatment with PD-1 inhibitors and chemotherapy.
Keywords: Fibroblast activation protein inhibitor, Gastric cancer, Positron emission tomography, Immunotherapy, Prognosis
INTRODUCTION
Gastric cancer represents a major global health burden, accounting for approximately 650,000 deaths annually [1]. Early-stage disease is associated with favorable postoperative survival; however, recurrence leads to dismal outcomes, with a 3-year survival rate below 20% and a median overall survival (OS) of less than 10 months [2,3,4]. Overcoming therapeutic resistance in recurrent settings necessitates novel treatment strategies [5], underscoring the need for reliable prognostic and predictive biomarkers.
Immune checkpoint blockade (ICB), particularly programmed cell death protein 1 (PD-1) inhibitors combined with chemotherapy, has improved outcomes in patients with recurrent gastric cancer [6,7]. However, only a subset of patients derives meaningful benefit from chemoimmunotherapy, and accurately predicting treatment responders remains challenging. The prognostic utility of [18F] FDG PET/CT in gastric cancer is constrained by histological heterogeneity and nonspecific physiological uptake [8,9]. Given these limitations of [18F]FDG PET/CT in predicting responses to chemoimmunotherapy, the development of novel radiopharmaceuticals for treatment response assessment is warranted.
The tumor microenvironment (TME) plays a critical role in therapeutic resistance. Cancer-associated fibroblasts (CAFs) expressing fibroblast activation protein (FAP) contribute to immunosuppression and tumor progression [10,11]. FAP is a transmembrane serine protease that is overexpressed in approximately 80%–90% of gastric cancers [12,13] and inhibits cytotoxic T-cell function [10]. By targeting FAP, FAPI PET/CT provides higher tumor-to-background contrast and greater specificity than [18F]FDG in gastric cancer [8,12]. Furthermore, FAPI PET/CT has demonstrated potential utility in predicting chemotherapy response [14]. Nevertheless, its ability to predict long-term outcomes of immunotherapy, especially in patients with recurrent gastric cancer receiving combined chemoimmunotherapy, remains insufficiently explored.
Therefore, this study aimed to evaluate the prognostic value of [18F]ALF-NOTA-FAPI-04 PET/CT in patients with recurrent gastric cancer treated with PD-1 inhibitors and chemotherapy.
MATERIALS AND METHODS
Study Design and Participants
This retrospective cohort study was approved by the Institutional Review Board of the Hunan Cancer Hospital (2021 New Medical Technology Expedited Review [IRB No. 02]) and registered at ClinicalTrials.gov (NCT06543108).
Between October 2020 and May 2024, a total of 185 consecutive patients who underwent [18F]ALF-NOTA-FAPI-04 PET/CT for restaging after radical gastric cancer resection at our institution were screened. All patients underwent total/subtotal gastrectomy with D2 lymphadenectomy (advanced gastric cancer) or D1 + β/D2 lymphadenectomy (early gastric cancer). Postoperative follow-up was performed every two–three months during the first year and every six months thereafter. This schedule continued for three years for early gastric cancer and five years for advanced disease, followed by annual evaluations using clinical, laboratory, imaging, and endoscopic methods. The inclusion Criteria were as follows: 1) first recurrence of gastric cancer confirmed histologically or radiologically through multidisciplinary consensus after surgery, 2) initiation of anti-PD-1 immunotherapy combined with chemotherapy within 14 days following [18F]ALF-NOTA-FAPI-04 PET/CT, and 3) no previous exposure to immunotherapy. The exclusion criteria were as follows: 1) concurrent radiotherapy or interventional therapy administered during the immunotherapy period, 2) inability to complete ≥80% of the planned immunotherapy cycles, 3) diagnosis of another malignancy within the previous 5 years. The patient enrollment flowchart is shown in Figure 1. After exclusions, 51 patients (20 males, 31 females; mean age, 51.5; standard deviation, 11.8 years) were included. Moreover, 16 patients (6 males, 10 females; mean age, 46.5; standard deviation, 10.6 years) sequentially underwent [18F]FDG PET/CT within 7 days after [18F]ALF-NOTA-FAPI-04 PET/CT.
Fig. 1. Flowchart illustrating patient enrollment, treatment, and follow-up. PD = progressive disease, BOR = best objective response.
PET/CT
The synthesis of [18F]ALF-NOTA-FAPI-04 was performed manually following an established protocol [15]. [18F]FDG was synthesized according to standard procedures. All scans were acquired using the Discovery MI digital PET/CT (GE Healthcare, Milwaukee, WI, USA). Prior to [18F]FDG PET/CT, all patients fasted for at least 6 hours. Only those with blood glucose levels confirmed to be below 11.0 mmol/L received an intravenous injection of [18F]FDG. The intravenous dose for each patient was determined based on body weight, following a protocol of 3.7 MBq (0.1 mCi) per kilogram for either [18F]FDG or [18F]ALF-NOTA-FAPI-04. For both [18F] FDG and [18F]ALF-NOTA-FAPI-04 PET/CT, the protocol was as follows: patients remained in a quiet environment for 60 ± 5 minutes prior to imaging. A low-dose whole-body CT scan was performed using a 64-slice helical CT with the following settings: 110 kV; 200–360mAs with automated dose modulation; slice: 3.5 mm; matrix, 512 × 512; 0.5-second rotation time; pitch index 0.984. CT images were reconstructed with a slice thickness of 3.5 mm using adaptive statistical iterative reconstruction. Immediately after the CT scan, whole-body PET scanning (6–7 bed positions) was performed using the 3D acquisition mode (matrix: 256 × 256; 2 minutes/bed position). All PET data were reconstructed using the Bayesian penalized likelihood algorithm (Q.Clear, GE Healthcare, Waukesha, WI, USA), with a regularization parameter (β) set to 750.
