Research ArticleClinical Studies
Open Access

Prediction of Early-stage Liver Fibrosis Using FDG-PET/CT

RIN SHINZATO, AKIHIRO NISHIE, TOMOKO TAMAKI, NAOKI WADA, MITSUHISA TAKATSUKI and GYO IIDA
Anticancer Research September 2023, 43 (9) 4221-4227; DOI: https://doi.org/10.21873/anticanres.16614

Abstract

Background/Aim: 18F-fluorodeoxyglucose (FDG) is known to accumulate in the liver. We investigated whether accumulation of FDG was correlated with the degree of liver fibrosis and the grade of necro-inflammatory activity. Patients and Methods: This retrospective study included 35 patients who underwent FDG-positron emission tomography (PET)/computed tomography (CT) before liver surgery. On fusion images of CT and PET, by placing regions of interest on the lateral, anterior and posterior segments of the liver and the aorta, the standardized uptake value (SUV) mean, and SUV normalized by lean body mass (SUL) mean of the liver were measured, and the ratio SUVmean liver/SUVmean aorta was calculated. According to the New Inuyama Classification, subjects were classified into three groups based on the grade of liver-fibrosis degree, i.e., F0, F1+F2 and F3+F4, and into three groups based on the grade of necro-inflammatory activity, i.e., A0, A1 and A2. Each of the above parameters was then compared among the groups using a Tukey test. Results: Average SULmean liver values of the F0, F1+F2 and F3+F4 groups were 1.573±0.211, 1.845±0.220 and 1.716±0.119, respectively. The SULmean liver of the F1+F2 group was significantly higher than that of the F0 group (p=0.0296). No significant difference was observed for the other two parameters. None of the parameters exhibited significant difference among the A0, A1, and A2 groups. Conclusion: FDG accumulation in the liver may be increased in the early stage of liver fibrosis. SULmean liver could be used to determine the necessity for therapeutic intervention in chronic liver disease.

Key Words:

Chronic hepatitis is a risk factor of liver cirrhosis and hepatocarcinogenesis (1-3). In recent years, the incidence of nonalcoholic steatohepatitis (NASH) has been increasing along with the prevalence of nonalcoholic fatty liver disease. NASH has a worse prognosis than nonalcoholic fatty liver disease and a significantly higher probability of liver-related death (4-7). In such chronic liver diseases, the degree of fibrosis, necrosis and inflammation can serve as indices for determining whether therapeutic intervention is necessary to prevent progression to liver cirrhosis. In general, the liver histology is diagnosed by needle biopsy. However, this method has a risk of various complications, including hemorrhage and infection (8). In addition, the likelihood of diagnostic disagreement among pathologists and the sampling error are causes of concern when employing a liver biopsy (8). For all these reasons, it would be of clinical value if we could obtain an accurate, non-invasive histology of the liver parenchyma.

18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography (CT) is a modality in which FDG is administered into the body and its accumulation is detected by a PET camera. FDG accumulates in areas where glycometabolism is active, such as tumors and regions of inflammation (9). It is also known to accumulate in normal organs such as the brain, myocardium, and liver (10). In inflammatory foci, increases in vascular flow occur, followed in order by telangiectasis, rupture and growth of capillaries, infiltration of inflammatory cells, and finally fibrosis (11). Among inflammatory cells, neutrophils, macrophages and activated lymphocytes use anaerobic glycolysis as an energy source (12), and we therefore hypothesized that FDG accumulation in the liver may change with the degree of inflammation and fibrosis.

In this study, we examined whether accumulation of FDG was correlated with the degree of liver fibrosis and the grade of necro-inflammatory activity.

Patients and Methods

Patients. This retrospective study was approved by the institutional review board of our hospital. The requirements for informed consent were waived. The institutional ethics committees comply with the Declaration of Helsinki.

