Liver Background Uptake of Prostate-specific Membrane Antigen–targeted PET Radiotracers: A Systematic Review and Meta-Analysis

Abstract

Purpose

To perform a systematic review and meta-analysis of liver background uptake across four prostate-specific membrane antigen (PSMA)–targeting PET radiotracers (fluorine 18 [¹⁸F]-piflufolastat, gallium 68 [⁶⁸Ga]-PSMA-11, ¹⁸F-flotufolastat, and ¹⁸F-PSMA-1007) and evaluate the potential impact on patient selection for PSMA-targeted therapy.

Materials and Methods

A comprehensive literature search was conducted in PubMed, Embase, Web of Science, and Cochrane through May 12, 2025, to identify human studies reporting quantitative liver background uptake for the four PSMA-targeted PET radiotracers of interest. For each eligible study, liver background uptake was extracted as standardized uptake values (SUVs) and summarized and compared by radiotracer using random effects models with robust variance estimation.

Results

Among 652 unique records, 17 studies with 1497 total patients met inclusion criteria and reported liver background SUVs. Summary mean liver background SUV was 5.0 (95% CI: 3.6, 6.3) for ¹⁸F-piflufolastat, 5.1 (95% CI: 4.3, 5.9) for ⁶⁸Ga-PSMA-11 (P = .579 compared with ¹⁸F-piflufolastat), 7.2 (95% CI: 6.3, 8.2) for ¹⁸F-flotufolastat (P = .005), and 12.1 (95% CI: 11.4, 12.9) for ¹⁸F-PSMA-1007 (P < .001).

Conclusion

Liver background uptake differed across PSMA-targeted PET radiotracers and may influence patient eligibility for PSMA-targeted therapy when liver activity is used as a reference for defining PSMA-positive metastatic prostate cancer.

Keywords: PET, Genital/Reproductive, Prostate, ⁶⁸Ga-PSMA-11, ¹⁸F-piflufolastat, ¹⁸F-flotufolastat, ¹⁸F-PSMA-1007, PSMA-targeted PET, PSMA-targeted Therapy

Supplemental material is available for this article.

© RSNA, 2026

Summary

Higher liver background uptake was observed on prostate-specific membrane antigen (PSMA)–targeted PET imaging with fluorine 18 (¹⁸F)-flotufolastat and ¹⁸F-PSMA-1007 compared with ¹⁸F-piflufolastat and gallium 68 (⁶⁸Ga)-PSMA-11, which may influence patient eligibility for PSMA-targeted therapy.

Key Points

• ■ In this systematic review and meta-analysis of 17 studies comparing liver background uptake across prostate-specific membrane antigen (PSMA)–targeted PET radiotracers, mean liver background standardized uptake value was lowest for fluorine 18 (¹⁸F)-piflufolastat (5.0) and gallium 68 (⁶⁸Ga)-PSMA-11 (5.1), higher for ¹⁸F-flotufolastat (7.2), and highest for ¹⁸F-PSMA-1007 (12.1).

• ■ Differences in liver background uptake among PSMA-targeted PET radiotracers may influence patient eligibility for PSMA-targeted therapy when liver activity is used as the reference to classify PSMA-positive prostate cancer.

Introduction

Prostate-specific membrane antigen (PSMA) is a transmembrane glutamate carboxypeptidase known to be overexpressed in most high-grade prostate cancers (1). PSMA-targeted therapy with lutetium 177 vipivotide tetraxetan (Pluvicto) has demonstrated efficacy in patients with PSMA-positive metastatic prostate cancer, leading to its use in both posttaxane and pretaxane clinical scenarios (2–4). Several additional PSMA-targeted therapies are also currently being evaluated in trials.

PSMA-targeted imaging is used to identify PSMA-positive prostate cancer before PSMA-targeted therapy. For example, the VISION trial used predefined PSMA-targeted imaging criteria to select patients for lutetium 177 vipivotide tetraxetan therapy, which have become known as the VISION criteria (2,5). The VISION criteria specify that a positive PSMA-targeted PET scan includes at least one lesion with PSMA-avidity greater than background liver avidity and no soft-tissue lesions greater than or equal to 1 cm in short-axis diameter (or lymph nodes ≥ 2.5 cm in short-axis diameter) with PSMA-avidity lower than background liver avidity (2,5). Thus, the VISION criteria use background liver avidity as a key reference when determining eligibility for PSMA-targeted therapy.