Image Analysis
Two nuclear medicine physicians, each with over four years of diagnostic experience in PET/CT, independently analyzed the [18F]ALF-NOTA-FAPI-04 PET/CT images using the Advantage Workstation (v4.7, GE Healthcare). Any disagreements were resolved by a third senior nuclear medicine physician with 15 years of experience. Positive uptake was defined as visually identifiable focal uptake exceeding background levels, after excluding physiological uptake and benign etiologies, such as trauma, infection, or inflammation. Subsequently, the whole-body volume of interest (VOI) encompassing all target lesions was automatically delineated using an AI-Assisted Analysis System for Oncological PET/CT Imaging (uAI Discover-PET/CT Oncology V1.0). The semi-quantitative parameters obtained from [18F]FDG PET/CT included the SUVmax, metabolic tumor volume (MTV), and total lesion glycolysis (TLG). Similarly, the [18F]ALF-NOTA-FAPI-04 PET/CT parameters included SUVmax, FAPI-avid tumor volume (FTV), and total lesion FAP expression (TLF). Quantitative parameters were obtained using an adaptive threshold-based three-dimensional (3D) segmentation algorithm with a 30% SUVmax threshold [16].
Follow-Up and Clinical Endpoint
All patients received anti-PD-1 immunotherapy combined with chemotherapy. The treatment duration was adjusted according to the patient’s condition. Regular follow-ups, comprising clinical assessments, laboratory tests, and imaging evaluations (contrast-enhanced CT or MRI), were performed every two to three treatment cycles. Six months after treatment completion, the treatment response was assessed using the modified RECIST criteria (complete response, partial response, stable disease, or progressive disease) [17]. The primary endpoints were durable clinical benefits (DCB) and nondurable clinical benefits (NDB). DCB was defined as a complete response, partial response, or stable disease lasting more than six months after treatment. NDB was defined as progressive disease or death within six months [18,19]. Secondary endpoints included progression-free survival (PFS) and OS. PFS was defined as the time from chemoimmunotherapy initiation to disease progression or death. OS was defined as the time from chemoimmunotherapy initiation to death from any cause. Patients without events were censored at the last follow-up on September 5, 2025.
Survival outcomes (OS and PFS) were retrieved from medical records and structured telephone interviews. For five patients with out-of-hospital mortality who lacked family-reported time-of-death documentation, time-of-death was estimated from the mid-point of the interview interval using the mid-point method [20].
Statistical Analysis
For group comparisons, continuous and categorical variables were analyzed using parametric/non-parametric tests (t-test/Mann-Whitney U test) and χ2/Fisher’s exact tests, respectively. The Wilcoxon signed-rank test was used to compare tracer uptake between [18F]FDG and [18F]ALF-NOTA-FAPI-04 PET/CT scans. The diagnostic accuracy for predicting DCB was evaluated using receiver operating characteristic (ROC) curve analysis, with pairwise comparisons of the area under the curve (AUC) performed using the DeLong test. Survival analysis was performed using the Cox proportional hazards model. Significant variables (P < 0.05) from univariable analyses were evaluated for multicollinearity (variance inflation factor [VIF] <5) before inclusion in the multivariable Cox models. Kaplan-Meier curves were generated to illustrate survival outcomes. Details of the sample size estimation are described in the Supplement. All analyses were performed using R version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria), with P < 0.05 considered statistically significant.
RESULTS
Patient Characteristics
This retrospective study analyzed 51 patients with recurrent gastric cancer who underwent radical resection, followed by immunotherapy combined with chemotherapy (Table 1). Histologically confirmed recurrence was observed in 35 patients (68.6%), and 16 patients (31.4%) were diagnosed radiologically. Advanced gastric cancer accounted for 96.1% of cases. The 8th AJCC III stage was predominant (78.4%). Signet ring cell carcinoma (SRCC) or mucinous adenocarcinoma (MA) was diagnosed in 21 patients (41.2%), while adenocarcinoma was diagnosed in 30 patients (58.8%). The Lauren classification revealed the diffuse type (64.7%, n = 33) and intestinal type (35.3%, n = 18). Median [18F]ALF-NOTA-FAPI-04 PET parameters included SUVmax 11.00 (interquartile range [IQR]: 6.2–15.1), FTV 44.17 cm3 (IQR: 16.0–146.4), and TLF 188.88 SUVbw·cm3 (IQR: 44.27–621.70).
Table 1. Characteristics of patients and comparison between DCB and NDB patients.
| Characteristic | Total (n = 51) | DCB (n = 30) | NDB (n = 21) | P | |
|---|---|---|---|---|---|
| Age, yr | 51.5 ± 11.8 | 49.6 ± 11.2 | 54.1 ± 12.3 | 0.185 | |
| Sex | 0.304 | ||||
| Female | 31 (60.8) | 20 (66.7) | 11 (52.4) | ||
| Male | 20 (39.2) | 10 (33.3) | 10 (47.6) | ||
| Type of gastrectomy | 0.463 | ||||
| Subtotal | 36 (70.6) | 20 (66.7) | 16 (76.2) | ||
| Total | 15 (29.4) | 10 (33.3) | 5 (23.8) | ||
| Histology | 0.174 | ||||
| Adenocarcinoma | 30 (58.8) | 20 (66.7) | 10 (47.6) | ||
| SRCC/MA | 21 (41.2) | 10 (33.3) | 11 (52.4) | ||
| Lauren histotype | 0.726 | ||||
| Intestinal | 18 (35.3) | 10 (33.3) | 8 (38.1) | ||
| Diffuse | 33 (64.7) | 20 (66.7) | 13 (61.9) | ||
| Pathologic T stage* | 0.446 | ||||
| T1 | 2 (3.9) | 1 (3.3) | 1 (4.8) | ||
| T2 | 4 (7.8) | 3 (10.0) | 1 (4.8) | ||
| T3 | 15 (30.0) | 11 (36.7) | 4 (19.0) | ||
| T4 | 30 (58.8) | 15 (50.0) | 15 (71.4) | ||
| Pathologic N stage* | 0.581 | ||||
| N0 | 15 (29.4) | 10 (33.3) | 5 (23.8) | ||
| N1 | 7 (13.7) | 5 (16.7) | 2 (9.5) | ||
| N2 | 6 (11.8) | 4 (13.3) | 2 (9.5) | ||
| N3 | 23 (45.1) | 11 (36.7) | 12 (57.2) | ||
| TNM stage* | 0.158 | ||||
| I | 1 (2.0) | 0 (0) | 1 (4.7) | ||
| II | 10 (19.6) | 8 (26.7) | 2 (9.4) | ||
| III | 40 (78.4) | 22 (73.3) | 18 (85.7) | ||
| Sites of recurrence | 0.098 | ||||
| Locoregional recurrence only | 8 (15.7) | 7 (23.3) | 1 (4.8) | ||
| Distant metastasis only | 30 (58.8) | 18 (60.0) | 12 (57.1) | ||
| Locoregional and distant failure | 13 (25.5) | 5 (16.7) | 8 (38.1) | ||
| Chemotherapy regimens* | 0.959 | ||||
| SOX | 22 (43.1) | 13 (43.3) | 9 (42.9) | ||
| TP | 13 (25.5) | 8 (26.7) | 5 (23.8) | ||
| XELOX | 16 (31.4) | 9 (30.0) | 7 (33.3) | ||
| Immunotherapy regimens | 0.646 | ||||
| Nivolumab | 11 (21.6) | 8 (26.7) | 3 (14.3) | ||
| Tislelizumab | 7 (13.7) | 4 (13.3) | 3 (14.3) | ||
| Sintilimab | 33 (64.7) | 18 (60.0) | 15 (71.4) | ||
| CEA, ng/mL | 3.00 [2.0, 7.3] | 3.01 [2.0, 7.2] | 2.74 [1.5, 7.0] | 0.432 | |
| CA19-9, U/mL | 20.41 [9.0, 60.6] | 25.33 [10.0, 60.6] | 18.17 [7.2, 45.9] | 0.464 | |
| CA125, U/mL | 14.69 [9.4, 28.1] | 14.00 [8.3, 26.2] | 14.77 [11.9, 28.7] | 0.504 | |
| [18F]ALF-NOTA-FAPI-04 SUVmax | 11.00 [6.2, 15.1] | 7.85 [4.9, 13.8] | 11.80 [9.9, 16.2] | 0.039 | |
| FTV, cm3 | 44.17 [16.0, 146.3] | 25.07 [6.8, 57.3] | 82.00 [48.8, 249.8] | 0.001 | |
| TLF, SUVbw·cm3 | 188.88 [44.2, 621.7] | 65.20 [17.9, 179.0] | 476.60 [195.4, 899.7] | <0.001 | |
Data are presented as mean ± standard deviation, number (percentage), and median [interquartile range].
*According to 8th American Joint Committee on Cancer staging system.
DCB = durable clinical benefit, NDB = nondurable clinical benefit, SRCC = signet ring cell carcinoma, MA = mucinous adenocarcinoma, TNM = tumor-node-metastasis, SOX = S1 plus oxaliplatin, TP = paclitaxel plus cisplatin, XELOX = oxaliplatin plus capecitabine, CEA = carcinoembryonic antigen, CA = carbohydrate antigen, SUVmax = maximum standardized uptake value, FTV = FAPI-avid tumor volume, TLF = total lesion fibroblast activation protein expression
Treatment Efficacy
Immunotherapy combined with chemotherapy was initiated within 2 weeks following [18F]ALF-NOTA-FAPI-04 PET/CT. Thirty-three, eleven, and seven patients received sintilimab, nivolumab, and tislelizumab immunotherapy regimens, respectively. Twenty-two, thirteen, and sixteen patients received the SOX (S1 plus oxaliplatin), TP (paclitaxel plus cisplatin), and XELOX (oxaliplatin plus capecitabine) chemotherapy regimens, respectively. The selection of anti-PD-1 and chemotherapy drugs was determined by various factors, including disease stage, therapeutic sequencing, and patient-specific variables, such as treatment preference and drug formulary availability under insurance plans. The median follow-up duration was nine months (range, 3.0–43.5). Median PFS and OS for the cohort were 7 months (IQR, 5.0–9.0) and 10 months (IQR, 8.7–14.4), respectively. Six months after treatment completion, the overall response rate was 35.3% (18/51), consisting of 2 complete responses (CR) and 16 partial responses (PR). The remaining cases (33/51, 65.7%) were classified as having stable disease (SD; n = 16) or progressive disease (PD; n = 17). DCB was observed in 58.8% of patients (30/51) and 41.2% (21/51) of patients with NDB.
Comparative Analysis Between DCB and NDB Groups
The key predictive factors for the DCB and NDB groups are summarized in Table 1. SUVmax, FTV, and TLF showed statistically significant differences (all P < 0.05). Comparisons of semi-quantitative [18F]ALF-NOTA-FAPI-04 PET parameters between the DCB and NDB groups are presented in Figure 2, and representative [18F]ALF-NOTA-FAPI-04 PET/CT images are displayed in Figure 3. Detailed descriptions of Figure 3 are presented in Supplementary Figures 1-8.
Fig. 2. Comparison of [18F]ALF-NOTA-FAPI-04 SUVmax (A), FTV (B), and TLF (C) between DCB and NDB patients. SUVmax = maximum standardized uptake value, FTV = FAPI-avid tumor volume, TLF = total lesion fibroblast activation protein expression, DCB = durable clinical benefit, NDB = nondurable clinical benefit.
Fig. 3. Representative [18F]ALF-NOTA-FAPI-04 PET/CT images in patients with DCB versus NDB. A-D: Representative [18F]ALF-NOTA-FAPI-04 PET/CT images for four study participants with DCB. Four patients had low [18F]ALF-NOTA-FAPI-04 avidity and metabolic tumor burden. E-H: Representative [18F]ALF-NOTA-FAPI-04 PET/CT images for four study participants with NDB. Four patients had high [18F]ALF-NOTA-FAPI-04 avidity and metabolic tumor burden. The red arrow indicates the location of the lesion. DCB = durable clinical benefit, NDB = nondurable clinical benefit, SUVmax = maximum standardized uptake value, wb = whole body.
ROC analysis was performed to assess the diagnostic accuracy of [18F]ALF-NOTA-FAPI-04 FAPI PET/CT in distinguishing DCB from NDB. While the AUC values for TLF (0.80; IQR: 0.67–0.94), FTV (0.76; IQR: 0.63–0.90), and SUVmax (0.67; IQR: 0.52–0.82) showed a descending trend, however, pairwise comparisons (TLF vs. SUVmax, P = 0.065; TLF vs. FTV, P = 0.186; FTV vs. SUVmax, P = 0.277) did not demonstrate statistically significant differences (Table 2).
Table 2. Performance of [18F]ALF-NOTA-FAPI-04 SUVmax, FTV and TLF for discriminating DCB from NDB.
| Parameters | AUC | Sensitivity* | Specificity* |
|---|---|---|---|
| SUVmax | 0.67 (0.52–0.82) | 0.53 (0.35–0.71) [16/30] | 0.86 (0.71–1.00) [18/21] |
| FTV | 0.76 (0.63–0.90) | 0.60 (0.42–0.78) [18/30] | 0.90 (0.78–1.00) [19/21] |
| TLF | 0.80 (0.67–0.94) | 0.73 (0.58–0.89) [22/30] | 0.90 (0.78–1.00) [19/21] |
Values in parentheses are 95% confidence interval, and numbers of patients are in brackets.
*The diagnostic cutoff values for these results were determined using the median values of the [18F]ALF-NOTA-FAPI-04 PET/CT-derived parameters (11.0 for SUVmax, 44.17 cm3 for FTV and 188.88 SUVbw·cm3 for TLF, respectively).
SUVmax = maximum standardized uptake value, FTV = FAPI-avid tumor volume, TLF = total lesion fibroblast activation protein expression, DCB = durable clinical benefit, NDB = nondurable clinical benefit, AUC = area under the receiver operating characteristics curve
Survival Analysis
For the prediction of PFS and OS, patients with recurrent gastric cancer were stratified using median values of SUVmax = 11.0, FTV = 44.17 cm3, and TLF = 188.88 SUVbw·cm3. In univariable survival analysis, all PET parameters, including SUVmax, FTV, and TLF, were significantly associated with both PFS and OS. Multicollinearity assessment revealed that the VIF values for SUVmax, FTV, and TLF were 1.304, 1.647, and 2.028, respectively. As all VIF values were below the threshold of 5, all three variables were retained for subsequent multivariable analysis. In multivariable analysis for PFS, TLF ≥ 188.88 SUVbw·cm3 (hazard ratio [HR]; 7.29 [95% confidence interval [CI], 2.16–24.56]; P = 0.001; Table 3) was the only independent factor associated with poor PFS. In multivariable analyses for OS, FTV ≥ 44.17 cm3 (HR; 5.16 [95% CI, 1.47–18.13]; P = 0.010) was independently associated with poor OS. The Kaplan-Meier plots are shown in Figures 4 and 5.
Table 3. Univariable and multivariable Cox proportional-hazards regression analysis for PFS and OS.
| Variables | PFS | OS | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Univariable analysis | Multivariable analysis | Univariable analysis | Multivariable analysis | ||||||
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | ||
| Age | 1.00 (0.97–1.03) | 0.932 | 0.98 (0.95–1.02) | 0.434 | |||||
| Sex | |||||||||
| Female | 1.00 (reference) | 1.00 (reference) | |||||||
| Male | 1.11 (0.57–2.17) | 0.765 | 0.85 (0.35–2.07) | 0.728 | |||||
| Pathologic T stage* | |||||||||
| T1 | 1.00 (reference) | 1.00 (reference) | |||||||
| T2 | 1.49 (0.32–6.92) | 0.612 | 1.68 (0.27–10.51) | 0.581 | |||||
| T3 | 1.14 (0.25–5.13) | 0.862 | 0.68 (0.09–4.83) | 0.697 | |||||
| T4 | 2.91 (0.83–10.14) | 0.094 | 1.68 (0.38–7.45) | 0.495 | |||||
| Pathologic N stage* | |||||||||
| N0 | 1.00 (reference) | 1.00 (reference) | |||||||
| N1 | 0.67 (0.18–2.50) | 0.556 | 1.61 (0.29–8.95) | 0.589 | |||||
| N2 | 0.68 (0.18–2.54) | 0.571 | 0.69 (0.08–6.23) | 0.743 | |||||
| N3 | 1.80 (0.81–3.97) | 0.146 | 2.82 (0.92–8.59) | 0.069 | |||||
| TNM stage* | |||||||||
| I | 1.00 (reference) | 1.00 (reference) | |||||||
| II | 0.70 (0.14–3.51) | 0.668 | 0.34 (0.03–3.78) | 0.381 | |||||
| III | 2.02 (0.57–7.15) | 0.274 | 1.34 (0.31–5.86) | 0.695 | |||||
| Type of gastrectomy | |||||||||
| Subtotal | 1.00 (reference) | 1.00 (reference) | |||||||
| Total | 1.12 (0.55–2.29) | 0.760 | 0.60 (0.20–1.81) | 0.367 | |||||
| Histology | |||||||||
| Adenocarcinoma | 1.00 (reference) | 1.00 (reference) | |||||||
| SRCC/MA* | 1.60 (0.83–3.10) | 0.161 | 1.01 (0.42–2.40) | 0.983 | |||||
| Lauren histotype | |||||||||
| Intestinal | 1.00 (reference) | 1.00 (reference) | |||||||
| Diffuse | 1.04 (0.51–2.09) | 0.918 | 1.87 (0.68–5.16) | 0.228 | |||||
| Sites of recurrence | |||||||||
| Locoregional recurrence only | 1.00 (reference) | 1.00 (reference) | |||||||
| Distant metastasis only | 1.89 (0.63–5.68) | 0.254 | 1.18 (0.31–4.46) | 0.807 | |||||
| Locoregional and distant failure | 2.49 (0.78–7.94) | 0.122 | 1.54 (0.41–5.86) | 0.524 | |||||
| Chemotherapy regimens* | |||||||||
| SOX | 1.00 (reference) | 1.00 (reference) | |||||||
| TP or XELOX | 1.09 (0.57–2.12) | 0.788 | 1.38 (0.57–3.32) | 0.476 | |||||
| Immunotherapy regimens | |||||||||
| Sintilimab | 1.00 (reference) | 1.00 (reference) | |||||||
| Nivolumab or tislelizumab | 1.04 (0.52–2.05) | 0.915 | 1.07 (0.42–2.73) | 0.881 | |||||
| [18F]ALF-NOTA-FAPI-04 SUVmax | |||||||||
| <11.0 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | |||||
| ≥11.0 | 2.12 (1.07–4.19) | 0.030 | 0.77 (0.32–1.82) | 0.544 | 2.71 (1.05–7.00) | 0.039 | 2.41 (0.72–8.05) | 0.153 | |
| FTV | |||||||||
| <44.17 cm3 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | |||||
| ≤44.17 cm3 | 3.15 (1.55–6.43) | 0.002 | 1.25 (0.53–2.94) | 0.616 | 5.39 (1.76–16.47) | 0.003 | 5.16 (1.47–18.13) | 0.010 | |
| TLF | |||||||||
| <188.88 SUVbw·cm3 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) | |||||
| ≥188.88 SUVbw·cm3 | 6.94 (3.00–16.04) | <0.001 | 7.29 (2.16–24.56) | 0.001 | 3.76 (1.44–9.85) | 0.007 | 1.41 (0.38–5.26) | 0.607 | |
*According to 8th American Joint Committee on Cancer staging system.
PFS = progression-free survival, OS = overall survival, HR = hazard ratio, CI = confidence interval, TNM = tumor-node-metastasis, SRCC = signet ring cell carcinoma, MA = mucinous adenocarcinoma, SOX = S1 plus oxaliplatin, TP = paclitaxel plus cisplatin, XELOX = oxaliplatin plus capecitabine, SUVmax = maximum standardized uptake value, FTV = FAPI-avid tumor volume, TLF = total lesion fibroblast activation protein expression
Fig. 4. Comparison of PFS in patients stratified by the median for [18F]ALF-NOTA-FAPI-04 TLF. Kaplan–Meier analysis was performed to assess unadjusted PFS (HR, 6.94 [95% CI, 3.00–16.04]; P < 0.001) and multivariable Cox regression for adjusted PFS (HR, 7.29 [95% CI, 2.16–24.56]; P = 0.001), with “TLF = low” as reference. PFS = progression-free survival, TLF = total lesion fibroblast activation protein expression, HR = hazard ratio, CI = confidence interval.
Fig. 5. Comparison of OS in patients stratified by the median for [18F]ALF-NOTA-FAPI-04 FTV. Kaplan–Meier analysis was performed to assess unadjusted OS (HR, 5.39 [95% CI, 1.76–16.47]; P = 0.003) and multivariable Cox regression for adjusted OS (HR, 5.16 [95% CI, 1.47–18.13]; P = 0.010), with “FTV = low” as reference. OS = overall survival, FTV = FAPI-avid tumor volume, HR = hazard ratio, CI = confidence interval.
Comparative Analysis of Semi-Quantitative Parameters Between [18F]FDG and [18F]ALF-NOTA-FAPI-04 PET/CT
A subgroup analysis was performed in 16 patients who sequentially underwent both [18F]FDG and [18F]ALF-NOTA-FAPI-04 PET/CT scans within a one-week interval prior to the initiation of chemoimmunotherapy. Recurrent gastric lesions were identified in all 16 patients using [18F]ALF-NOTA-FAPI-04 PET/CT, whereas [18F]FDG PET/CT failed to detect lesions in four cases. Among the matched lesions in 12 patients, [18F]ALF-NOTA-FAPI-04–derived parameters were significantly higher than those derived from [18F]FDG (all P < 0.05; Table 4).
Table 4. Comparison of semi-quantitative parameters in 12 patients with positive findings on both and [18F]ALF-NOTA-FAPI-04 PET/CT and [18F]FDG.
| Parameters | [18F]ALF-NOTA-FAPI-04 (n = 12) | [18F]FDG (n = 12) | P |
|---|---|---|---|
| SUVmax | 13.60 [11.0, 15.6] | 4.60 [3.6, 10.9] | 0.002 |
| MTV/FTV, cm3 | 53.15 [25.2, 200.3] | 2.30 [1.5, 8.6] | 0.001 |
| TLG/TLF, SUVbw·cm3 | 243.84 [98.5, 654.3] | 6.65 [3.0, 31.5] | 0.001 |
Data are presented as median [interquartile range].
SUVmax = maximum standardized uptake value, MTV = metabolic tumor volume, FTV = FAPI-avid tumor volume, TLG = total lesion glycolysis, TLF = total lesion fibroblast activation protein expression
DISCUSSION
This study investigated the potential of baseline [18F] ALF-NOTA-FAPI-04 PET/CT to predict outcomes in patients with recurrent gastric cancer receiving PD-1 inhibitor plus chemotherapy. Baseline [18F]ALF-NOTA-FAPI-04 PET/CT parameters effectively distinguished DCB from NDB, with lower values correlating with longer PFS and OS. Multivariable analysis revealed that TLF was an independent predictor of poor PFS, whereas FTV independently predicted reduced OS. Additionally, [18F]ALF-NOTA-FAPI-04 showed significantly higher tracer uptake values than [18F]FDG, suggesting its potential for superior lesion detection. Therefore, baseline [18F]ALF-NOTA-FAPI-04 PET/CT parameters may offer a potential tool for risk stratification prior to chemoimmunotherapy in patients with recurrent gastric cancer.