Referring to the medical data recorded at our hospital, we enrolled 346 patients who underwent FDG-PET/CT for evaluation of hepatobiliary diseases, and 776 patients who underwent FDG-PET/CT for reasons other than hepatobiliary diseases and were found to have liver tumors between July 2016 and July 2022. From the total group of 1,122 patients, we extracted 47 patients who underwent hepatic resection or liver transplantation. Among them, 12 patients for whom only partial hepatectomy was performed were excluded because we considered that the histology of peritumoral liver parenchyma did not represent that of the whole liver due to cytokine interaction between a tumor and peritumoral liver parenchyma and changes in the blood stream associated with a liver tumor itself. The inclusion and exclusion criteria are summarized in Figure 1. The remaining 35 patients included 20 men and 15 women (age range=27-82 years; mean age=65 years). Hepatitis B surface antigen was present in 2 cases and hepatitis C virus antibody in 1 case; neither was seen in the remaining 32 cases. The grading of liver dysfunction was pre-operatively evaluated based on the Child-Pugh classification, and 27, 7, and 1 patient were categorized into Grades A, B, and C, respectively. The underlying liver diseases for which surgery and liver transplantation were performed are summarized in Table I. The FDG-PET/CT examination and surgery were performed within 4 months (median, 1 month) for each patient. Two of the 8 patients with HCCs were treated with tyrosine kinase inhibitors, while 11 of the 15 patients with liver metastases were treated with oral or intravenous chemotherapy.

Figure 1.
Figure 1.

Inclusion and exclusion criteria of the enrolled patients.

View this table:
Table I.

Underlying liver diseases of the patients.

FDG-PET/CT. FDG-PET/CT examinations were performed according to a previously reported method (13). After fasting for 6 h, FDG was injected intravenously at 3.7 MBq/kg, and the image was taken after about 60 min of rest. PET/CT scans from the skull to upper thighs were performed using an integrated Biograph mCT PET/CT scanner (Siemens Healthcare, Erlangen, Germany). Prior to PET image acquisition, low-dose CT (tube voltage, 120 kV; tube current, auto mA) was scanned for the purpose of attenuation correction and precise anatomical localization. Time-of-flight PET imaging in three-dimensional mode was reconstructed on a 200×200 image matrix using the iterative reconstruction algorithm provided by Siemens Healthcare. The section thickness was 2.0 mm. For providing attenuation-corrected FDG-PET images, the CT data and an ordered subset expectation maximization algorithm were used. A Gaussian filter, set at full width at half maximum, and scatter correction were applied for smoothing. CT and PET images were fused using dedicated software (Syngo.via VA30A; Siemens Healthcare, Erlangen, Germany).

Image analysis. Three observers independently performed image analysis using CT-PET fusion images and the above-described software. First, they selected the best image quality slice at the hepatic portal region level. They measured the SUV (standardized uptake value) mean and SUL (SUV normalized by lean body mass) mean by placing 2-2.5 cm2 round regions of interest (ROIs) on the lateral, anterior and posterior segments of the liver and the aorta, avoiding vessels, tumors, and artifacts. SUL was calculated using the following equation: SUL=SUV×lean body weight/body weight. The averages of three SUV and three SUL values in the liver of an individual patient were defined as the “SUVmean liver” and “SULmean liver” for that patient, respectively. For analysis, the average values of SUVmean liver and SULmean liver of three observers were used. The average SUVmean values of the aorta of three observers was defined as the “SUVmean aorta”, and the ratio SUVmean liver/SUVmean aorta was also calculated.

To ascertain the reproducibility of SUVmean liver, SULmean liver and the SUVmean liver/SUV mean aorta ratio, intraclass correlation coefficients (ICCs) were also calculated between observers 1 and 2, observers 2 and 3, and observers 1 and 3, respectively.

Pathologic analysis. Two pathologists, who were unaware of the imaging data, reviewed the hematoxylin-eosin-stained glass slides of each patient to determine the histological findings of the liver parenchyma in a consensus fashion. According to the New Inuyama Classification, the degree of liver fibrosis was classified into five groups: F0 (no fibrosis), F1 (fibrous portal expansion), F2 (bridging fibrosis), F3 (bridging fibrosis with architectural distortion) and F4 (liver cirrhosis). Similarly, the grade of necro-inflammatory activity was scored as A0 (no necro-inflammatory reaction), A1 (mild), A2 (moderate) and A3 (severe) (14).

