T1rho MRI as a quantitative biomarker of radiation-induced liver disease in normal and fibrotic rat models

Introduction

Hepatocellular carcinoma (HCC) is a highly prevalent malignancy worldwide. Although surgical resection offers the best chance for long-term survival in early-stage cases, it is feasible for only about 20% of patients at diagnosis due to advanced disease or impaired liver function from cirrhosis (1,2). Most patients rely on alternative treatments, including local ablation, transarterial chemoembolization, radiation therapy (RT), liver transplantation, and systemic therapy (3-5). As a moderately radiosensitive tumor, HCC has seen expanded application of RT, especially stereotactic body radiotherapy (SBRT), enabled by advances in technology (3-5). However, high-dose radiation may also affect surrounding normal liver tissue, leading to radiation-induced liver disease (RILD) (6,7). Thus, it’s necessary to reach a delicate balance between achieving therapeutic tumor doses and minimizing the risk of RILD.

Current imaging approaches to evaluate post-radiotherapy liver injury are largely qualitative. Hepatobiliary phase imaging with gadoxetic acid-enhanced magnetic resonance imaging (EOB-MRI) is commonly used, and non-uptake areas have been associated with subsequent liver dysfunction (8). However, these techniques lack quantitative sensitivity and may not capture subtle or early pathophysiological changes. In this context, quantitative magnetic resonance imaging (MRI) biomarkers are attracting increasing attention. Among them, T1rho (T1ρ) relaxation, which reflects spin-lattice relaxation in the rotating frame under transverse radiofrequency (RF) fields, has emerged as a promising noninvasive biomarker for liver injury in both animal (9-14) and human studies (15-19). As reported by Allkemper et al., the T1rho value had a significant correlation with patients’ Child-Pugh staging (16). For patients with chronic liver diseases, Takayama et al. established two key findings: liver T1rho values correlated positively with serum total bilirubin, direct bilirubin, and indocyanine green (ICG-R15), and correlated negatively with serum albumin and γ-glutamyl transpeptidase—all correlations were statistically significant (17). Most recently, Peng et al. reported that mild chronic hepatitis B patients showed elevated T1rho versus healthy controls. T1rho value exhibited strong positive correlations with histopathological severity and moderate correlations with liver function parameters (19). After pegylated interferon-α (PEG-IFNα) therapy, liver T1rho decreased significantly compared to pre-treatment value (19).

Animal models are essential for mechanistic studies and biomarker validation. Acute RT-alone models primarily reproduce radiation injury without significant fibrosis, while combined thioacetamide (TAA) plus RT models better simulate the clinical scenario of patients with underlying chronic liver disease who undergo radiotherapy and subsequently develop complications. By comparing these two models, it is possible to delineate how pre-existing fibrosis modifies the trajectory of RILD and to evaluate whether T1rho imaging can noninvasively capture these differences.

The present study aimed to investigate longitudinal changes in liver T1rho in rat models of RILD with and without background fibrosis, and to correlate T1rho with histological markers of fibrosis and inflammation. We hypothesized that T1rho would show a time-dependent increase after irradiation, that pre-existing fibrosis would amplify these changes, and that T1rho values would strongly correlate with histological indices of collagen deposition. This work provides preclinical evidence supporting T1rho MRI as a sensitive, quantitative biomarker for assessing RILD severity, with potential implications for early detection and individualized risk stratification in patients receiving liver radiotherapy. We present this article in accordance with the ARRIVE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2076/rc).

Methods

Animals and radiation dose selection

All animal experiments were performed under a project license (No. 2019-826) granted by the Animal Experimentation Ethics Committee of the First Affiliated Hospital, Zhejiang University School of Medicine, in compliance with institutional guidelines for the care and use of animals. Under a 12-h light/dark cycle and at a temperature of 25 ℃, male Sprague-Dawley (SD) rats (200 g) were housed, and they were provided with free access to food and water. All animals were acclimatized for one week prior to experimentation. Bodyweight was recorded daily. Based on the QUANTEC liver report, which recommends <13–20 Gy in 3–6 fractions for SBRT to avoid RILD (20), and our pilot data, 20 and 25 Gy were selected to irradiate the right liver lobe in rats with or without liver fibrosis. A protocol was prepared before the study without registration.

Animal model building

A total of 87 rats were assigned to two experimental groups. The RT group (n=41) modeled RILD without fibrosis, with 9 rats serving as normal controls and 32 rats receiving a single fraction of 25 Gy targeted to the right hepatic lobe. The TAA + RT group (n=46) modeled RILD with pre-existing fibrosis induced by intraperitoneal TAA injections (200 mg/kg, twice weekly for 6 weeks); after fibrosis induction, 38 rats received 20 Gy right-lobe irradiation (Figure 1). Group size differences were based on pilot data indicating higher mortality in fibrotic livers, thereby ensuring sufficient statistical power. A partial-liver irradiation approach was adopted to minimize exposure to adjacent organs (21). Radiotherapy was performed using a clinical medical linear accelerator (Varian TrueBeam, Varian Medical Systems, Palo Alto, CA, USA) instead of a small-animal irradiator. A 6 MV photon beam was delivered at a dose rate of approximately 400 MU/min. CT simulation (slice thickness 1 mm) was conducted with rats immobilized in the supine position. The right hepatic lobe was contoured as the target, and treatment plans were generated using a three-dimensional conformal radiotherapy (3D-CRT) technique with parallel-opposed or multi-field beam arrangements. Multileaf collimators (MLCs) were adjusted to shield the contralateral lobe and adjacent structures whenever feasible. Because the left hepatic lobe received low-dose scatter irradiation and could be secondarily affected by systemic and paracrine inflammatory responses triggered by right-lobe irradiation, it did not serve as a reliable internal control. Therefore, no intra-animal comparison between the right and left lobes was performed.

Figure 1 Image-guided irradiation plan for rat liver, and experimental design and timeline for the two rat models of RILD. (A) CT simulation imaging showing the transverse view of the rat liver. A vertical line through the spinal cord (red line) is used to approximately define the right liver (pink filling). (B-D) Dose distribution of the radiation plan targeting the right liver displayed in the transverse plane, the coronal plane, and in the sagittal plane, respectively. (E) RT group: 9 rats served as normal controls and underwent histological analysis without radiation. The remaining 32 rats received image-guided RT with a 25-Gy single dose targeting the right liver, with histological evaluations performed at 2, 4, 8, and 12 weeks post-RT. (F) TAA + RT group: liver fibrosis was induced with intraperitoneal TAA injections (200 mg/kg, twice weekly for 6 weeks) in 46 rats. After 6 weeks, 38 rats received a 20-Gy dose of image-guided RT to the right liver, and histological evaluations were conducted at 6 weeks post-TAA + 2 weeks post-RT, 6 weeks post-TAA + 4 weeks post-RT, 6 weeks post-TAA + 8 weeks post-RT, and 6 weeks post-TAA + 12 weeks post-RT. CT, computed tomography; MRI, magnetic resonance imaging; RILD, radiation-induced liver disease; RT, radiation therapy; SD, Sprague-Dawley; TAA, thioacetamide.

MRI protocol

For MRI data acquisition, a 3T Prisma scanner (Siemens Healthineers) was utilized. To cover the liver, a 15-channel knee coil was employed as the signal receiver, and the scanner’s in-built body coil was used as the signal transmitter. A standard axial T2-weighted TSE sequence was utilized to conduct liver anatomical imaging that covers the whole liver. To perform T1rho imaging, representative axial slices passing through the liver were selected, taking the axial T2-weighted image as the reference. The T1rho images were acquired using a single slice, inversion pulse-prepared (fluid-attenuated) T1rho pre-encoded turbo spin-echo (TSE) pulse sequence. For T1rho mapping, the spin-lock frequency was configured at 340 Hz, and spin-lock times (TSL) including 0, 10, 30, and 50 ms were used. The echo time (TE) and repetition time (TR) for TSE acquisition were set at 8.8 and 2,000 ms, respectively. The field of view (FOV) was 100×100 mm², with a voxel size of 0.4×0.4×2 mm³. The flip angle was 180 degrees. Each image of single TSL was acquired four times to improve signal-to-noise ratio (SNR) and a total imaging time of 8 min for four images.

Image analysis

T1rho maps were computed on a pixel-by-pixel basis using a mono-exponential decay (Eq. [1]) model:

M(TSL)=M0∗exp(−TSLT1rho)[1]

Where M0 and M (TSL) refer to the equilibrium magnetization and T1rho-prepared magnetization at TSL, respectively.

Regions of interest (ROIs) of 100–200 mm² were manually drawn over the irradiated right lobe, carefully avoiding large vessels, bile ducts, and artifacts. The placement and boundaries of ROIs are illustrated in the Figure S1. In a preliminary reproducibility analysis of 10 randomly selected rats, ROIs were independently delineated by two radiologists (10 and 12 years of experience), yielding excellent inter-observer agreement for T1rho values [intraclass correlation coefficient (ICC) =0.91, 95% confidence interval (CI): 0.87–0.95]. Given this high reproducibility, subsequent ROI delineation for the full dataset was performed by a single experienced observer.