Our study demonstrated that [18F]ALF-NOTA-FAPI-04 PET/CT-derived parameters effectively differentiated DCB from NDB in patients with recurrent gastric cancer undergoing chemoimmunotherapy. Previous studies have suggested that FAPI PET/CT holds potential value in predicting responses to immunotherapy. Rong et al. [21] reported that elevated [68Ga]FAPI-04 PET/CT SUVmax correlated with progressive disease and an immunosuppressive tumor microenvironment in patients with gastric cancer following immunotherapy. Similarly, Wu et al. [18] identified [68Ga]FAPI-derived volumetric parameters (FTV and TLF) as strong predictors of NDB in patients with unresectable hepatocellular carcinoma treated with PD-1 inhibitors and lenvatinib, findings that are consistent with the present study. To date, the application of FAPI PET in predicting DCB/NDB in patients with recurrent gastric cancer treated with chemoimmunotherapy has not been previously described, highlighting the novelty of our findings.
TLF and FTV were identified as independent predictors of PFS and OS, consistent with emerging evidence supporting the prognostic utility of FAPI PET/CT. Sun et al. [22] reported that the [68Ga]FAPI-derived peritoneal cancer index independently predicted poor PFS in gastric mucinous adenocarcinoma and signet-ring cell carcinoma. Similarly, Watanabe et al. [23] established [68Ga]FAPI SUVmax as a prognostic biomarker for both PFS and OS across multiple solid tumors. However, none of these studies specifically evaluated the long-term prognostic value of FAPI PET-derived volumetric parameters, such as FTV and TLF. Notably, Zhu et al. [16] demonstrated that changes in MTV/TLF during chemotherapy predicted PFS in pancreatic ductal adenocarcinoma, supporting the broader prognostic relevance of volumetric PET metrics. Among non-FAPI tracers, [18F]FDG PET/CT parameters have demonstrated robust prognostic value across malignancies [24,25,26]. Nevertheless, their performance in gastric cancer remains suboptimal due to low FDG avidity in certain histological subtypes. We also found that [18F]FDG PET lacked accuracy in predicting patient prognosis, which further underscores the limitations of [18F]FDG PET/CT in gastric cancer. Critically, our work established [18F]ALF-NOTA-FAPI-04 PET/CT-derived volumetric parameters as predictors of long-term survival in patients with recurrent gastric cancer undergoing chemoimmunotherapy, offering clinically potential biomarkers with direct biological relevance.
The prognostic predictive value of FAPI PET/CT is fundamentally rooted in its ability to non-invasively quantify tumor stromal abundance and the associated immunosuppressive TME [27,28]. The radiotracer [18F]ALF-NOTA-FAPI-04 targets FAP expressed on CAFs within the tumor stroma. Consequently, the FAPI PET/CT parameters quantitatively reflect both the abundance and biological activity of the protumoral stromal compartment. In our multivariable analysis, the superior prognostic performance of volumetric parameters (FTV and TLF) over SUVmax underscores critical biological distinctions. SUVmax reflects the single voxel with the highest FAP uptake, and may therefore be influenced by focal inflammation or sampling bias. In contrast, FTV and TLF quantify the total functional mass of FAP-positive CAFs, providing a global assessment of tumor stromal burden. This distinction is mechanistically significant because the pro-tumoral effects of CAFs are largely determined by their total abundance and spatial architecture. A high stromal burden, as reflected by elevated FTV and TLF levels, is mechanistically associated with an immunosuppressive TME. Dense fibroblast networks act as a physical barrier that impedes T-cell infiltration into the tumor core. In addition, CAFs actively contribute to immune evasion by secreting cytokines, chemokines, and extracellular matrix components that inhibit cytotoxic T-cell function. They also promote the recruitment of immunosuppressive cells, including myeloid-derived suppressor cells and regulatory T-cells [29,30,31]. This FAP-positive, CAF-rich niche thereby fosters an immunologically “cold” TME that is inherently resistant to PD-1 inhibitor therapy. Therefore, high [18F]ALF-NOTA-FAPI-04 volumetric parameters identified in our study may serve as integrative biomarkers of this treatment-resistant niche.
However, while our baseline [18F]ALF-NOTA-FAPI-04 PET/CT parameters demonstrated robust overall predictive value, discordant cases warrant further analysis. Notably, early disease progression occurred in a subset of patients with low baseline FAPI uptake. We observed that these individuals frequently presented with distant metastatic involvement (e.g., bone or peritoneal lesions) at recurrence, suggesting that aggressive, site-specific disease phenotypes may override global FAPI-based quantification. Although our analysis detected no statistically significant association between distant metastasis and DCB (potentially limited by the sample size), this observation merits validation in larger cohorts. Conversely, durable clinical benefit was achieved in some patients with high baseline FAPI uptake. In these cases, successful consolidation therapies, exemplified by metastasectomy after chemoimmunotherapy-induced tumor shrinkage, may have contributed substantially to favorable outcomes. Collectively, these findings emphasize that imaging biomarkers should be interpreted within the context of comprehensive clinical profiles, including metastatic patterns, and planned therapeutic interventions.
This single-center, retrospective analysis has several limitations. First, the relatively small sample size (n = 51) may limit statistical power and generalizability. Multicenter validation studies with larger cohorts are required to refine cutoff values and confirm prognostic thresholds. In addition, our cohort exclusively comprised patients with postoperative recurrence, and the predictive role of FAPI PET/CT in treatment-naive gastric cancer requires further exploration. Second, the retrospective study design inherently lacks standardized treatment protocols, as therapeutic regimens (e.g., chemotherapy combinations and immunotherapy dosing) were individualized according to patient-specific clinical conditions. This heterogeneity in treatment administration may have confounded the observed associations between FAPI PET/CT parameters and clinical outcomes. Future prospective studies with protocol-driven therapies are needed to minimize potential confounding variables. Third, the use of midpoint imputation for five patients with out-of-hospital deaths may have introduced potential bias into the overall survival analysis. Finally, correlative studies integrating FAPI uptake with histopathological assessment of FAP expression and immune infiltrate profiling are essential to elucidate the biological mechanistic links between imaging biomarkers and therapeutic resistance.
In conclusion, [18F]ALF-NOTA-FAPI-04 FAPI PET/CT-derived parameters represent promising noninvasive biomarkers for predicting clinical outcomes in patients with recurrent gastric cancer undergoing chemoimmunotherapy. This imaging modality may hold potential to guide personalized therapeutic strategies in recurrent gastric cancer by enhancing patient selection and predicting clinical outcomes.
Footnotes
Conflicts of Interest: The authors have no potential conflicts of interest to disclose.
- Conceptualization: Hui Ye, Yongzhi Xie.
- Data curation: Haifeng He, Chengzhi Jiang, Wanjing Zhou.
- Formal analysis: Haifeng He.
- Investigation: Haifeng He, Yongzhi Xie, Hui Ye.
- Methodology: Haifeng He, Chengzhi Jiang, Wanjing Zhou.
- Project administration: Hui Ye.
- Supervision: Hui Ye.
- Validation: Haifeng He, Chengzhi Jiang.
- Visualization: Haifeng He.
- Writing—original draft: Haifeng He, Yongzhi Xie.
- Writing—review & editing: Hui Ye.
Funding Statement: This work was supported by Hunan Provincial Natural Science Foundation of China (Grant No. 2024JJ9250), Health Research Project of Hunan Provincial Health Commission (Grant No. W20243245).
Availability of Data and Material
The data generated for this study are available on request to the corresponding author.
Supplement
The Supplement is available with this article at https://doi.org/10.3348/kjr.2025.145.
References
- 1.Sundar R, Nakayama I, Markar SR, Shitara K, van Laarhoven HWM, Janjigian YY, et al. Gastric cancer. Lancet. 2025;405:2087–2102. doi: 10.1016/S0140-6736(25)00052-2. [DOI] [PubMed] [Google Scholar]
- 2.Katai H, Ishikawa T, Akazawa K, Isobe Y, Miyashiro I, Oda I, et al. Five-year survival analysis of surgically resected gastric cancer cases in Japan: a retrospective analysis of more than 100,000 patients from the nationwide registry of the Japanese Gastric Cancer Association (2001-2007) Gastric Cancer. 2018;21:144–154. doi: 10.1007/s10120-017-0716-7. [DOI] [PubMed] [Google Scholar]
- 3.Moon YW, Jeung HC, Rha SY, Yoo NC, Roh JK, Noh SH, et al. Changing patterns of prognosticators during 15-year follow-up of advanced gastric cancer after radical gastrectomy and adjuvant chemotherapy: a 15-year follow-up study at a single Korean institute. Ann Surg Oncol. 2007;14:2730–2737. doi: 10.1245/s10434-007-9479-4. [DOI] [PubMed] [Google Scholar]
- 4.Kim SH, Song BI, Kim HW, Won KS, Son YG, Ryu SW. Prognostic value of restaging F-18 fluorodeoxyglucose positron emission tomography/computed tomography to predict 3-year post-recurrence survival in patients with recurrent gastric cancer after curative resection. Korean J Radiol. 2020;21:829–837. doi: 10.3348/kjr.2019.0672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Waldman AD, Fritz JM, Lenardo MJ. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nat Rev Immunol. 2020;20:651–668. doi: 10.1038/s41577-020-0306-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Moehler M, Xiao H, Blum SI, Elimova E, Cella D, Shitara K, et al. Health-related quality of life with nivolumab plus chemotherapy versus chemotherapy in patients with advanced gastric/gastroesophageal junction cancer or esophageal adenocarcinoma from CheckMate 649. J Clin Oncol. 2023;41:5388–5399. doi: 10.1200/JCO.23.00170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Janjigian YY, Kawazoe A, Yañez P, Li N, Lonardi S, Kolesnik O, et al. The KEYNOTE-811 trial of dual PD-1 and HER2 blockade in HER2-positive gastric cancer. Nature. 2021;600:727–730. doi: 10.1038/s41586-021-04161-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lv J, Zheng K, Jiang C, Yang J, Peng X, Ye H, et al. Evaluating the diagnostic performance of [18F]ALF-NOTA-FAPI-04 PET/CT in gastric cancer: a comparative study with [18F]FDG PET/CT. Eur Radiol. 2025;35:3314–3323. doi: 10.1007/s00330-024-11219-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dassen AE, Lips DJ, Hoekstra CJ, Pruijt JF, Bosscha K. FDG-PET has no definite role in preoperative imaging in gastric cancer. Eur J Surg Oncol. 2009;35:449–455. doi: 10.1016/j.ejso.2008.11.010. [DOI] [PubMed] [Google Scholar]
- 10.Peltier A, Seban RD, Buvat I, Bidard FC, Mechta-Grigoriou F. Fibroblast heterogeneity in solid tumors: from single cell analysis to whole-body imaging. Semin Cancer Biol. 2022;86(Pt 3):262–272. doi: 10.1016/j.semcancer.2022.04.008. [DOI] [PubMed] [Google Scholar]
- 11.Zhang H, Deng T, Liu R, Ning T, Yang H, Liu D, et al. CAF secreted miR-522 suppresses ferroptosis and promotes acquired chemo-resistance in gastric cancer. Mol Cancer. 2020;19:43. doi: 10.1186/s12943-020-01168-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Xu W, Cai J, Peng T, Meng T, Pang Y, Sun L, et al. Fibroblast activation protein-targeted PET/CT with 18F-fibroblast activation protein inhibitor-74 for evaluation of gastrointestinal cancer: comparison with 18F-FDG PET/CT. J Nucl Med. 2024;65:40–51. doi: 10.2967/jnumed.123.266329. [DOI] [PubMed] [Google Scholar]
- 13.Ruan D, Zhao L, Cai J, Xu W, Sun L, Li J, et al. Evaluation of FAPI PET imaging in gastric cancer: a systematic review and meta-analysis. Theranostics. 2023;13:4694–4710. doi: 10.7150/thno.88335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Miao Y, Feng R, Yu T, Guo R, Zhang M, Wang Y, et al. Value of 68Ga-FAPI-04 and 18F-FDG PET/CT in early prediction of pathologic response to neoadjuvant chemotherapy in locally advanced gastric cancer. J Nucl Med. 2024;65:213–220. doi: 10.2967/jnumed.123.266403. [DOI] [PubMed] [Google Scholar]
- 15.Jiang X, Wang X, Shen T, Yao Y, Chen M, Li Z, et al. FAPI-04 PET/CT using [18F]AlF labeling strategy: automatic synthesis, quality control, and in vivo assessment in patient. Front Oncol. 2021;11:649148. doi: 10.3389/fonc.2021.649148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhu Z, Cheng K, Yun Z, Zhang X, Hu X, Liu J, et al. [18F] AlF-NOTA-FAPI-04 PET/CT can predict treatment response and survival in patients receiving chemotherapy for inoperable pancreatic ductal adenocarcinoma. Eur J Nucl Med Mol Imaging. 2023;50:3425–3438. doi: 10.1007/s00259-023-06271-8. [DOI] [PubMed] [Google Scholar]
- 17.Lencioni R, Llovet JM. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis. 2010;30:52–60. doi: 10.1055/s-0030-1247132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wu M, Wang Y, Yang Q, Wang X, Yang X, Xing H, et al. Comparison of baseline 68Ga-FAPI and 18F-FDG PET/CT for prediction of response and clinical outcome in patients with unresectable hepatocellular carcinoma treated with PD-1 inhibitor and lenvatinib. J Nucl Med. 2023;64:1532–1539. doi: 10.2967/jnumed.123.265712. [DOI] [PubMed] [Google Scholar]
- 19.Lin X, Liu Z, Zhou K, Li Y, Huang G, Zhang H, et al. Intratumoral and peritumoral PET/CT-based radiomics for non-invasively and dynamically predicting immunotherapy response in NSCLC. Br J Cancer. 2025;132:558–568. doi: 10.1038/s41416-025-02948-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dugué AE, Pulido M, Chabaud S, Belin L, Gal J. How to deal with interval-censored data practically while assessing the progression-free survival: a step-by-step guide using SAS and R software. Clin Cancer Res. 2016;22:5629–5635. doi: 10.1158/1078-0432.CCR-16-1017. [DOI] [PubMed] [Google Scholar]
- 21.Rong X, Lv J, Liu Y, Wang Z, Zeng D, Li Y, et al. PET/CT imaging of activated cancer-associated fibroblasts predict response to PD-1 blockade in gastric cancer patients. Front Oncol. 2022;11:802257. doi: 10.3389/fonc.2021.802257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sun Y, Xu J, Qiao Y, Zhang J, Pan H, Xu X, et al. Assessing the value of 68Ga-FAPI PET/CT in gastric mucinous adenocarcinoma or signet ring cell carcinoma. Radiol Imaging Cancer. 2024;6:e230195. doi: 10.1148/rycan.230195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Watanabe M, Fendler WP, Grafe H, Hirmas N, Hamacher R, Lanzafame H, et al. Prognostic implications of 68Ga-FAPI-46 PET/CT-derived parameters on overall survival in various types of solid tumors. J Nucl Med. 2024;65:1027–1034. doi: 10.2967/jnumed.123.266981. [DOI] [PubMed] [Google Scholar]
- 24.Leithner D, Flynn JR, Devlin SM, Mauguen A, Fei T, Zeng S, et al. Conventional and novel [18F]FDG PET/CT features as predictors of CAR-T cell therapy outcome in large B-cell lymphoma. J Hematol Oncol. 2024;17:21. doi: 10.1186/s13045-024-01540-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Iravani A, Wallace R, Lo SN, Galligan A, Weppler AM, Hicks RJ, et al. FDG PET/CT prognostic markers in patients with advanced melanoma treated with ipilimumab and nivolumab. Radiology. 2023;307:e221180. doi: 10.1148/radiol.221180. [DOI] [PubMed] [Google Scholar]
- 26.Dall’Olio FG, Zrafi W, Roelants V, Ambrosini V, Fourquet A, Mitea C, et al. Metabolic tumor volume assessed by 18F FDG-PET CT scan as a predictive biomarker for immune checkpoint blockers in advanced NSCLC and its biological correlates. Clin Cancer Res. 2025;31:352–364. doi: 10.1158/1078-0432.CCR-24-1993. [DOI] [PubMed] [Google Scholar]
- 27.Shi X, Xing H, Yang X, Li F, Yao S, Zhang H, et al. Fibroblast imaging of hepatic carcinoma with 68Ga-FAPI-04 PET/CT: a pilot study in patients with suspected hepatic nodules. Eur J Nucl Med Mol Imaging. 2021;48:196–203. doi: 10.1007/s00259-020-04882-z. [DOI] [PubMed] [Google Scholar]
- 28.Hirmas N, Hamacher R, Sraieb M, Ingenwerth M, Kessler L, Pabst KM, et al. Fibroblast-activation protein PET and histopathology in a single-center database of 324 patients and 21 tumor entities. J Nucl Med. 2023;64:711–716. doi: 10.2967/jnumed.122.264689. [DOI] [PubMed] [Google Scholar]
- 29.Kieffer Y, Hocine HR, Gentric G, Pelon F, Bernard C, Bourachot B, et al. Single-cell analysis reveals fibroblast clusters linked to immunotherapy resistance in cancer. Cancer Discov. 2020;10:1330–1351. doi: 10.1158/2159-8290.CD-19-1384. [DOI] [PubMed] [Google Scholar]
- 30.Li K, Liu W, Yu H, Chen J, Tang W, Wang J, et al. 68Ga-FAPI PET imaging monitors response to combined TGF-βR inhibition and immunotherapy in metastatic colorectal cancer. J Clin Invest. 2024;134:e170490. doi: 10.1172/JCI181374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sahai E, Astsaturov I, Cukierman E, DeNardo DG, Egeblad M, Evans RM, et al. A framework for advancing our understanding of cancer-associated fibroblasts. Nat Rev Cancer. 2020;20:174–186. doi: 10.1038/s41568-019-0238-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data generated for this study are available on request to the corresponding author.