Statistical analysis. The correlations of SUVmean liver, SULmean liver and SUVmean liver/SUVmean aorta with the degree of liver fibrosis or the grade of necro-inflammatory activity were assessed using a Spearman’s rank correlation test, and a correlation coefficient (rho) was also obtained. Average values of SUVmean liver, SULmean liver and SUVmean liver/SUVmean aorta were compared among the groups F0, F1+F2 and F3+F4, and among the groups A0, A1 and A2 using a Tukey test.

For all tests, a p-value of <0.05 was considered to indicate a statistically significant difference.

Results

The average (minimum–maximum) values of SUVmean liver, SULmean liver and the SUVmean liver/SUVmean aorta ratio were 2.37 (1.64-3.52), 1.79 (1.19-2.57), and 1.37 (1.03-1.95), respectively. The results of pathologic analysis for the liver parenchyma are summarized in Table II. Table III shows the correlation coefficients between the FDG-PET/CT parameters of the degree of liver fibrosis or the grade of necro-inflammatory activity. No significant correlation was obtained for any of the parameters. Table IV shows the results of the comparison of FDG-PET/CT parameters among the F0, F1+F2 and F3+F4 groups. The average SULmean liver of the F1+F2 group was significantly higher than that of the F0 group (p=0.0296) (Figure 2). Table V shows the results of the comparison of FDG-PET/CT parameters among the A0, A1 and A2 groups. No significant difference was found between any two of the three groups. Table VI shows the ICC results. All ICCs between each pair of observers for SULmean liver were 0.868 or more. Representative cases are shown in Figure 3, Figure 4, and Figure 5.

View this table:
Table II.

Pathologic analysis for the liver parenchyma.

View this table:
Table III.

Correlation coefficients between FDG-PET/CT parameters and the degree of liver fibrosis or the grade of necro-inflammatory activity.

View this table:
Table IV.

Comparison of the FDG-PET/CT parameters among the F0, F1+F2 and F3+F4 groups.

Figure 2.
Figure 2.

Comparison of the SULmean liver among the F0, F1+F2 and F3+F4 groups. The average SULmean liver of the F1+F2 group was significantly higher than that of the F0 group (p=0.0296).

View this table:
Table V.

Comparison of the FDG-PET/CT parameters among the A0, A1 and A2 groups.

View this table:
Table VI.

Intraclass correlation coefficients.

Figure 3.
Figure 3.

A 63-year-old woman with liver metastasis. A) An FDG-PET/CT fusion image. Round regions of interest were placed on the lateral, anterior and posterior segments of the liver and the aorta. The SUVmean liver was 2.02, the SULmean liver was 1.32 and the SUVmean liver/SUVmean aorta ratio was 1.27. An arrow indicates a tumor. B) Hematoxylin-eosin staining (objective, ×10). The degree of liver fibrosis was F0, while the grade of necro-inflammatory activity was A0.

Figure 4.
Figure 4.

An 80-year-old woman with intraductal papillary neoplasm of the bile duct. A) An FDG-PET/CT fusion image. Round regions of interest were placed on the lateral, anterior and posterior segments of the liver and the aorta. The SUVmean liver was 2.79, the SULmean liver was 1.95 and the SUVmean liver/SUVmean aorta ratio was 1.49. B) Hematoxylin-eosin staining (objective, ×10). The degree of liver fibrosis was F2, while the grade of necro-inflammatory activity was A1.

Figure 5.
Figure 5.

A 45-year-old man with hepatocellular carcinoma. A) An FDG-PET/CT fusion image. Round regions of interest were placed on the lateral, anterior and posterior segments of the liver and the aorta. The SUVmean liver was 2.25, the SULmean liver was 1.77 and the SUVmean liver/SUVmean aorta ratio was 1.45. An arrow indicates a tumor. B) Hematoxylin-eosin staining (objective, ×4). The degree of liver fibrosis was F4, while the grade of necro-inflammatory activity was A2.

Discussion

In this study there was no significant rank correlation between any of the FDG-PET/CT parameters and the degree of fibrosis. Although no significant differences were observed in the SUVmean liver or the SUVmean liver/SUVmean aorta ratio, the SULmean liver was significantly higher in the F1+F2 group compared to the F0 group. There are two possible reasons for the significant difference in SULmean liver. First, the effect of fat was excluded from SUVmean. In general, no FDG uptake is seen in fat tissue, and therefore the SULmean may reflect the degree of accumulation more accurately than the SUVmean. Second, because the SUVmean aorta values were likely to fluctuate depending on the slice, the SUVmean liver/SUVmean ratio might have been unstable. Indeed, the ICCs between each pair of observers for the SUVmean liver/SUVmean aorta ratio were low. In a previous report, the SUVmean liver was found to increase from F0 to F1+F2 and to significantly decrease from F3+F4 (15). This study was performed using rabbits in vivo. In a clinical study, the liver to blood ratio increased from F0 to F2 and decreased from F3+F4. This previous clinical study was performed using images taken 75 min after injection of FDG, and volumes of interest were placed only in the right lobe for calculation of SUVmean in the liver (16). The results were similar to those of our present study, although the parameter showing a significant difference differed between the two reports. Based on all the above, the SULmean liver might serve as a useful index of the necessity of therapeutic intervention in patients with chronic liver disease.

Why was the SULmean liver higher in the F1+F2 group than the other groups? Plausible reasons are as follows. As mentioned earlier, increased vascular flow in inflammatory foci causes telangiectasis, followed in order by rupture and growth of capillaries, infiltration of inflammatory cells, and finally fibrosis (11). Among inflammatory cells, neutrophils, macrophages and activated lymphocytes use anaerobic glycolysis as an energy source (12). Therefore, we hypothesized that the accumulation of FDG in the liver might change especially with the degree of inflammation. However, there was no significant rank correlation between each FDG-PET/CT parameter and the degree of inflammation. According to a previous report (11), liver fibrosis is induced by activation of Kupffer cells, which are resident macrophages in the liver, followed by activation of stellate cells and increase and deposition of extracellular matrices. Therefore, we consider that the higher SULmean liver in the F1+F2 group may reflect activation of Kupffer cells at the early stage of liver fibrosis. Unless macrophages infiltrating the liver accompanied with inflammation induce liver fibrosis, this could account for all of our results. In a previous study using superparamagnetic iron oxide (SPIO), an iron particle which is phagocytosed by Kupffer cells, the F1+F2 group showed increased SPIO uptake capacity (8). In addition, it has been shown that activated monocytes in vivo show increased SPIO uptake capacity (17). These reports support the hypothesis that the higher SULmean liver of F1+F2 may reflect activation of Kupffer cells at the early stage of liver fibrosis.

Our study has several limitations. First, the number of cases was small. Patients who underwent FDG-PET/CT before hepatic resection for liver tumor or liver transplantation were limited. Second, the etiology of chronic liver disease varied. Some patients might have had mild pathological changes in the liver due to chemotherapy. Third, the number of cases in each stage of fibrosis and necro-inflammatory activity varied, probably because we collected patients consecutively. Fourth, the FDG-PET scan was performed only once and at a fixed time during an examination. Therefore, the dynamic changes in FDG accumulation could not be evaluated. Fifth, the ROI may have contained tiny blood vessels, although we made an effort to avoid visible vessels.

Conclusion

FDG accumulation in the liver may be increased at the early stage of liver fibrosis. SULmean liver could be used to determine the necessity for therapeutic intervention in patients with chronic liver disease.

Footnotes

  • Authors’ Contributions

    Conceptualization: Akihiro Nishie; Data curation: Rin Shinzato, Akihiro Nishie, Tomoko Tamaki, Naoki Wada, Mitsuhisa Takatsuki, Gyo Iida; Formal analysis: Akihiro Nishie; Investigation: Rin Shinzato, Akihiro Nishie, Tomoko Tamaki, Naoki Wada, Mitsuhisa Takatsuki, Gyo Iida; Methodology: Akihiro Nishie, Tomoko Tamaki, Naoki Wada, Gyo Iida; Project administration: Akihiro Nishie; Resources: Naoki Wada, Mitsuhisa Takatsuki, Gyo Iida; Supervision: Akihiro Nishie; Writing - original draft: Rin Shinzato; Writing - review & editing: Akihiro Nishie, Tomoko Tamaki, Naoki Wada, Mitsuhisa Takatsuki, Gyo Iida.

  • Conflicts of Interest

    No potential conflicts of interest were disclosed in relation to this study.

  • Received June 24, 2023.
  • Revision received July 19, 2023.
  • Accepted July 20, 2023.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).

References

  1. ↵
    1. Kim AI,
    2. Saab S
    : Treatment of hepatitis C. Am J Med 118(8): 808-815, 2005. DOI: 10.1016/j.amjmed.2005.01.073
    OpenUrlCrossRefPubMed
    1. Kumada H,
    2. Okanoue T,
    3. Onji M,
    4. Moriwaki H,
    5. Izumi N,
    6. Tanaka E,
    7. Chayama K,
    8. Sakisaka S,
    9. Takehara T,
    10. Oketani M,
    11. Suzuki F,
    12. Toyota J,
    13. Nomura H,
    14. Yoshioka K,
    15. Seike M,
    16. Yotsuyanagi H,
    17. Ueno Y, The Study Group for the Standardization of Treatment of Viral Hepatitis Including Cirrhosis, Ministry of Health, Labor and Welfare of Japan
    : Guidelines for the treatment of chronic hepatitis and cirrhosis due to hepatitis B virus infection for the fiscal year 2008 in Japan. Hepatol Res 40(1): 1-7, 2010. DOI: 10.1111/j.1872-034X.2009.00633.x
    OpenUrlCrossRefPubMed
  3. ↵
    1. Kumada H,
    2. Okanoue T,
    3. Onji M,
    4. Moriwaki H,
    5. Izumi N,
    6. Tanaka E,
    7. Chayama K,
    8. Sakisaka S,
    9. Takehara T,
    10. Oketani M,
    11. Suzuki F,
    12. Toyota J,
    13. Nomura H,
    14. Yoshioka K,
    15. Seike M,
    16. Yotsuyanagi H,
    17. Ueno Y, The Study Group for the Standardization of Treatment of Viral Hepatitis Including Cirrhosis, Ministry of Health, Labour and Welfare of Japan
    : Guidelines for the treatment of chronic hepatitis and cirrhosis due to hepatitis C virus infection for the fiscal year 2008 in Japan. Hepatology Research 40(1): 8-13, 2010. DOI: 10.1111/j.1872-034X.2009.00634.x
    OpenUrlCrossRefPubMed
  4. ↵
    1. Adams LA,
    2. Lymp JF,
    3. St. Sauver J,
    4. Sanderson SO,
    5. Lindor KD,
    6. Feldstein A,
    7. Angulo P
    : The natural history of nonalcoholic fatty liver disease: A Population-Based Cohort Study. Gastroenterology 129(1): 113-121, 2005. DOI: 10.1053/j.gastro.2005.04.014
    OpenUrlCrossRefPubMed
    1. Ekstedt M,
    2. Franzén LE,
    3. Mathiesen UL,
    4. Thorelius L,
    5. Holmqvist M,
    6. Bodemar G,
    7. Kechagias S
    : Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 44(4): 865-873, 2006. DOI: 10.1002/hep.21327
    OpenUrlCrossRefPubMed
    1. Rafiq N,
    2. Bai C,
    3. Fang Y,
    4. Srishord M,
    5. Mccullough A,
    6. Gramlich T,
    7. Younossi ZM
    : Long-term follow-up of patients with nonalcoholic fatty liver. Clin Gastroenterol Hepatol 7(2): 234-238, 2009. DOI: 10.1016/j.cgh.2008.11.005
    OpenUrlCrossRefPubMed
  7. ↵
    1. Musso G,
    2. Gambino R,
    3. Cassader M,
    4. Pagano G
    : Meta-analysis: Natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity. Ann Med 43(8): 617-649, 2011. DOI: 10.3109/07853890.2010.518623
    OpenUrlCrossRefPubMed
  8. ↵
    1. Nishie A,
    2. Asayama Y,
    3. Ishigami K,
    4. Tajima T,
    5. Kakihara D,
    6. Nakayama T,
    7. Takayama Y,
    8. Okamoto D,
    9. Taketomi A,
    10. Shirabe K,
    11. Fujita N,
    12. Obara M,
    13. Yoshimitsu K,
    14. Honda H
    : MR prediction of liver fibrosis using a liver-specific contrast agent: Superparamagnetic iron oxide versus Gd-EOB-DTPA. J Magn Reson Imaging 36(3): 664-671, 2012. DOI: 10.1002/jmri.23691
    OpenUrlCrossRefPubMed
  9. ↵
    1. Alavi A,
    2. Kung JW,
    3. Zhuang H
    : Implications of PET based molecular imaging on the current and future practice of medicine. Semin Nucl Med 34(1): 56-69, 2004. DOI: 10.1053/j.semnuclmed.2003.09.007
    OpenUrlCrossRefPubMed
  10. ↵
    1. Zincirkeser S,
    2. Şahin E,
    3. Halac M,
    4. Sager S
    : Standardized uptake values of normal organs on 18F-fluorodeoxyglucose positron emission tomography and computed tomography imaging. J Int Med Res 35(2): 231-236, 2007. DOI: 10.1177/147323000703500207
    OpenUrlCrossRefPubMed
  11. ↵
    1. Mormone E,
    2. George J,
    3. Nieto N
    : Molecular pathogenesis of hepatic fibrosis and current therapeutic approaches. Chem Biol Interact 193(3): 225-231, 2011. DOI: 10.1016/j.cbi.2011.07.001
    OpenUrlCrossRefPubMed
  12. ↵
    1. Palestro CJ
    : FDG-PET in musculoskeletal infections. Semin Nucl Med 43(5): 367-376, 2013. DOI: 10.1053/j.semnuclmed.2013.04.006
    OpenUrlCrossRefPubMed
  13. ↵
    1. Toguchi M,
    2. Ishigami K,
    3. Goya M,
    4. Saito S,
    5. Murayama S,
    6. Nishie A
    : Efficacy of pre-operative 18F-FDG PET/CT in prognostic prediction in patients with renal cell carcinoma. Cancer Diagn Progn 2(2): 216-222, 2022. DOI: 10.21873/cdp.10097
    OpenUrlCrossRefPubMed
  14. ↵
    1. Nishie A,
    2. Akahori S,
    3. Asayama Y,
    4. Ishigami K,
    5. Ushijima Y,
    6. Kakihara D,
    7. Nakayama T,
    8. Takayama Y,
    9. Fujita N,
    10. Morita K,
    11. Ishimatsu K,
    12. Takao S,
    13. Yoshizumi T,
    14. Kohashi K,
    15. Li Y,
    16. Honda H
    : Prediction of liver fibrosis using ct under respiratory control: new attempt using deformation vectors obtained by non-rigid registration technique. Anticancer Res 39(3): 1417-1424, 2019. DOI: 10.21873/anticanres.13257
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Pan S,
    2. Wang L,
    3. Xin J
    : Combining 18F-FDG PET and Gd-EOB-DTPA-enhanced MRI for staging liver fibrosis. Life Sci 269: 119086, 2021. DOI: 10.1016/j.lfs.2021.119086
    OpenUrlCrossRefPubMed
  16. ↵
    1. Verloh N,
    2. Einspieler I,
    3. Utpatel K,
    4. Menhart K,
    5. Brunner S,
    6. Hofheinz F,
    7. van den Hoff J,
    8. Wiggermann P,
    9. Evert M,
    10. