Comparison of 11C‐Acetate and 18F‐FDG PET/CT for Immune Infiltration and Prognosis in Hepatocellular Carcinoma

Abstract

ABSTRACT

Immunotherapy has revolutionized cancer treatment, making it a challenge to noninvasively monitor immune infiltration. Metabolic reprogramming in cancers, including hepatocellular carcinoma (HCC), is closely linked to immune status. In this study, we aimed to evaluate the ability of carbon‐11 acetate (¹¹C‐acetate) and fluorine‐18 fluorodeoxyglucose (¹⁸F‐FDG) PET/CT findings in predicting overall survival (OS) and immune infiltration in HCC patients. Totally 32 patients who underwent preoperative ¹⁸F‐FDG and ¹¹C‐acetate PET/CT, followed by liver resection for HCC, were prospectively enrolled at authors' institute between January 2019 and October 2021. Tracer uptake was qualified. Densities of CD3⁺, CD8⁺, and granzyme B⁺ CD8⁺ immune cells were assessed and the Immunoscore was defined by combining the densities of CD3⁺ and CD8⁺ in tumor interior (TI) and invasion margin (IM). Patients with avid HCCs in ¹¹C‐acetate PET/CT demonstrated a longer OS. Those with only ¹¹C‐acetate‐avid HCCs exhibited a longer OS compared to those with only ¹⁸F‐FDG uptake. In contrast to ¹⁸F‐FDG uptake, ¹¹C‐acetate uptake was positively associated with CD3⁺, CD8⁺, and granzyme B⁺ CD8⁺ cell infiltration. Patients with a higher Immunoscore exhibited a longer OS and an increased uptake of ¹¹C‐acetate rather than ¹⁸F‐FDG. The sensitivity of ¹¹C‐acetate PET/CT in the detection of patients with immune infiltration was superior to that of ¹⁸F‐FDG PET/CT (88% [21 of 24] vs. 58% [14 of 24]). These data show that preoperative ¹¹C‐acetate PET/CT may be a promising approach for the evaluation of immune status and postoperative outcome of HCCs.

Abbreviations

• ¹¹C-acetate
• carbon-11 acetate
• ¹⁸F-FDG
• fluorine-18 fluorodeoxyglucose
• HCC
• hepatocellular carcinoma
• IM
• invasion margin
• IQR
• interquartile range
• OS
• overall survival
• SUV_max
• maximum standardized uptake value
• TBR
• tumor-to-background ratio
• TI
• tumor interior

Introduction

Hepatocellular carcinoma (HCC) is the fourth most common cause of cancer-related death worldwide [1]. Its prevalence is influenced by genetic susceptibility and multiple risk factors in different populations, contributing to the substantial intertumoral genomic heterogeneity, resulting in unfavorable clinical outcomes [2]. Notably, ongoing metabolic reprogramming during the processes of tumorigenesis, progression, and metastasis is involved in the development of this pronounced heterogeneity [3]. Previous studies have identified several molecular HCC subtypes, with a particular emphasis on metabolism alterations [2].

Differing from conventional imaging modalities like ultrasound, CT, and MRI, PET has the capacity to offer valuable biological insights, with metabolic tracer imaging being particularly adept at highlighting the anomalous uptake of various metabolites by tumors [4, 5]. However, the sensitivity of the clinical routine Fluorine-18 fluorodeoxyglucose (¹⁸F-FDG) for HCC detection is limited, whereas carbon-11 acetate (¹¹C-acetate) demonstrates a better diagnostic performance [6]. It is important to recognize that each metabolic PET tracer possesses the ability to reflect diverse metabolic patterns of tumor.

Immunotherapy has revolutionized the treatment of HCC recently [7]. While the factors determining the effectiveness of immunotherapy remain enigmatic, a pivotal role is attributed to the immune microenvironment, particularly the presence and activation of CD8⁺ T cells, which has been linked to the response to immune-checkpoint inhibitors [7]. Increasing evidence has established important contributions of intratumoral CD3⁺ and CD8⁺ cell densities to the progression [8], recurrence and long-term prognosis of HCC [9]. To further consolidate the quantitative information regarding the infiltration of CD3⁺ and CD8⁺ immune cells in tumor lesions, the Immunoscore has emerged as a valuable prognostic scoring system for patients with solid malignancies [9, 10]. Currently, the assessment of immune infiltration relies on postoperative pathology and further advancements for the noninvasive preoperative visualization of immune infiltration are urgently needed.

Within the diverse cell types in the tumor microenvironment, nutrient competition can trigger immunosuppression, while metabolic crosstalk has been recognized for its role in shaping immune responses [11, 12]. Previous studies have demonstrated the association between ¹⁸F-FDG uptake and tumor-infiltrating lymphocytes in lung and breast cancer [13, 14]. However, the relationships between the metabolic patterns identified through ¹¹C-acetate and ¹⁸F-FDG PET/CT, and their association with immune infiltration and prognosis in HCC, are still unknown.

The purpose of this study was to investigate the potential correlations between the metabolic variables as indicated by ¹¹C-acetate and ¹⁸F-FDG PET/CT with overall survival (OS), and immune infiltration in HCC. We sought to answer the question that whether metabolic PET imaging can contribute to distinguishing specific HCC patients with better postoperative outcome and clarify the immune status aided in the formulation of individualized immunotherapy.

Materials and Methods

Patients and Inclusion Criteria

Patients who underwent preoperative ¹¹C-acetate and ¹⁸F-FDG PET/CT scans, and liver resections for HCC were enrolled in authors' institute from January 2019 to October 2021. The definitive pathological diagnosis of HCC was determined by two independent pathologists. Patients were excluded if they were pathologically diagnosed with non-HCC or had accepted regional or systemic treatments before surgery. Clinical and pathological information of the patients including age at surgery, sex, serum α-fetoprotein (AFP) level, etiology (HBV or HCV infection background), tumor size, stage, histological grade, microvascular invasion, and liver cirrhosis were collected. Patients were regularly followed up every 2–3 months, and overall survival (OS) was defined as the time from the date of surgery to death or the last follow-up. This prospective study was approved by the ethics committee at Huashan Hospital, Fudan University (Ethics approval number: HS-KY2020-729) and conducted in accordance with the principles of the Helsinki Declaration. Informed consent was obtained from all participants.

Image Acquisition and Interpretation of ¹¹C-acetate and ¹⁸F-FDG PET/CT Scans

For ¹¹C-acetate PET/CT, static PET/CT scans were obtained 15 min after the intravenous injection of ~370 MBq (~10 mCi) of ¹¹C-acetate using a dedicated PET/CT scanner (Biograph mCT Flow scanner, Siemens, Germany). For ¹⁸F-FDG PET/CT, all patients were instructed to fast for at least 6 h before the examination. After 60 min of the intravenous injection of ¹⁸F-FDG (5.5 MBq/kg, 0.15 mCi/kg), whole-body PET/CT scans were obtained on the same PET/CT scanner. ¹¹C-acetate and ¹⁸F-FDG PET/CT scans were obtained within 1 or 2 days of each other. PET imaging was acquired in three-dimensional mode, with image reconstruction carried out using the ordered subset expectation maximization three-dimensional method.

PET/CT images were interpreted by two experienced nuclear medicine physicians. Tracer uptake was qualified with maximum standardized uptake value (SUV_max) and tumor-to-background ratio (TBR). SUV_max was measured by delineating a spherical region of interest for each lesion. Regions of interest in the noncancerous liver parenchyma were selected for normalization in the mid-to-lower section of the right hepatic lobe, avoiding large blood vessels and abnormal tissues such as tumors or cysts. Mean standardized uptake values were then extracted from circular areas measuring 2–5 cm², depending on liver size. TBR was calculated by dividing the SUV_max of tumor by the mean standardized uptake value of the noncancerous liver parenchyma. A lesion was considered avid based on visual judgment of elevated tracer uptake in the tumor lesion.

Immunohistochemistry and Immunofluorescence Assays of HCC Tissues

The tissue slides from the formalin-fixed and paraffin-embedded blocks of HCC and nontumor liver tissues of the enrolled patients were immunohistochemically stained with antibodies against CD3 (ab16669, Abcam) and CD8 (ab217344, Abcam). The tumor interior (TI) and invasion margin (IM) were marked for biomarker imaging [9, 15]. Representative images were captured under high-power magnification in the areas of focally high staining density. The densities of infiltrating CD3⁺ and CD8⁺ cells in both TI and IM were determined using ImageJ software (National Institute of Health, USA).

Immunofluorescence was also performed on HCC tissues. The tissues were stained with antibodies against CD8 (ab217344, Abcam), granzyme B (ab4059, Abcam), and 4′,6-Diamidino-2-phenylindole. Images were captured in the areas with high staining density in TI region with a Zeiss Axioimager Z2 Apotome. The number of granzyme B⁺ cells and the percentage of granzyme B⁺ on CD8⁺ T cells were quantified using ImageJ software (National Institute of Health, USA).

The Immunoscore was used for the assessment of T-cell infiltration [9, 10], which was determined by two types of immune cells (CD3⁺ and CD8⁺ cells) in each tumor region (TI and IM). This scoring system assigned a value of 0 (low) for densities ranging from 0% to 25%, and a value of 1 (high) for densities ranging from 25% to 100%. The Immunoscore, with a scale from 0 to 4, was obtained by summing up four individual score values, thus categorizing patients into five groups denoted as Im0, Im1, Im2, Im3, and Im4, respectively.

Statistical Analysis

Clinical characteristics, PET tracer uptake, and immune-related data were summarized as numbers with percentages for categorical variables and medians with ranges or medians with interquartile ranges (IQRs) for continuous variables. Normality was clarified by the Kolmogorov–Smirnov test. Differences in PET tracer uptake among clinical subgroups were compared using unpaired t, Mann–Whitney U, or Kruskal–Wallis test. Spearman correlation analysis was used to analyze the relationships between tracer uptakes and immune-related data. The densities of immune cells were compared between PET tracer-avid and nonavid subgroups using unpaired t or the Mann–Whitney U test, and categorical data were analyzed with Fisher's exact tests. Kaplan–Meier overall survival curves were calculated from the date of surgery to the date of death or the last date of follow-up (July 2023). The log-rank test or the log-rank test for trend was employed to compare the survival difference. All the statistical analyses were performed using GraphPad Prism (GraphPad Prism, version 7.0) and SPSS (SPSS, version 24). Two-tailed p-values less than 0.05 indicated a statistically significant difference.

Results

Patient Characteristics

A total of 32 patients (31 men; median age, 62 years; IQR, 55–70 years) were enrolled in this study. The last follow-up was performed in July 2023, during which three patients (10%) were lost to follow-up, and one patient who died during the postoperative period was excluded from the survival analysis. The median overall survival (OS) was 26 months (range, 10–36 months). The patient selection process is depicted in Figure 1. In six patients with multifocal lesions, a PET-avid result indicated the presence of one or more avid lesions. Among the 32 patients, 13 (41%) displayed avid results in both ¹¹C-acetate and ¹⁸F-FDG PET/CT, 5 (16%) were nonavid in both tracers, 10 (31%) were only ¹¹C-acetate-avid, and 4 (12%) were only ¹⁸F-FDG-avid.

[IMAGE OMITTED. SEE PDF]

The clinical characteristics and PET/CT parameters of patients are summarized in Table 1. Patients with larger tumor size exhibited higher SUV_max and TBR in both ¹¹C-acetate and ¹⁸F-FDG (all p < 0.05). The SUV_max of ¹⁸F-FDG was positively associated with higher BCLC stage and histological grade, while the TBR was associated with microvascular invasion (MVI, all p < 0.05). In contrast, no significant association was found between ¹¹C-acetate uptake and the BCLC stage, histological grade and MVI of HCC (all p > 0.05).

TABLE 1 Demographic characteristics of patients with resectable HCC who underwent preoperative ¹¹C-acetate and ¹⁸F-FDG PET/CT.

Characteristics	All patients (n = 32)a	Median 11C-acetate SUVmax	p	Median 11C-acetate TBR	p	Median 18F-FDG SUVmax	p	Median 18F-FDG TBR	p
Age(y)b	62 (55–70)	4.37 (2.13–9.76)	NA	1.57 (0.83–4.48)	NA	3.05 (1.40–12.64)	NA	1.56 (0.96–6.51)	NA
Sex
Female	1 (3)	5.9	NA	1.97	NA	2.99	NA	1.29	NA
Male	31 (97)	4.34 (2.13–9.76)		1.55 (0.83–4.48)		3.10 (1.40–12.64)	1.61 (0.96–6.51)
Serum AFP
Low (≤ 400 ng/mL)	28 (88)	4.75 (2.13–9.76)	0.85	1.58 (0.83–4.48)	0.34	2.97 (1.64–12.64)	0.25	1.46 (0.96–6.28)	0.15
High (> 400 ng/mL)	4 (12)	3.86 (2.98–9.44)		1.42 (1.13–1.76)		8.74 (1.40–12.10)	3.61 (1.26–6.51)
Etiology
HepB	29 (91)	4.34 (2.13–9.76)	0.58	1.55 (0.83–4.48)	0.53	3.23 (1.40–12.64)	0.14	1.62 (0.97–6.51)	0.08
Other	3 (9)	6.10 (4.31–6.79)		1.83 (1.29–2.98)		2.55 (1.64–2.94)	1.28 (0.96–1.34)
Tumor sizec
Small (≤ 3 cm)	22 (56)	3.40 (0.41–9.10)	0.03*	1.45 (0.16–3.07)	0.03*	2.53 (1.37–12.10)	0.003*	1.25 (0.72–6.51)	< 0.001*
Large (> 3 cm)	17 (44)	5.96 (2.13–9.76)		1.76 (0.83–4.48)		4.85 (1.40–12.64)	2.38 (1.26–6.28)
BCLC stage
Stage 0	9 (28)	3.46 (2.46–9.10)	0.16	1.45 (1.11–2.15)	0.28	2.55 (1.64–5.50)	0.04*	2.55 (1.64–5.50)	0.21
Stage A	23 (72)	5.45 (2.13–9.76)		1.61 (0.83–4.48)		3.70 (1.40–12.64)	2.02 (1.09–6.51)
Histological grade
1–2	18 (56)	3.44 (2.13–9.57)	0.07	1.49 (1.03–3.47)	0.06	2.83 (1.40–12.10)	0.004*	1.42 (0.96–6.51)	0.05
3–4	14 (44)	5.93 (2.31–9.76)		1.87 (0.83–4.48)		4.64 (2.55–12.64)	2.43 (1.27–6.28)
MVI
M0	9 (28)	5.90 (2.46–9.10)	0.84	1.83 (1.11–3.07)	0.64	2.85 (1.64–5.50)	0.07	1.28 (0.96–2.86)	0.01*
M1-2	23 (72)	4.34 (2.13–9.76)		1.55 (0.83–4.48)		3.57 (1.40–12.64)	2.10 (0.97–6.51)
Cirrhosis
Yes	22 (69)	3.89 (2.13–9.76)	0.08	1.53 (0.83–3.47)	0.13	3.17 (1.40–12.10)	0.91	1.52 (0.97–6.51)	0.93
No	10 (31)	6.45 (2.46–9.57)		1.80 (1.11–4.48)		2.92 (1.64–12.64)	1.56 (0.96–5.45)
PET findings
Both avid	13 (41)	7.91 (2.90–9.62)	NA	2.15 (1.45–4.48)	NA	5.50 (2.89–12.64)	NA	2.83 (1.34–5.45)	NA
Both nonavid	5 (16)	2.98 (2.46–4.31)		1.15 (1.11–1.29)		2.51 (1.40–2.85)	1.26 (1.20–1.34)
Only 11C-acetate-avid	10 (31)	4.73 (2.89–9.76)		1.72 (1.23–2.98)		2.56 (1.64–3.23)	1.29 (0.96–2.26)
Only 18F-FDG-avid	4 (12)	2.54 (2.13–3.39)		1.13 (0.83–1.26)		8.43 (2.80–12.10)	4.53 (1.61–6.51)

Survival Outcomes for Patients With Distinct PET/CT Findings

The patients were divided into subgroups based on their tracer uptake as observed in preoperative PET imaging. Among these 32 patients, 23 (72%) were ¹¹C-acetate-avid and 17 (53%) were ¹⁸F-FDG-avid. Kaplan–Meier survival curves for OS were generated and the log-rank test revealed that the presence of avid lesions in ¹¹C-acetate PET/CT was associated with a longer OS (p = 0.002; Figure 2A). In contrast, no statistical difference in OS was found between the patients with ¹⁸F-FDG-avid lesions and those without (p = 0.18; Figure 2B). Moreover, patients with only ¹¹C-acetate-avid lesions exhibited the longest OS, while those with only ¹⁸F-FDG uptake were the poorest (p < 0.0001; Figure 2C). For the 39 HCC lesions found in these 32 patients, no significant correlation of the SUV_max and TBR was observed between ¹¹C-acetate and ¹⁸F-FDG (SUV_max: ρ = 0.29, p = 0.07; TBR: ρ = 0.19, p = 0.24; Figure 2D,E). These results suggest that avid findings in ¹¹C-acetate PET/CT may be used to predict the postoperative outcome of HCC patients.

[IMAGE OMITTED. SEE PDF]

Association of ¹¹C-acetate and ¹⁸F-FDG Uptakes With CD3⁺ and CD8⁺ Cell Infiltration

The demarcation ranges of the TI and IM are shown in Figure 3A, and the representative immunohistochemical staining of CD3 and CD8 in TI and IM regions is presented in Figure 3B. The association of the densities of CD3⁺ and CD8⁺ cells with PET findings was further analyzed. Patients with ¹¹C-acetate-avid lesions showed higher densities of CD3⁺ cells in both TI region (p = 0.02) and IM region (p = 0.04), as well as CD8⁺ cells in IM region (p = 0.005) compared to those without (Table 2). In addition, the SUV_max of ¹¹C-acetate was positively associated with the density of CD3⁺ immune cells in TI region (ρ = 0.44, p = 0.01) and CD8⁺ cells in both TI region (ρ = 0.51, p = 0.003) and IM region (ρ = 0.57, p < 0.001), while TBR was also related to the densities of CD3⁺ cells in TI region (ρ = 0.38, p = 0.03) and CD8⁺ cells in IM region (ρ = 0.47, p = 0.007) (Figure 3C,D). However, the SUV_max and TBR of ¹⁸F-FDG only mildly correlated with the density of CD3⁺ cells in TI region (SUV_max: ρ = 0.39, p = 0.03; TBR: ρ = 0.35, p = 0.05) (Figure 3E,F). Taken together, these data indicated that the uptake of ¹¹C-acetate, rather than ¹⁸F-FDG, was closely correlated with an increased infiltration of both CD3⁺ and CD8⁺ cells.

[IMAGE OMITTED. SEE PDF]

TABLE 2 Comparison of CD3⁺ and CD8⁺ cell infiltration in patients with different PET tracer uptake.

	11C-acetate-avid	Non-11C-acetate-avid	p	18F-FDG-avid	Non-18F-FDG-avid	p
No. of patientsa	23 (72)	9 (28)	NA	17 (53)	15 (47)	NA
Median number of CD3+ cells in TI (cells per mm2)	107 (59.5–188.5)	26 (6–66)	0.02*	124 (62–198)	32 (5–100.5)	0.03*
Median number of CD3+ cells in IM (cells per mm2)	320 (190–481)	174 (133–222)	0.04*	320 (174–439)	213 (161–331.5)	0.33
Median number of CD8+ cells in TI (cells per mm2)	62 (33–143.5)	30 (10–69)	0.07	108 (13–131)	54 (35–79)	0.76
Median number of CD8+ cells in IM (cells per mm2)	347 (260–431)	125 (116–237)	0.005*	286 (148–384)	307 (159.5–396)	0.96

Efficacy of ¹¹C-acetate and ¹⁸F-FDG PET/CT in the Detection of Immune Infiltration

Kaplan–Meier survival curves for CD3⁺ and CD8⁺ cell infiltration in TI and IM regions were conducted separately and revealed associations between higher CD3⁺ and CD8⁺ cell infiltration in IM region and prolonged OS (CD3: p = 0.04; CD8: p = 0.01) (Figure 4A,B). According to the Immunoscore grouping method, the patients were classified as follows: eight (25%) in Im0, eight (25%) in Im1, nine (28%) in Im2, two (6%) in Im3, and five (16%) in Im4. Among these groups (Figure 4C, Table 3), the high Immunoscore group exhibited a statistical association with improved OS (p = 0.04) and showed higher SUV_max and TBR of ¹¹C-acetate (p = 0.005, p = 0.02, respectively), rather than ¹⁸F-FDG (all p > 0.05). By defining Im1-4 group as patients with immune infiltration (Figure 5) and Im0 group as patients without (Figure 6) based on the immune-desert phenotype criteria, it was observed that patients with immune infiltration had a longer OS compared with those Im0 ones (p = 0.02) (Figure 4D). In our patient-based analysis (Table 4), the sensitivity, specificity, and accuracy of ¹¹C-acetate PET/CT in detection of patients with immune infiltration were 88% (21 of 24), 75% (6 of 8), and 84% (27 of 32), respectively, whereas those of ¹⁸F-FDG PET/CT were 58% (14 of 24), 63% (5 of 8), and 59% (19 of 32), respectively (Table 4). The sensitivity of ¹¹C-acetate PET/CT was superior to that of ¹⁸F-FDG PET/CT (88% [21 of 24] vs. 58% [14 of 24], p = 0.05). However, the specificity of ¹¹C-acetate PET/CT was not significantly different from that of ¹⁸F-FDG (75% [6 of 8] vs. 63% [5 of 8], p > 0.99). Our analysis elucidated the correlation between immune infiltration and patient prognosis, and further suggested a superior efficacy of ¹¹C-acetate PET/CT in the detection of immune infiltration.

[IMAGE OMITTED. SEE PDF]

TABLE 3 Relationship of Immunoscore with ¹¹C-acetate and ¹⁸F-FDG uptake.

Variable	Im0	Im1	Im2	Im3/Im4	p
No. of patients	8 (25)	8 (25)	9 (28)	2 (6)/5 (16)	NA
11C-acetate
Avid	2 (6)	6 (19)	8 (25)	7 (22)	NA
Nonavid	6 (19)	2 (6)	1 (3)	0 (0)
18F-FDG
Avid	3 (9)	3 (9)	6 (19)	5 (16)	NA
Nonavid	5 (16)	5 (16)	3 (9)	2 (6)
Median 11C-acetate SUVmaxa	2.93 (2.13–4.31)	4.76 (2.52–9.76)	5.45 (3.12–9.11)	7.91 (4.39–9.57)	0.005*
Median 11C-acetate TBRa	1.24 (0.83–1.93)	1.68 (1.03–2.98)	1.61 (1.23–4.48)	1.97 (1.53–3.47)	0.02*
Median 18F-FDG SUVmaxa	3.04 (1.40–12.10)	2.78 (1.64–5.70)	3.70 (1.81–12.64)	4.85 (1.91–11.96)	0.35
Median 18F-FDG TBRa	1.80 (1.23–6.51)	1.28 (0.96–2.97)	2.02 (1.20–5.45)	2.38 (0.97–4.38)	0.22

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

TABLE 4 Diagnostic performance of ¹¹C-acetate and ¹⁸F-FDG PET/CT in assessment of immune infiltration.

Imaging modality	Sensitivity (%)	Specificity (%)	Accuracy (%)
11C-acetate PET/CT	88 (21/24) [69–96]	75 (6/8) [41–96]	84 (27/32) [75–92]
18F-FDG PET/CT	58 (14/24) [39–76]	63 (5/8) [31–86]	59 (19/32) [47–71]

Association of ¹¹C-acetate and ¹⁸F-FDG Uptakes With Granzyme B–Mediated Cytotoxic T-Cell Activity

The activation of cytotoxic CD8⁺ T cells in TI region was assessed through granzyme B expression, as illustrated in Figure 7A, depicting representative immunofluorescence staining displaying colocalization of granzyme B⁺ and CD8⁺ T cells. The median density of granzyme B⁺ CD8⁺ T cells in TI region was 7 cells/mm² (IQR: 0.5–17.5 cells/mm²) and the median percentage of granzyme B⁺ on CD8⁺ T cells was 7% (IQR: 0%–16%). Our analysis revealed positive correlations between both the density of granzyme B⁺ CD8⁺ T cells (ρ = 0.47, p = 0.007) and the percentage of granzyme B⁺ on CD8⁺ T cells (ρ = 0.42, p = 0.02) in TI region with the SUV_max of ¹¹C-acetate, but not the TBR (Figure 7B). However, no evidence of correlations was found with ¹⁸F-FDG (all p > 0.05) (Figure 7C). These findings indicated that the SUV_max of ¹¹C-acetate was positively associated with the cytotoxic activity of intratumoral CD8⁺ cells.

[IMAGE OMITTED. SEE PDF]

Discussion

Metabolic tumor imaging through PET/CT examinations provides accurate information regarding the biological behavior of tumor lesions and indicates tumor characteristics and patient prognosis preoperatively [16, 17]. In this study, we evaluated the relationship between preoperative ¹¹C-acetate and ¹⁸F-FDG uptake, immune infiltration, and prognosis of patient with HCC after surgical resection. Our study findings revealed that, although not related to clinical characteristics, patients with ¹¹C-acetate-avid lesions had a longer overall survival (p = 0.002), while no statistically significant difference was observed in the prognostic analysis of ¹⁸F-FDG (p = 0.18). We also demonstrated that patients with only ¹¹C-acetate-avid HCCs exhibited a longer OS compared to those with only ¹⁸F-FDG uptake (p < 0.0001). Consistently, higher SUV_max of ¹¹C-acetate was associated with increased CD3⁺ and CD8⁺ cell infiltration, heightened granzyme B-mediated cytotoxic T-cell activity, and a more robust Immunoscore which related to improved overall survival. Furthermore, ¹¹C-acetate PET/CT scans exhibited superior sensitivity in the detection of immune infiltration as compared to ¹⁸F-FDG PET/CT (88% [21 of 24] vs. 58% [14 of 24], respectively; p = 0.05), highlighting the value of ¹¹C-acetate PET/CT for screening HCCs with immune infiltration and provided a noninvasive immune evaluation method.

Glucose and acetate serve as energy sources and nutritional suppliers for tumor growth [18, 19]. In a prospective study involving 39 patients with HCC who underwent paired ¹¹C-acetate and ¹⁸F-FDG PET/CT, the results indicated that ¹⁸F-FDG PET/CT tended to detect poorly differentiated HCC lesions and ¹¹C-acetate PET/CT tended to detect well-differentiated types which suggested that patients with ¹¹C-acetate-avid findings may have a better prognosis [20], while Park et al. further demonstrated that patients with avid findings in ¹⁸F-FDG PET/CT had lower survival rates but no prognostic value was found for ¹¹C-acetate PET/CT [21]. Of note, patients recruited in these prospective studies encompassed all stages of HCC, even including patients with metastases [20, 21]. While our findings aligned with prior research that has shown a correlation between elevated ¹⁸F-FDG uptake, advanced tumor stage, and poorer differentiation [22], our study did not identify a statistical association between ¹⁸F-FDG and OS. This lack of association might be attributed to our small sample size and specific selection criteria, which included only early-stage HCC. Acetate uptake has been reported to be primarily regulated by ACSS2 expression, which is known to promote HCC progression under metabolic stress [18, 23]. However, Jung et al. [24] demonstrated that acetate-use-related tumors with increased ACSS2 levels exhibit decreased malignancy. They found that inhibiting ACSS2 reduced anabolic lipid metabolism but increased glycolysis, indicating a complementary relationship between acetate and glucose uptake in providing sufficient carbon sources for HCC progression [24]. In contrast, our study found poorly differentiated HCCs exhibited more ¹⁸F-FDG uptake but no significant difference in ¹¹C-acetate uptake between well differentiated and poorly differentiated HCCs, suggesting that enhanced glycolysis, rather than acetate uptake, may be associated with poorer differentiation, leading to poor prognosis in HCC. Considering the disparate prognostic value observed for ¹¹C-acetate and ¹⁸F-FDG PET/CT in early-stage HCC within our study, the absence of a correlation between ¹¹C-acetate and ¹⁸F-FDG uptake suggested that the metabolic patterns of HCC, as indicated by the uptake of metabolic tracers, may have distinctive implications for postoperative prognosis.

As no significant correlation was found between ¹¹C-acetate uptake and clinical characteristics associated with improved prognosis, we evaluated CD3⁺ and CD8⁺ cell infiltration in tumor microenvironment and demonstrated the relation to prolonged OS, aligning with the established roles of these cells in modulating tumor survival and influencing patient prognosis [9, 25]. Correlations between ¹⁸F-FDG parameters and immune-related markers have been explored in several tumors, yet inconsistent findings have given rise to a debated connection between ¹⁸F-FDG uptake and immune infiltration [13, 26, 27]. In this study, ¹⁸F-FDG was only associated with CD3⁺ cell infiltration in TI region suggesting a limited predictive value for immune infiltration in patients with early-stage HCC. Intriguingly, our results herein demonstrated a positive correlation between ¹¹C-acetate uptake and CD3⁺ and CD8⁺ cell infiltration, highlighting the potential immuno-predictive value of ¹¹C-acetate PET/CT. The Immunoscore serves as a comprehensive quantitative indicator reflecting the density of CD3⁺ and CD8⁺ immune cells in TI and IM regions of tumor [9, 15]. Consistent with the previously proposed prognostic value of the Immunoscore [28, 29], our data also demonstrated a higher Immunoscore positively correlated with prolonged OS and further revealed that HCC lesions with a higher Immunoscore exhibited higher SUV_max and TBR of ¹¹C-acetate rather than ¹⁸F-FDG, indicating the superiority of ¹¹C-acetate PET/CT in representing the overall immune status of tumor. Collectively, these findings suggested an inherent connection between immune infiltration and acetate accumulation, rather than glucose uptake, which aligned with previous research indicating that acetate can serve as a fuel source, bolstering T-cell proliferation [30].

Previous studies have stratified the immune status into three subtypes: the immune-infiltrated with CD8⁺ T cells inside the tumor; the immune-excluded with CD8⁺ T cells surrounding the tumor; and the immune-desert phenotype with CD8⁺ T cells absent in the tumor [31]. Consistent with the definition of immune-desert phenotype [31], our study found that the group with an Immunoscore of 0 (Im0), lacking CD8⁺ T cells infiltration in the tumor, had the poorest postoperative outcomes. Immune status is crucial, and predictive models based on noninvasive imaging have been developed [32, 33]. Tong et al. employed a ¹⁸F-FDG PET/CT radiomics model to assess CD8⁺ cell infiltration in patients with lung cancer, achieving good predictive performance [32]. However, our results indicated that ¹¹C-acetate PET/CT was more sensitive than ¹⁸F-FDG PET/CT for detecting HCC with immune infiltration. Moreover, the immune-infiltrated phenotype involves cytotoxic CD8⁺ T cells diffused in TI region of tumor, primarily acting through the perforin–granzyme-mediated mechanism [34]. Our study revealed a positive association between ¹¹C-acetate uptake and granzyme B⁺ CD8⁺ T-cell infiltration, which was consistent with acetate's potential to rescue impaired T-cell function [35]. Taken together, these findings suggested that ¹¹C-acetate PET/CT may be a potential predictive tool for immune infiltration and activation preoperatively.

Noninvasive prediction and monitoring of immunotherapy efficacy through PET imaging has gained prominence and ¹⁸F-FDG PET/CT has been explored for assessing the response of solid tumors to immunotherapy, primarily through metabolic response evaluation [36]. Given that immune infiltration status, including CD8⁺ T-cell infiltration and Immunoscore, has proven effective in predicting immunotherapy outcomes [10, 37], we suggested that ¹¹C-acetate PET/CT, with its positive association with immune infiltration, held the potential to extend its role in predicting immunotherapy effectiveness. However, large-scale prospective studies are required to validate this potential.

Several limitations of the present study should be addressed. First, a primary limitation was the small sample size, which may indicate a selection bias. Second, due to the limited size of tumor foci in patients with multifocal HCC lesions, we only collected samples from the largest lesion, thus restricting our analysis to a patient-based approach. To enhance accuracy, we matched the pathological features and immunohistochemistry results with tracer uptake data based on tumor size in surgical pathological records and PET/CT imaging. In addition, the findings from this study suggested the association between ¹¹C-acetate uptake and immune infiltration, which should be validated in future, larger-scale studies.

In conclusion, the metabolic tracers ¹¹C-acetate and ¹⁸F-FDG indicated different metabolic statuses of HCC and the uptake of ¹¹C-acetate emerged as an independent prognostic marker for patients with hepatocellular carcinoma. This uptake was closely linked to CD3⁺ and CD8⁺ cell infiltration, Immunoscore, and granzyme B–mediated T-cell activation implying that ¹¹C-acetate PET/CT may have the potential to preoperatively predict immune status and postoperative prognosis.

Author Contributions

Hao Xu: conceptualization, data curation, formal analysis, investigation, project administration, writing – original draft. Hao Wang: data curation, formal analysis, resources. Dong-Lang Jiang: investigation, methodology, validation. Yan-Fei Wu: formal analysis, resources. Sun-Zhe Xie: formal analysis, writing – review and editing. Ying-Han Su: formal analysis, writing – review and editing. Yi-Hui Guan: data curation, resources, writing – review and editing. Fang Xie: resources, writing – review and editing. Wen-Wei Zhu: funding acquisition, investigation, writing – review and editing. Lun-Xiu Qin: funding acquisition, investigation, project administration, writing – original draft.

Acknowledgments

We gratefully acknowledge the support and contributions of the individuals who participated in this study.

Ethics Statement

Approval of the research protocol by an Institutional Reviewer Board: The studies involving human participants were reviewed and approved by the ethics committee at Huashan Hospital, Fudan University (Registered No. HS-KY2020-729).

Informed Consent: All informed consent was obtained from the patients.

Registry and the Registration of the Study: N/A.

Animal Studies: N/A.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

All data obtained during the studies are available from the corresponding authors upon reasonable request.

References

H. Sung, J. Ferlay, R. L. Siegel, et al., “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 71 (2021): 209–249.

Q. Zhang, Y. Lou, J. Yang, et al., “Integrated Multiomic Analysis Reveals Comprehensive Tumour Heterogeneity and Novel Immunophenotypic Classification in Hepatocellular Carcinomas,” Gut 68 (2019): 2019–2031.

D. Hanahan, “Hallmarks of Cancer: New Dimensions,” Cancer Discovery 12 (2022): 31–46.

V. Goh, D. Sarker, S. Osmany, and G. J. Cook, “Functional Imaging Techniques in Hepatocellular Carcinoma,” European Journal of Nuclear Medicine and Molecular Imaging 39 (2012): 1070–1079.

S. A. Kwee, M. Tiirikainen, M. M. Sato, et al., “Transcriptomics Associates Molecular Features With (¹⁸)F‐Fluorocholine PET/CT Imaging Phenotype and Its Potential Relationship to Survival in Hepatocellular Carcinoma,” Cancer Research 79 (2019): 1696–1704.

T. T. Cheung, C. L. Ho, C. M. Lo, et al., “¹¹C‐Acetate and ¹⁸F‐FDG PET/CT for Clinical Staging and Selection of Patients With Hepatocellular Carcinoma for Liver Transplantation on the Basis of Milan Criteria: Surgeon's Perspective,” Journal of Nuclear Medicine 54 (2013): 192–200.

J. M. Llovet, F. Castet, M. Heikenwalder, et al., “Immunotherapies for Hepatocellular Carcinoma,” Nature Reviews. Clinical Oncology 19 (2022): 151–172.

C. Ma, A. H. Kesarwala, T. Eggert, et al., “NAFLD Causes Selective CD4(+) T Lymphocyte Loss and Promotes Hepatocarcinogenesis,” Nature 531 (2016): 253–257.

A. Gabrielson, Y. Wu, H. Wang, et al., “Intratumoral CD3 and CD8 T‐Cell Densities Associated With Relapse‐Free Survival in HCC,” Cancer Immunology Research 4 (2016): 419–430.

D. Bruni, H. K. Angell, and J. Galon, “The Immune Contexture and Immunoscore in Cancer Prognosis and Therapeutic Efficacy,” Nature Reviews. Cancer 20 (2020): 662–680.

K. C. Kao, S. Vilbois, C. H. Tsai, and P. C. Ho, “Metabolic Communication in the Tumour‐Immune Microenvironment,” Nature Cell Biology 24 (2022): 1574–1583.

W. Yu, Q. Lei, L. Yang, et al., “Contradictory Roles of Lipid Metabolism in Immune Response Within the Tumor Microenvironment,” Journal of Hematology & Oncology 14 (2021): 187.

Y. Wang, N. Zhao, Z. Wu, et al., “New Insight on the Correlation of Metabolic Status on (¹⁸)F‐FDG PET/CT With Immune Marker Expression in Patients With Non‐Small Cell Lung Cancer,” European Journal of Nuclear Medicine and Molecular Imaging 47 (2020): 1127–1136.

K. Kitajima, T. Higuchi, Y. Fujimoto, et al., “Relationship Between FDG‐PET and the Immune Microenvironment in Breast Cancer,” European Journal of Radiology 158 (2023): [eLocator: 110661].

D. Malka, A. Lièvre, T. André, J. Taïeb, M. Ducreux, and F. Bibeau, “Immune Scores in Colorectal Cancer: Where Are We?,” European Journal of Cancer 140 (2020): 105–118.

Y. Li, Y. Zhang, Q. Fang, et al., “Radiomics Analysis of [(¹⁸)F]FDG PET/CT for Microvascular Invasion and Prognosis Prediction in Very‐Early‐ and Early‐Stage Hepatocellular Carcinoma,” European Journal of Nuclear Medicine and Molecular Imaging 48 (2021): 2599–2614.

J. An, M. Oh, S. Y. Kim, et al., “PET‐Based Radiogenomics Supports mTOR Pathway Targeting for Hepatocellular Carcinoma,” Clinical Cancer Research 28 (2022): 1821–1831.

Z. T. Schug, B. Peck, D. T. Jones, et al., “Acetyl‐CoA Synthetase 2 Promotes Acetate Utilization and Maintains Cancer Cell Growth Under Metabolic Stress,” Cancer Cell 27 (2015): 57–71.

A. F. Abdel‐Wahab, W. Mahmoud, and R. M. Al‐Harizy, “Targeting Glucose Metabolism to Suppress Cancer Progression: Prospective of Anti‐Glycolytic Cancer Therapy,” Pharmacological Research 150 (2019): [eLocator: 104511].

C. L. Ho, S. C. Yu, and D. W. Yeung, “¹¹C‐Acetate PET Imaging in Hepatocellular Carcinoma and Other Liver Masses,” Journal of Nuclear Medicine 44 (2003): 213–221.

J. W. Park, J. H. Kim, S. K. Kim, et al., “A Prospective Evaluation of ¹⁸F‐FDG and ¹¹C‐Acetate PET/CT for Detection of Primary and Metastatic Hepatocellular Carcinoma,” Journal of Nuclear Medicine 49 (2008): 1912–1921.

A. Sabaté‐Llobera, J. Mestres‐Martí, G. Reynés‐Llompart, et al., “2‐[(¹⁸)F]FDG PET/CT as a Predictor of Microvascular Invasion and High Histological Grade in Patients With Hepatocellular Carcinoma,” Cancers (Basel) 13 (2021): [eLocator: 2554].

S. A. Comerford, Z. Huang, X. Du, et al., “Acetate Dependence of Tumors,” Cell 159 (2014): 1591–1602.

K. H. Jung, S. Lee, H. S. Kim, et al., “Acetyl‐CoA Synthetase 2 Contributes to a Better Prognosis for Liver Cancer by Switching Acetate‐Glucose Metabolism,” Experimental & Molecular Medicine 56 (2024): 721–733.

S. Itoh, T. Yoshizumi, K. Yugawa, et al., “Impact of Immune Response on Outcomes in Hepatocellular Carcinoma: Association With Vascular Formation,” Hepatology 72 (2020): 1987–1999.

M. Togo, T. Yokobori, K. Shimizu, et al., “Diagnostic Value of (¹⁸)F‐FDG‐PET to Predict the Tumour Immune Status Defined by Tumoural PD‐L1 and CD8(+) Tumour‐Infiltrating Lymphocytes in Oral Squamous Cell Carcinoma,” British Journal of Cancer 122 (2020): 1686–1694.

K. J. Na and H. Choi, “Tumor Metabolic Features Identified by (¹⁸)F‐FDG PET Correlate With Gene Networks of Immune Cell Microenvironment in Head and Neck Cancer,” Journal of Nuclear Medicine 59 (2018): 31–37.

Q. F. Chen, W. Li, P. H. Wu, L. J. Shen, and Z. L. Huang, “Significance of Tumor‐Infiltrating Immunocytes for Predicting Prognosis of Hepatitis B Virus‐Related Hepatocellular Carcinoma,” World Journal of Gastroenterology 25 (2019): 5266–5282.

B. Mlecnik, M. Van den Eynde, G. Bindea, et al., “Comprehensive Intrametastatic Immune Quantification and Major Impact of Immunoscore on Survival,” Journal of the National Cancer Institute 110 (2018): 97–108.

K. D. Miller, S. O'Connor, K. A. Pniewski, et al., “Acetate Acts as a Metabolic Immunomodulator by Bolstering T‐Cell Effector Function and Potentiating Antitumor Immunity in Breast Cancer,” Nature Cancer 4 (2023): 1491–1507.

M. Desbois, A. R. Udyavar, L. Ryner, et al., “Integrated Digital Pathology and Transcriptome Analysis Identifies Molecular Mediators of T‐Cell Exclusion in Ovarian Cancer,” Nature Communications 11 (2020): 5583.

H. Tong, J. Sun, J. Fang, et al., “A Machine Learning Model Based on PET/CT Radiomics and Clinical Characteristics Predicts Tumor Immune Profiles in Non‐Small Cell Lung Cancer: A Retrospective Multicohort Study,” Frontiers in Immunology 13 (2022): [eLocator: 859323].

W. Huang, Y. Jiang, W. Xiong, et al., “Noninvasive Imaging of the Tumor Immune Microenvironment Correlates With Response to Immunotherapy in Gastric Cancer,” Nature Communications 13 (2022): 5095.

P. Golstein and G. M. Griffiths, “An Early History of T Cell‐Mediated Cytotoxicity,” Nature Reviews. Immunology 18 (2018): 527–535.

J. Qiu, M. Villa, D. E. Sanin, et al., “Acetate Promotes T Cell Effector Function During Glucose Restriction,” Cell Reports 27 (2019): 2063–2074.

D. J. Hughes, M. Subesinghe, B. Taylor, et al., “(¹⁸)F FDG PET/CT and Novel Molecular Imaging for Directing Immunotherapy in Cancer,” Radiology 304 (2022): 246–264.

A. Mezheyeuski, M. Backman, J. Mattsson, et al., “An Immune Score Reflecting Pro‐ and Anti‐Tumoural Balance of Tumour Microenvironment Has Major Prognostic Impact and Predicts Immunotherapy Response in Solid Cancers,” eBioMedicine 88 (2023): [eLocator: 104452].