In the VISION trial, PSMA-targeted PET scans used to select patients for PSMA-targeted therapy were performed with the [⁶⁸Ga]Glu-Urea-Lys(Ahx)-HBED-CC radiotracer (also known as gallium 68 [⁶⁸Ga]-PSMA-11) (2), which is marketed as Illuccix, Gozellix, and Locametz (6–8). Three fluorine 18 (¹⁸F)–labeled radiotracers are also widely used for PSMA-targeted PET imaging, including: ¹⁸F-piflufolastat, previously known as ¹⁸F-DCFPyL, and marketed in the United States as Pylarify and in the European Union as Pylclari; ¹⁸F-flotufolastat, previously known as ¹⁸F-rhPSMA-7.3 and marketed as Posluma (7–10); and ¹⁸F-PSMA-1007, marketed in the European Union as Radelumin (11,12). There are no controlled clinical trials comparing these four PSMA-targeted PET radiotracers for selecting patients for PSMA-targeted therapy; therefore, European Society for Medical Oncology (13) and National Comprehensive Cancer Network (14) guidelines recommend that any regionally approved PSMA-targeted imaging agent may be used.

Although PSMA-targeted PET radiotracers share substantial similarities in biodistribution, lesional uptake, and background avidity (15), differences in chemical structure can result in variation in relative renal and hepatic excretion (9). Differences in hepatic background uptake may therefore influence which patients meet imaging-based eligibility criteria for PSMA-targeted therapy (9). This systematic review and meta-analysis aims to evaluate and compare liver background uptake across commonly used PSMA-targeting PET radiotracers to inform their applicability for selecting patients for PSMA-targeted therapy.

Materials and Methods

This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The objective was to evaluate and compare liver background uptake across ¹⁸F-piflufolastat, ⁶⁸Ga-PSMA-11, ¹⁸F-flotufolastat, and ¹⁸F-PSMA-1007. Institutional review board approval was not required because this study analyzed previously published data and did not involve collection of new participant data.

Search Strategy

A comprehensive literature search was conducted by one author (J.W.S., with 40 years of literature search experience) across multiple databases, including PubMed, Embase (OVID version), Web of Science, the Cochrane Central Register of Controlled Trials (CENTRAL), and Academic Search Premier, through May 12, 2025. The search strategy was designed to identify human studies evaluating PSMA-targeted imaging with quantification of radiotracer background liver uptake. To ensure the broadest capture of relevant studies, a combination of Medical Subject Headings and free-text keywords was used. The detailed search strategy is provided in Appendix S1.

Study Selection

Abstracts of the retrieved studies were screened by two authors (G.A.U., a clinical scientist with 16 years of academic experience, and A.S.P., a clinical scientist with 5 years of academic experience), and full texts of potentially eligible studies were reviewed for inclusion based on predefined eligibility criteria: PSMA-targeted imaging in human participants and reported quantification of background liver radiotracer uptake using standardized uptake values (SUVs). Studies were excluded if they did not evaluate one of the four PSMA-targeted PET radiotracers of interest (¹⁸F-piflufolastat, ⁶⁸Ga-PSMA-11, ¹⁸F-flotufolastat, and ¹⁸F-PSMA-1007), were nonhuman studies, or lacked quantitative SUV data for background liver radiotracer uptake.

Data Extraction

For the studies reporting background liver uptake for one or more of the four PSMA-targeted radiotracers of interest, data were extracted by one author (G.A.U., a clinical scientist with 16 years of academic experience), including (a) study characteristics (author, year, study design, sample size); (b) PSMA-targeted PET/CT details—radiotracer(s) evaluated, administered dose, imaging protocol, volume of interest (VOI) measurement method, and uptake metrics (mean SUV [SUVmean], SUV normalized to lean body mass, maximum SUV [SUVmax], and peak SUV [SUVpeak]), when available; and (c) quantitative measurements of background liver uptake. All data were cross-checked to ensure consistency.

Data Analysis

SUVmean was chosen as the primary metric for liver background SUV as all but one included study reported SUVmean. A summary mean liver background SUVmean was calculated for each radiotracer using random effects models with robust variance estimation to account for inclusion of multiple estimates per study and the small number of studies per radiotracer. When the SD was not reported, it was estimated from the IQR (ie, as the average of [P25 – mean]/−0.675 and [P75 – mean]/0.675), if available, or imputed using the average SD for all studies by radiotracer. τ² was calculated to provide an estimate of variance of the true effects across studies. To investigate the influence of individual studies, the analyses were repeated leaving one study out at a time and reporting the range of resulting estimates. Finally, although publication bias was not expected for reporting of SUVmean values, it was investigated using the Egger test. StataSE18 (StataCorp, stata.com) was used for the analysis.

Results

Literature Search Outcomes

Figure 1 shows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart illustrating the study identification and selection. The search across all electronic databases yielded 1330 total records. After excluding 678 duplicates, 652 abstracts from Pubmed (n = 363), Embase (n = 151), Web of Science (n = 116), Cochrane Library (n = 11), and Academic Search Premier (n = 11) underwent initial screening. After excluding 610 records that did not meet study criteria for PSMA-targeted imaging of humans with reported liver findings, 42 full-text articles were assessed for eligibility. Twenty-five studies were excluded for lack of background liver radiotracer uptake measurements, yielding a final total of 17 studies, including 1497 unique patients, that were included in this systematic review and meta-analysis (16–32).

Figure 1:

Preferred Reporting Items for Systematic Review and Meta-Analyses flowchart shows study identification and selection. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses, PSMA = prostate-specific membrane antigen.

Study Characteristics

Table 1 provides detailed characteristics of the included studies. The 17 studies were published between June 2017 and January 2025. Study designs included 11 retrospective studies, one post hoc analysis of a prior prospective study, and five prospective studies. Data were reported for ¹⁸F-piflufolastat in six studies, ⁶⁸Ga-PSMA-11 in eight studies, ¹⁸F-flotufolastat in four studies, and ¹⁸F-PSMA-1007 in seven studies. Six studies evaluated two PSMA-targeted agents, and one study evaluated three agents.

Table 1:

Characteristics of the 17 Studies Included in this Systematic Review and Meta-Analysis

First Author and Reference No.	Publication Year	Study Design	PET Scanners Used in Study	PSMA-targeted Imaging Agents Evaluated
				18F-piflufolastat	68Ga-PSMA-11	18F-flotufolastat	18F-PSMA-1007
Li (16)	2017	Retrospective	Discovery RX or Biograph mCT	Yes	No	No	No
Giesel (17)	2018	Prospective	Biograph mCT 40	Yes	No	No	Yes
Rahbar (18)	2018	Retrospective	Biograph mCT	No	No	No	Yes
Jansen (19)	2019	Retrospective	Ingenuity TF, Gemini TOF, Biograph 40, or Discovery 710	Yes	Yes	No	No
Oh (20)	2020	Retrospective	Biograph mCT Flow	No	No	Yes	No
Werner (21)	2020	Retrospective	Discovery RX 64 or Biograph mCT 128	Yes	No	No	No
Rahbar (22)	2020	Prospective	Biograph mCT	No	No	No	Yes
Emmett (23)	2021	Prospective	Ingenuity TOF 64	No	Yes	No	No
Hoberück (24)	2021	Retrospective	Biograph Vision 600	No	Yes	No	Yes
Knorr (25)	2022	Retrospective	Biograph mCT Flow	No	No	Yes	No
Langbein (26)	2022	Retrospective	Biograph mCT Flow	No	No	Yes	No
Pattison (27)	2022	Prospective	Biograph mCT	No	Yes	No	No
Sharma (28)	2022	Prospective	Discovery MIDR or Discovery 710	No	Yes	No	Yes
Penny (29)	2024	Post hoc analysis	Not stated	No	No	Yes	No
Popescu (30)	2024	Retrospective	Discovery MI, 690, VCT, and ST, or Biograph mCT Flow	No	Yes	No	Yes
Yadav (31)	2024	Retrospective	Not stated	Yes	Yes	No	No
Heilinger (32)	2025	Retrospective	Biograph mCT 128 Flow-Edge or 16 TruePoint	Yes	Yes	No	Yes

Note.—PSMA = prostate-specific membrane antigen.

Liver Background Uptake Measurements by Radiotracer

A summary of extracted data for each of the four PSMA-targeted PET radiotracers is provided in Table 2. For ¹⁸F-piflufolastat, liver background uptake data were available from 222 patients. For ⁶⁸Ga-PSMA-11, data were available from 339 patents. For ¹⁸F-flotufolastat, data were available from a total of 865 patients. For ¹⁸F-PSMA-1007, data were available from 210 patients, including 40 patients imaged at two separate time points. There were a total of 1497 unique patients, with some patients undergoing more than one scan.

Table 2:

Summary of Included Studies Reporting Liver Background SUV Metrics by PSMA PET Radiotracer

Dose (mCi)	No. of Patients	VOI for SUV Measurement	SUVmean	SULmean	SUVpeak	SUVmax	First Author and Reference No.
18F-Piflufolastat (Pylarify, Pylclari)
<9	64	3-cm sphere	5.1 ± 0.7	3.8 ± 0.6			Li (16)
6.5–7.0	12	2-cm sphere				16.9	Giesel (17)
8.1–8.6	50	3-cm sphere	5.9 ± 1.3			6.8 ± 1.4	Jansen (19)
<9	50	3-cm sphere	3.5 ± 0.7	2.6 ± 0.5			Werner (21)
7.0–11.7	33	Not stated	4.3 (IQR, 3.5–5.0)				Yadav (31)
4.4–11.1	13	Sphere	6.2 ± 1.6				Heilinger (32)
Summary mean SUVmean			5.0 (95% CI: 3.6, 6.3), τ2 = 1.11
68Ga-PSMA-11 (Illuccix, Gozellix, and Locametz)
3.8	87	3-cm sphere	4.8 ± 1.6			6.4 ± 1.9	Jansen (19)
0.54 mCi/kg	14	Not stated	5.4 ± 1.3				Emmett (23)
3.0–4.3	46	3-cm sphere	7.0				Hoberück (24)
2.7–4.1	50	1-cm sphere	4.4 ± 1.3				Pattison (27)
1.5	4	3.5-cm sphere	4.9 ± 1.9			11.1 ± 4.2	Sharma (28)
3.3–4.2	50	Sphere	5.4 ± 1.7				Popescu (30)
4.9–11.4	47	Not stated	4.2 (IQR, 3.3–5.0)				Yadav (31)
1.7–5.9	41	Sphere	4.6 ± 1.0				Heilinger (32)
Summary mean SUVmean			5.1 (95% CI: 4.3, 5.9), τ2 = 0.80	P = .579 compared with 18F-Piflufolastat
18F-Flotufolastat (Posluma)
6.0–8.0	65	2–3-cm sphere	7.9 ± 2.5				Oh (20)
8.1–10.0	74	2–3-cm sphere	8.3 ± 2.6				Oh (20)
10.1–12	63	2–3-cm sphere	7.9 ± 2.3				Oh (20)
6.4–11.3	33	Sphere	7.3 ± 2.2				Knorr (25)
0.11 mCi/kg	84	3-cm sphere	7.2 (IQR, 6.2–8.2)				Langbein (26)
5.8–10.7	546	6-cm sphere	6.7 (IQR, 5.8–7.6)		8.2 (IQR, 7.1–9.3)		Penny (29)
Summary SUVmean			7.2 (95% CI 6.3–8.2), τ2 = 0.20	P = .005 compared with 18F-Piflufolastat
18F-PSMA-1007 (Radelumin)
6.5–7.0	12	2-cm sphere				9.1	Giesel (17)
7.9–10.4	40	Circular	9.7 (at 60 min: IQR, 8.1–12.2)				Rahbar (18)
7.9–10.4	40	Circular	12.0 (at 120 min: IQR, 10.6–14.1)				Rahbar (18)
6.1–11.1	40	Circular	11.2 (at 120 min)				Rahbar (22)
3.3–4.7	46	3-cm sphere	13.0				Hoberück (24)
7.2	4	3.5-cm sphere	12.2 ± 1.0			20.5 ± 2.8	Sharma (28)
5.6–7.2	50	Sphere	11.7 ± 3.9				Popescu (30)
4.4–11.1	18	Sphere	12.9 ± 3.2				Heilinger (32)
Summary Mean SUVmean			12.1 (95% CI: 11.4, 12.9),*; τ2 = 0.28	P < .001 compared with 18F-Piflufolastat

Note.—SUVmean is shown with mean ± SD, except when noted as IQR. PSMA = prostate-specific membrane antigen, SUL = SUV normalized to lean body mass, SUV = standardized uptake value, VOI = volume of interest.

^*Does not include the value from Rahbar et al (18) estimated at 60 minutes.

In most studies, liver background uptake was measured using spherical VOIs ranging from 1 to 6 cm, although VOI size was not disclosed in four studies. Two studies by the same first author reported a circular VOI, and two studies did not report the VOI of measurement. All but one study reported liver background uptake as SUVmean. Most studies reported mean SUVmean values with SDs, whereas others reported mean SUVmean values with IQRs. Some studies also reported SUV normalized to lean body mass, SUVpeak, or SUVmax.

Mean liver background SUVmean values ranged from 3.5 to 6.2 for ¹⁸F-piflufolastat, 4.2 to 7.0 for ⁶⁸Ga-PSMA-11, 6.7 to 8.3 for ¹⁸F-flotufolastat, and 9.7 to 13.0 for ¹⁸F-PSMA-1007. Summary background SUVmean values were 5.0 (95% CI: 3.6, 6.3) for ¹⁸F-piflufolastat (n = 210 scans), 5.1 (95% CI: 4.3, 5.9) for ⁶⁸Ga-PSMA-11 (n = 339 scans; P = .579 compared with ¹⁸F-piflufolastat), 7.2 (95% CI: 6.3, 8.2) for ¹⁸F-flotufolastat (n = 865 scans; P = .005), and 12.1 (95% CI: 11.4, 12.9) for ¹⁸F-PSMA-1007 (n = 238 scans; P < .001). Figure 2 presents the data on liver background uptake for the four radiotracers with summary SUVmeans and standard errors. Leaving one study out one at a time and repeating the analyses resulted in similar estimates (¹⁸F-piflufolastat: range of SUVmean values = 4.7–5.3; ⁶⁸Ga-PSMA-11: range = 4.8–5.2, range of P values compared with ¹⁸F-piflufolastat = .40–.83; ¹⁸F-flotufolastat: 7.0–7.5, .018–.006; and ¹⁸F-PSMA-1007: 11.9–12.3, <.001–<.001). Finally, no evidence of publication bias was observed using the Egger test.

Figure 2:

Summary SUVmeans and standard errors for liver background uptake of the four PSMA-targeted imaging radiotracers. Error bars represent the standard error of each summary SUVmean. P values are given for each tracer compared with ¹⁸F-Piflufolastat. PSMA = prostate-specific membrane antigen, SUV = standardized uptake value.

Discussion

European Society for Medical Oncology and National Comprehensive Cancer Network guidelines currently state that multiple U.S. Food and Drug Administration–approved PSMA-targeted imaging agents can be equivalently used to select appropriate patients for PSMA-targeted therapy. However, liver background uptake differs across PSMA-targeted PET radiotracers, which may affect patient eligibility when liver uptake is used as a reference standard, as in the VISION trial. In this systematic review and meta-analysis, weighted mean liver background SUVmean values were similar for ¹⁸F-piflufolastat (5.0; n = 210 scans) and ⁶⁸Ga-PSMA-11 (5.1; n = 339 scans), but higher for ¹⁸F-flotufolastat (7.2; n = 865 scans) and highest for ¹⁸F-PSMA-1007 (12.1; n = 238 scans). Thus, if liver background is used to define PSMA-positive and PSMA-negative lesions, the choice of radiotracer may influence which patients are deemed eligible for PSMA-targeted therapy.

Although the number of patients that may be selected for PSMA-targeted therapy when evaluated by one radiotracer, but not selected when evaluated by another is not known, this variation in liver background by more than 100% between the two extremes (lowest liver background with ¹⁸F-piflufolastat, highest liver background with ¹⁸F-PSMA-1007) suggests that number may be substantial.

The VISION criteria define lesions as PSMA-positive or PSMA-negative based on comparison to liver background SUV by visual assessment, thus it is not the SUV per se that selects patients for PSMA-targeted therapy. Nevertheless, a substantial difference in liver background SUV will influence whether a lesion has tracer avidity greater than the liver background.

There were differences in methodology within the included studies which deserve mention. The administered radiotracer dose was nonuniform, ranging from 1.5 mCi to 12 mCi. SUVs, which are semiquantitative measurements of radiotracer uptake within a VOI relative to the administered dose, would be expected to remain comparable across this range. The most common VOI used for liver background measurements was a sphere. Sphere size ranged from 1 to 6 cm in studies that reported this parameter. This variability would not be expected to substantially affect SUVmeans, which were most commonly reported, but may influence SUVmaxes, which were reported in three studies. There were also differences in study sample sizes, although most studies included sufficient numbers of patients to yield relatively low SDs of SUV measurements. Differences in radiotracer uptake time and PET/CT scanner characteristics may also affect SUV measurements, as demonstrated in Rahbar et al (18), where a 120-minute uptake time for ¹⁸F-PSMA-1007 yielded a statistically significantly higher mean liver background SUVmean of 12.0 compared with 9.7 at a 60-minute uptake time.

This analysis was not a comprehensive review of all PSMA-targeted imaging molecules. Multiple other PSMA-targeted PET radiotracers that have been used in clinical trials or in clinical practice in other regions, such as ¹⁸F-CTT1057 or copper 64-SAR-bisPSMA, were not included.

There are additional considerations regarding SUVs. In most included studies, liver uptake was measured using spherical VOIs ranging from 1 to 6 cm³. VOI placement could affect SUV measurements, and whole-liver contouring may yield different values. In addition, lesional SUVs may differ between radiotracers; for example, ¹⁸F-piflufolastat may demonstrate higher lesional SUVs than ⁶⁸Ga-PSMA-11 (33), which could also influence patient selection for PSMA-targeted therapy. Alternative approaches to tumor-to-background assessment have been proposed, such as the PSMA PET tumor-to-salivary gland ratio (34), although data on PSMA PET tumor-to-salivary gland ratio remain limited.

The VISION trial included patients with metastatic prostate cancer who had completed chemotherapy. Nearly identical criteria have now been used for patients who are chemotherapy naive, such as in the PSMAfore trial. Selection of these patients in clinical practice for PSMA-targeted therapy could be similarly affected by difference in liver background between PSMA-targeted tracers (4).

Some limitations to the data collection in this systematic review and meta-analysis must be acknowledged. These include the inherent limitations of pooling data from heterogeneous studies and the potential for publication bias. Other factors may affect liver background SUV, such as extent of disease burden. A formal quality appraisal of each study, such as QUADAS-2 (35), was not performed. Indeed, many included studies were observational or post hoc analyses, with potential biases.

In conclusion, this systematic review and meta-analysis demonstrates that liver background uptake of ¹⁸F-piflufolastat and ⁶⁸Ga-PSMA-11 are nearly equivalent, whereas liver background uptake is higher for ¹⁸F-flotufolastat and substantially higher for ¹⁸F-PSMA-1007. Although current European Society for Medical Oncology and National Comprehensive Cancer Network guidelines state that multiple PSMA-targeted PET radiotracers can be used to select appropriate patients for PSMA-targeted therapy, differences in liver background uptake between these radiotracers may result in differences in patient eligibility. Specifically, ¹⁸F-piflufolastat and ⁶⁸Ga-PSMA-11 may select similar patients when applying VISION criteria, whereas ¹⁸F-flotufolastat and ¹⁸F-PSMA-1007 could select fewer eligible patients due to higher liver background uptake. Future research will refine data on background liver SUVs for PSMA-targeted radiotracers discussed here.

Disclosures of conflicts of interest

Please see ICMJE form(s) for author conflicts of interest. These have been provided as supplemental materials.