Pathological examination

Within 4 hours post-MRI, animals were sacrificed for histological examination. Liver samples were fixed in 4% phosphate-buffered formaldehyde and then embedded in paraffin wax. Sections (5 µm thick) underwent dewaxing in xylene and subsequent rehydration in a series of ethanol concentrations. To assess steatosis and inflammation, hematoxylin-eosin (HE) staining was carried out. A liver histopathologist with 15 years’ experience independently evaluated the histology. The Ishak scoring system (22) and non-alcoholic fatty liver disease (NAFLD) activity scoring (NAS) system (23) was adapted for the semi-quantitative assessment of liver portal inflammation, and hepatocyte ballooning, respectively. In addition, four histological stains were applied: Masson’s trichrome (MT), Picro-Sirius Red (PSR), transforming growth factor-beta (TGF-β) Immunohistochemistry, and alpha-smooth muscle actin (α-SMA) Immunohistochemistry. MT and PSR are both used to evaluate fibrosis. TGF-β is a signaling molecule that initiates and promotes fibrosis. α-SMA is a specific marker of activated hepatic stellate cells and myofibroblasts, the primary effector cells in fibrosis. These four markers were selected to assess complementary aspects of fibrosis, with MT and PSR reflecting collagen deposition and TGF-β and α-SMA capturing upstream fibrogenic signaling and stellate cell activation. Using multiple markers allowed us to evaluate both structural and molecular components of the fibrotic process in this model. Observations were performed using a Leica 2500 microscope at ×400 magnification. For each sample, five regions exhibiting the most intense staining were systematically selected for image acquisition. For each image, the integrated optical density (IOD)—defined as the cumulative optical density of all positively stained pixels—was measured. For immunohistochemical staining (TGF-β, and α-SMA), IOD values were used to estimate protein expression levels, and were normalized by the corresponding area to calculate the mean optical density (IOD/area) for each region. For MT and PSR staining, which are used to visualize and quantify collagen deposition and fibrosis, IOD values were similarly obtained and normalized by area to assess the extent of fibrotic changes in liver tissue. All measurements were averaged across the five selected regions per sample.

Statistical analysis

Statistical analyses were performed using GraphPad Prism (version 8; GraphPad Software, La Jolla, CA, USA). Non-paired comparisons were performed using the Mann-Whitney U test. A P value <0.05 denoted statistical significance. The Pearson correlation coefficient (r) assessed the strength of correlation: r<0.3 indicated weak correlation, 0.3–0.7 indicated moderate correlation, and r>0.7 indicated strong correlation.

Results

Histological findings

In the RT group, fibrosis and cellular disorganization progressed in a time-dependent manner. Mild cellular edema appeared at 2 weeks post-RT and persisted throughout the observation period, accompanied by relatively mild architectural disorganization and hepatocyte misalignment. Ballooning degeneration was mild, appearing from 4 weeks onward, without major differences between later time-points. Inflammation remained minimal, with only limited inflammatory cell infiltration from 2 to 12 weeks. Focal mild necrosis was detected at 8 and 12 weeks. After 6 weeks of TAA injection, early-stage fibrosis developed, characterized by more severe ballooning degeneration than in the RT group. Inflammation was mild and primarily localized around portal tracts and fibrotic areas. Minimal steatosis (<5% of parenchyma) was observed. In the TAA + RT group, pre-existing fibrosis and irradiation acted synergistically, leading to more severe and diffuse fibrosis. Mild ballooning degeneration was consistently present, although its severity was lower than in the TAA-alone group. Inflammation was more pronounced, with significantly higher levels at 2 weeks post-RT compared with TAA alone. Minimal steatosis (<5%) and focal necrosis were observed at later stages. Histomorphometric analysis confirmed a treatment-duration-dependent increase in collagen and fibrosis markers in both models (Figure 2).

Figure 2 Histological progression of RILD in the RT and TAA + RT models. (A-D) show Masson (A), PSR (B), TGF-β (C), α-SMA (D) in the RT group at control, 2, 4, 8, and 12 weeks post-RT. (E-H) show the corresponding Masson (E), PSR (F), TGF-β (G), and α-SMA (H) staining results in the TAA + RT group, including control, TAA alone, and TAA + RT at 2, 4, 8, and 12 weeks post-RT. Note the control rat readings are shared for RT model and the TAA + RT model. AU, arbitrary units; PSR, Picro-Sirius Red; RILD, radiation-induced liver disease; RT, radiation therapy; TAA, thioacetamide; TGF-β, transforming growth factor-beta; α-SMA, alpha-smooth muscle actin.

T1rho changes

The baseline control rats had normal liver T1rho value of 37.2±0.93 ms [range, 36.5–39.1 ms, CoV (coefficient of variation) of 0.02]. An RT of 25 Gy led to an elevated liver T1rho of 40.14±1.06 ms 2 weeks post-RT (P<0.001). With a pre-treatment of TAA, an RT of 20 Gy led to an elevated liver T1rho of 45.24±1.24 ms 2 weeks post-RT, which was statistically significantly higher than the liver T1rho 2 weeks post-RT with a dosage of 25 Gy (P<0.001). Thus, pre-existing liver injury by TAA sensitized the liver to RT injury. For both the RT model and the TAA + RI model, there was a treatment-duration dependent elongation of liver T1rho value (Figures 3,4). For the TAA base-model, liver T1rho allowed complete separation between control rats and TAA treated rats. On the other hand, while there was a statistically significant difference in liver T1rho between control rats and 2 weeks’ RT treated rats (P<0.001), there were slight individual overlaps between the two groups. TAA + 2 weeks’ RT treated rats had liver T1rho values similar to those of 4 weeks’ RT treated rats. TAA + 4 weeks’ RT treated rats had liver T1rho values similar to those of 8 weeks’ RT treated rats. Twelve weeks’ RT treated rats had liver T1rho values longer than those of TAA + 8 weeks’ RT treated rats. TAA + 12 weeks’ RT treated rats had the longest liver T1rho value (Figure 4).

Figure 3 Representative T1rho maps and corresponding histological images of rat livers at different time points after irradiation. (A-D) T1rho-weighted images of a normal control rat acquired with TSL of 0, 10, 30, and 50 ms, respectively. (E) The corresponding T1ρ map. (F-I) T1rho maps from RT rats at 2, 4, 8, and 12 weeks after 25 Gy irradiation to the right lobe. (J-N) T1rho maps in the TAA + RT group, including 6 weeks of TAA-induced fibrosis (J) and combined TAA + RT injury at 2, 4, 8, and 12 weeks after irradiation (K-N). The color scale represents T1rho relaxation times (ms). NC, normal control; RT, radiation therapy; TAA, thioacetamide; TSL, spin-lock times.

Figure 4 For both the RT model and the TAA + RT model, there was a treatment-duration dependent elongation of liver T1rho value. For the TAA base-model, liver T1rho allowed complete separation between control rats and TAA treated rats. For liver T1rho between control rats and 2 weeks’ RT treated rats, there were slight individual overlaps between the two groups. Data are presented as mean ± standard deviation. RT, radiation therapy; TAA, thioacetamide.

Correlation of T1rho with histopathology

In the RT group, T1rho values correlated strongly with Masson (r=0.958), PSR (r=0.920), TGF-β (r=0.916), and α-SMA (r=0.851). In the TAA + RT group, similar correlations were observed with Masson (r=0.944), PSR (r=0.967), TGF-β (r=0.962), and α-SMA (r=0.941). Inter-marker correlations were also strong, including PSR vs. Masson (r=0.974), TGF-β vs. Masson (r=0.956), and α-SMA vs. Masson (r=0.932); all P<0.001 (Figure 5A-5K). These findings indicate that collagen deposition was the predominant contributor to T1rho prolongation. These correlations were as strong as those observed among histological markers themselves (Figure 5I-5K), indicating that collagen deposition was the dominant contributor to T1rho prolongation. Inflammation showed an additive effect: livers with grade-3 inflammation had higher T1rho values than predicted from grade-1/2 inflammation (Figure 6A), suggesting that inflammatory activity contributed to T1rho changes beyond collagen deposition. In contrast, ballooning degeneration did not significantly influence T1rho, as values overlapped regardless of its presence (Figure 6B).

Figure 5 Correlations between liver T1rho and histologic readouts in the RT group (A-D) and the TAA + RT group (E-H), and inter-marker correlations among histologic markers (I-K). Panels show linear regression fits with Pearson correlation coefficients (r) annotated in each plot. All correlations were statistically significant (all P<0.001). Red dashed circles indicate shared control data points included for reference. AU, arbitrary units; PSR, Picro-Sirius Red; RT, radiation therapy; TAA, thioacetamide; TGF-β, transforming growth factor-beta; α-SMA, alpha-smooth muscle actin.

Figure 6 Summary of the influence of inflammatory activity and hepatocyte ballooning on the relationship between T1rho values and fibrosis severity. (A) Data from TAA + RT group. When comparing the rat livers with grade-1/2 inflammation and the rat livers with grade-3 inflammation, the distribution of rat livers with grade-3 inflammation had liver T1rho values higher than could be predicted from rat livers with grade-1/2 inflammation. (B) Data from the controls and RT group. When comparing the rat livers with grade-1 inflammation without ballooning and the rat livers with grade-1 inflammation with grade-1 ballooning, the distribution of rat livers with grade-1 inflammation without ballooning could be predicted from the distribution of rat livers with grade-1 inflammation with ballooning. The line in B is copied from Figure 5A. RT, radiation therapy; TAA, thioacetamide.

Discussion

This study evaluated liver T1rho change for two rat RILD models, with the goal of confirming the suitability of T1rho in assessing RILD. RT induced rat RILD model and TAA + RT induced rat RILD model were established in this study. Histologically, the RT group exhibited mild edema (2–12 weeks), mild ballooning degeneration (4–12 weeks), and minimal inflammation (2–12 weeks). Collagen deposition and TGF-β expression increased significantly at 8–12 weeks (P<0.05), while α-SMA remained mildly elevated from 4 to 12 weeks (P>0.05). The TAA + RT group showed mild/severe ballooning degeneration, moderate inflammation, and markedly higher collagen/fibrosis versus RT (P<0.05). Both TGF-β and α-SMA in TAA + RT rose progressively, peaking at 12 weeks (P<0.05). This study established the liver T1rho time course of RT model and TAA + RT model, demonstrating the time dependent progress of liver injury as measured by liver T1rho.

A number of rat liver disease models have been tested with T1rho, with each model having its unique pathogenetic feature of liver disease initiation and progression. The methionine and choline-deficient (MCD) diet rat model is designed to mimic NAFLD in humans. In rodents, methionine and choline deficiency induce the development of steatohepatitis. After 3 weeks of MCD diet feeding, steatohepatitis is well-established; by weeks 8–10, pericellular and perisinusoidal fibrosis emerge. After 10 weeks on the diet, perivenular and pericellular fibrosis—the characteristic “chicken-wire” fibrosis seen in human NASH—also readily develops (24,25). For the biliary duct ligation (BDL) model, increased intraluminal pressure serves as the mechanism that initiates biliary epithelial cells entering the replicative cycle. When stellate cells are activated and transformed into myofibroblasts, this process leads to fibrogenesis. In contrast, inflammatory responses associated with this model remain mild—even 16 days after BDL (26,27). Compared with the BDL and MCD models, carbon tetrachloride (CCl4)-induced liver fibrosis exhibits greater inflammation severity. It also shows more obvious edema, inflammatory cell infiltration, and hepatocyte degeneration and necrosis (10,12). Recent works by Melin et al. (28) and by Kim et al. (29) reported the establishment of acute RILD in mice, without background liver fibrosis. However, in clinical practice, the majority of patients have underlying liver conditions, particularly liver fibrosis. To address this, we developed a rat RILD model with initially normal liver and a rat RILD model with initially TAA-induced liver fibrosis model. The TAA-induced fibrotic model, in particular, mimics the clinical scenario of liver cancer patients with underlying liver disease, offering a more accurate representation of RILD. Compared to the CCl4-induced liver fibrosis model, in which fibrosis is induced via intraperitoneal CCl4 injection, the TAA-induced model exhibits greater stability, with fibrosis persisting even after the cessation of TAA administration (30). The current study confirms that pre-existing liver injury by TAA sensitized the liver to RT injury. Histologically, in the RT group, the progression of liver fibrosis was time-dependent. Ballooning degeneration emerged at 8 W post-RT and aggravated by 12 W post-RT. Notably, inflammation was minimal throughout the observation period, with only limited inflammatory cell infiltration. Focal necrosis occurred in the later stages (8 and 12 W post-RT). Distinctive fibrotic features along the RT beam path, including bile canaliculi dilation and disruption of hepatic lobular architecture—pathognomonic of radiation-induced injury. Additionally, RILD amplified inflammatory responses, especially within portal regions, demonstrating the acute characteristics of radiation-induced fibrosis.

T1rho has emerged as a promising imaging biomarker for liver injury assessment (15-19). Three key advantages of the T1rho technique are as follows: no additional hardware is required, imaging post-processing is uncomplicated, and T1rho measurements exhibit excellent scan-rescan stability (31,32). In the current study, for the 9 control rats, liver T1rho measure CoV was only 0.02. Earlier studies reported liver T1rho measure CoV for control rats being 0.04 (11) and 0.03 (13), respectively. In a human volunteer’s liver study, Wáng et al. (31) described scan-rescan repeatability in the same scan session had ICC of 0.98, and inter-scan reproducibility ICC of 0.82. T1rho technique can also be potentially highly sensitive to tissue changes, with even menstrual changes of the liver and spleen compositions can be measured in women. Wáng et al. (32,33) noted that spleen T1rho in menstrual phase women was 10.7% lower (median, 95.9 vs. 85.6 ms, P=0.012) than that of non-menstrual phase women, while the liver T1rho in menstrual phase women was 3.8% lower than that of non-menstrual phase women (mean, 42.74 vs. 44.52 ms, P=0.064).

Withstanding the promising results in assessing liver injury with T1rho (9-19), several questions regarding the biochemical or histological mechanisms underlying T1rho changes after liver injury have not yet been fully addressed (32-34). Researchers in previous studies have noted T1rho elongation in conditions linked to the depletion of macromolecules (35). Xie et al. (12) used a CCl4-intoxicated rat model of liver inflammation and fibrosis to investigate T1rho values. Their results showed weak correlation between liver T1rho values and fibrosis stages (r=0.362), and moderate correlation with inflammation grades (r=0.568). For rats with the same inflammation grade, T1rho values did not differ significantly among various fibrosis stages. From this, they determined that inflammation grade was an independent variable linked to T1rho values, and that inflammatory activity affected liver T1rho values more than fibrosis (12). Additionally, in a clinical observational study, Xie et al. noted that simple steatosis does not cause elongation of liver T1rho (18). Though liver iron level has been hypothesized to affect T1rho measure (34), recent analyses support the notion that liver iron concentration has not major implications for T1rho measure (33,36). Consistent with a number of other studies (37,38), Wang et al. (9) and Zhao et al. (13) reported liver collagen deposition, with an increase of macromolecules in the liver, is positively and strongly correlated with T1rho elongation. According to Zhao et al. (13), a 1% rise in collagen contributed roughly to a 1.4 ms elongation of the T1rho relaxation time. The results of this study further support that, in RILD, liver T1rho elongation is dominantly associated with liver collagen deposition with the Pearson correlation r being mostly >0.9 (Figure 5), which is similar to the correlation strength among the four histological reading (Masson, PSR, TGF-β, α-SMA). Although these histologic markers represent different biological components of the fibrotic cascade, their temporal patterns in our study were largely parallel, suggesting that fibrosis progression in this radiation-induced model follows a tightly coupled biological trajectory. This explains why Masson, PSR, TGF-β, and α-SMA exhibited consistent trends despite targeting distinct aspects of liver fibrosis. In this aspect, it can be taken that liver T1rho can be a very good quantitative biomarker for collagen. With an experimental study using MCD diet rat model, Zhao et al. noted that fat deposition shortened liver T1rho measure, with 10% fat increase contributed to 1.55 ms T1rho shortening (14). Zhao et al. identified a weak trend of higher liver inflammation correlating with longer liver T1rho (10), whereas rats with no or minimal inflammation could still have very high T1rho values amid collagen deposition (13). Zhao et al. (14) further proposed that the correlation between liver T1rho and inflammation is more mediated via liver collagen content, as severe inflammation tends to induce more collagen deposition. The current study suggests that liver inflammation contributes to T1rho elongation beyond collagen depiction (Figure 6). The relative contribution of inflammation and collagen to T1rho is not static but varies with the stage of injury. Early after irradiation, inflammatory infiltration account for a larger proportion of T1rho elevation, while at later stages collagen accumulation becomes the predominant driver. Thus, T1rho serves as an integrated marker capturing both early microenvironmental injury and later fibrotic remodeling. Though it is a tentative result, for the first time, the current study suggests that liver ballooning may not contribute to T1rho change. The data of this study supports animal models being used for mechanistic investigation of RILD using T1rho as a non-invasive biomarker, particularly for clinical investigations in which histological samples are not readily available.

This study has several limitations. First, in this study, a spin-lock frequency of 340 Hz was adopted instead of the more commonly used 500 Hz to comply with specific absorption rate (SAR) limitations and avoid excessive RF heating in small animals, as similarly reported in a cardiac T1rho study by Wang et al. (39). Second, the T1rho values obtained here—37.2±0.93 ms (range, 36.5–39.1 ms) for control rat livers—were marginally lower than those documented in previous studies (9,11,13). This discrepancy might be associated with both the T1rho sequence configuration and the TSLs selected in this study. Third, the T1rho acquisition was limited to representative single slices rather than full liver coverage, which may have introduced sampling bias. Fourth, due to the ex vivo sampling for histology, the absence of serial in vivo T1rho imaging in the same animals over time may have reduced sensitivity to assess individual liver injury progression for the data presented in the study. Last, due to the limited sample size and the strong interdependence among histological parameters, multivariate analyses to disentangle the independent effects of collagen, steatosis, and ballooning on T1rho were not performed in this study. Future studies with larger cohorts will apply such advanced statistical approaches.

Conclusions

In conclusion, this study demonstrates that T1rho MRI quantitatively captures the severity of RILD in experimental rat models, with pre-existing fibrosis amplifying radiation-induced injury. T1rho values correlated strongly with histological indices of collagen deposition and were additionally modulated by inflammatory activity, while ballooning degeneration contributed minimally. These findings establish T1rho as a sensitive and noninvasive biomarker of liver injury that complements conventional imaging. Beyond its preclinical validation, T1rho shows promise for early detection, individualized risk assessment, and mechanistic exploration of RILD, particularly in patients with chronic liver disease.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the ARRIVE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2076/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2076/dss

Funding: This study was supported by the National Natural Science Foundation of China (No. 82171890).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-2076/coif). Y.X.J.W. serves as the Editor-in-Chief of Quantitative Imaging in Medicine and Surgery. Y.Z. and F.Z. serve as unpaid editorial board members of Quantitative Imaging in Medicine and Surgery. All authors declared that this study was supported by the National Natural Science Foundation of China (No. 82171890). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All animal experiments were performed under a project license (No. 2019-826) granted by the Animal Experimentation Ethics Committee of the First Affiliated Hospital, Zhejiang University School of Medicine, in compliance with institutional guidelines for the care and use of animals.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. EASL Clinical Practice Guidelines. Management of hepatocellular carcinoma. J Hepatol 2018;69:182-236. [Crossref] [PubMed]
2. Angelico R, Grimaldi C, Saffioti MC, Castellano A, Spada M. Hepatocellular carcinoma in children: hepatic resection and liver transplantation. Transl Gastroenterol Hepatol 2018;3:59. [Crossref] [PubMed]
3. Facciorusso A, Chierici A, Cincione I, Sacco R, Ramai D, Mohan BP, Chandan S, Ofosu A, Cotsoglou C. Stereotactic body radiotherapy vs. radiofrequency ablation for the treatment of hepatocellular carcinoma: a meta-analysis. Expert Rev Anticancer Ther 2021;21:681-8. [Crossref] [PubMed]
4. Bergamo AM, Kauweloa K, Gan G, Shi Z, Daniels J, Crownover R, Narayanasamy G, Stathakis S, Mavroidis P, Papanikolaou N, Gutierrez A. Correlation between Biological Effective Dose and Radiation-induced Liver Disease from Hypofractionated Radiotherapy. J Med Phys 2019;44:185-90. [Crossref] [PubMed]
5. Mathew AS, Atenafu EG, Owen D, Maurino C, Brade A, Brierley J, Dinniwell R, Kim J, Cho C, Ringash J, Wong R, Cuneo K, Feng M, Lawrence TS, Dawson LA. Long term outcomes of stereotactic body radiation therapy for hepatocellular carcinoma without macrovascular invasion. Eur J Cancer 2020;134:41-51. [Crossref] [PubMed]
6. Koay EJ, Owen D, Das P. Radiation-Induced Liver Disease and Modern Radiotherapy. Semin Radiat Oncol 2018;28:321-31. [Crossref] [PubMed]
7. De La Pinta Alonso C. Radiation-induced liver disease in the era of SBRT: a review. Expert Rev Gastroenterol Hepatol 2020;14:1195-201. [Crossref] [PubMed]
8. Kim JY, Kim MJ, Kim KA, Jeong HT, Park YN. Hyperintense HCC on hepatobiliary phase images of gadoxetic acid-enhanced MRI: correlation with clinical and pathological features. Eur J Radiol 2012;81:3877-82. [Crossref] [PubMed]
9. Wang YX, Yuan J, Chu ES, Go MY, Huang H, Ahuja AT, Sung JJ, Yu J. T1rho MR imaging is sensitive to evaluate liver fibrosis: an experimental study in a rat biliary duct ligation model. Radiology 2011;259:712-9. [Crossref] [PubMed]
10. Zhao F, Wang YX, Yuan J, Deng M, Wong HL, Chu ES, Go MY, Teng GJ, Ahuja AT, Yu J MR. T1ρ as an imaging biomarker for monitoring liver injury progression and regression: an experimental study in rats with carbon tetrachloride intoxication. Eur Radiol 2012;22:1709-16. [Crossref] [PubMed]
11. Koon CM, Zhang X, Chen W, Chu ES, San Lau CB, Wáng YX. Black blood T1rho MR imaging may diagnose early stage liver fibrosis: a proof-of-principle study with rat biliary duct ligation model. Quant Imaging Med Surg 2016;6:353-63. [Crossref] [PubMed]
12. Xie S, Qi H, Li Q, Zhang K, Zhang L, Cheng Y, Shen W. Liver injury monitoring, fibrosis staging and inflammation grading using T1rho magnetic resonance imaging: an experimental study in rats with carbon tetrachloride intoxication. BMC Gastroenterol 2020;20:14. [Crossref] [PubMed]
13. Zhao F, Zhou N, Wang JL, Zhou H, Zou LQ, Zhong WX, He J, Zheng CJ, Yan SX, Wáng YXJ. Collagen deposition in the liver is strongly and positively associated with T1rho elongation while fat deposition is associated with T1rho shortening: an experimental study of methionine and choline-deficient (MCD) diet rat model. Quant Imaging Med Surg 2020;10:2307-21. [Crossref] [PubMed]
14. Zhao F, Zhou N, Wang X, Wang JL, Zhong WX, Deng M, Zheng CJ, He J, Yan SX, Wáng YXJ. T1rho shortening effect of fat in liver steatosis after fat suppression: approximate estimation in a methionine and choline-deficient (MCD) diet rat model. Quant Imaging Med Surg 2021;11:870-5. [Crossref] [PubMed]
15. Wang YX, Zhao F, Wong VWS, Yuan J, Kwong KM, Chen HLY. Liver MR T1rho measurement in liver cirrhosis patients: a preliminary study with a 2D fast field echo sequence at 3T. Proc Intl Soc Mag Reson Med 2012;20:1289.
16. Allkemper T, Sagmeister F, Cicinnati V, Beckebaum S, Kooijman H, Kanthak C, Stehling C, Heindel W. Evaluation of fibrotic liver disease with whole-liver T1ρ MR imaging: a feasibility study at 1.5 T. Radiology 2014;271:408-15. [Crossref] [PubMed]
17. Takayama Y, Nishie A, Asayama Y, Ushijima Y, Okamoto D, Fujita N, Morita K, Shirabe K, Kotoh K, Kubo Y, Okuaki T, Honda H. T1 ρ Relaxation of the liver: A potential biomarker of liver function. J Magn Reson Imaging 2015;42:188-95. [Crossref] [PubMed]
18. Xie S, Li Q, Cheng Y, Zhang Y, Zhuo Z, Zhao G, Shen W. Impact of Liver Fibrosis and Fatty Liver on T1rho Measurements: A Prospective Study. Korean J Radiol 2017;18:898-905. [Crossref] [PubMed]
19. Peng XG, Zhao Y, Qin Z, Jiang Y, Dai J, Cui Y. Longitudinal T1ρ quantification in chronic hepatitis B: from fibrosis staging to pegylated interferon-α therapeutic response tracking. Eur J Radiol 2025;190:112203. [Crossref] [PubMed]
20. Bentzen SM, Constine LS, Deasy JO, Eisbruch A, Jackson A, Marks LB, Ten Haken RK, Yorke ED. Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC): an introduction to the scientific issues. Int J Radiat Oncol Biol Phys 2010;76:S3-9. [Crossref] [PubMed]
21. Pan CC, Kavanagh BD, Dawson LA, Li XA, Das SK, Miften M, Ten Haken RK. Radiation-associated liver injury. Int J Radiat Oncol Biol Phys 2010;76:S94-100. [Crossref] [PubMed]
22. Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, Denk H, Desmet V, Korb G, MacSween RN. Histological grading and staging of chronic hepatitis. J Hepatol 1995;22:696-9. [Crossref] [PubMed]
23. Pai RK, Jairath V, Hogan M, Zou G, Adeyi OA, Anstee QM, et al. Reliability of histologic assessment for NAFLD and development of an expanded NAFLD activity score. Hepatology 2022;76:1150-63. [Crossref] [PubMed]
24. Larter CZ, Yeh MM. Animal models of NASH: getting both pathology and metabolic context right. J Gastroenterol Hepatol 2008;23:1635-48. [Crossref] [PubMed]
25. Schattenberg JM, Galle PR. Animal models of non-alcoholic steatohepatitis: of mice and man. Dig Dis 2010;28:247-54. [Crossref] [PubMed]
26. Aller MA, Lorente L, Alonso S, Arias J. A model of cholestasis in the rat, using a microsurgical technique. Scand J Gastroenterol 1993;28:10-4. [Crossref] [PubMed]
27. Slott PA, Liu MH, Tavoloni N. Origin, pattern, and mechanism of bile duct proliferation following biliary obstruction in the rat. Gastroenterology 1990;99:466-77. [Crossref] [PubMed]
28. Melin N, Yarahmadov T, Sanchez-Taltavull D, Birrer FE, Brodie TM, Petit B, Felser A, Nuoffer JM, Montani M, Vozenin MC, Herrmann E, Candinas D, Aebersold DM, Stroka D. A new mouse model of radiation-induced liver disease reveals mitochondrial dysfunction as an underlying fibrotic stimulus. JHEP Rep 2022;4:100508. [Crossref] [PubMed]
29. Kim J, Jung Y. Radiation-induced liver disease: current understanding and future perspectives. Exp Mol Med 2017;49:e359. [Crossref] [PubMed]
30. Deng X, Zhang X, Li W, Feng RX, Li L, Yi GR, Zhang XN, Yin C, Yu HY, Zhang JP, Lu B, Hui L, Xie WF. Chronic Liver Injury Induces Conversion of Biliary Epithelial Cells into Hepatocytes. Cell Stem Cell 2018;23:114-122.e3. [Crossref] [PubMed]
31. Wáng YXJ, Deng M, Lo GG, Liang D, Yuan J, Chen W. Breath-hold black-blood T1rho mapping improves liver T1rho quantification in healthy volunteers. Acta Radiol 2018;59:257-65. [Crossref] [PubMed]
32. Wáng YXJ, Deng M, Lin J, Kwok AWL, Liu EKW, Chen W. Age- and Gender-Associated Liver Physiological T1rho Dynamics Demonstrated with a Clinically Applicable Single-Breathhold Acquisition. SLAS Technol 2018;23:179-87. [Crossref] [PubMed]
33. Wáng YXJ, Yu WL, Deng M. Age- and sex-related variations of normal spleen T1rho and the more stable liver T1rho to spleen T1rho ratio. Rofo 2025;197:829-35. [Crossref] [PubMed]
34. Wáng YXJ, Chen W, Deng M. How liver pathologies contribute to T1rho contrast require more careful studies. Quant Imaging Med Surg 2017;7:608-13. [Crossref] [PubMed]
35. Duvvuri U, Goldberg AD, Kranz JK, Hoang L, Reddy R, Wehrli FW, Wand AJ, Englander SW, Leigh JS. Water magnetic relaxation dispersion in biological systems: the contribution of proton exchange and implications for the noninvasive detection of cartilage degradation. Proc Natl Acad Sci U S A 2001;98:12479-84. [Crossref] [PubMed]
36. Wáng YXJ. Physiological variation of liver iron concentration may not be dominantly responsible for the liver T1rho variations associated with age and gender. Quant Imaging Med Surg 2021;11:1668-73. [Crossref] [PubMed]
37. Hectors SJ, Bane O, Kennedy P, El Salem F, Menon M, Segall M, Khaim R, Delaney V, Lewis S, Taouli B T. (1ρ) mapping for assessment of renal allograft fibrosis. J Magn Reson Imaging 2019;50:1085-91. [Crossref] [PubMed]
38. Hectors SJ, Bane O, Stocker D, Carbonell G, Lewis S, Kennedy P, Schiano TD, Thung S, Fischman A, Taouli B. Splenic T(1ρ) as a noninvasive biomarker for portal hypertension. J Magn Reson Imaging 2020;52:787-94. [Crossref] [PubMed]
39. Wang C, Zheng J, Sun J, Wang Y, Xia R, Yin Q, Chen W, Xu Z, Liao J, Zhang B, Gao F. Endogenous contrast T1rho cardiac magnetic resonance for myocardial fibrosis in hypertrophic cardiomyopathy patients. J Cardiol 2015;66:520-6. [Crossref] [PubMed]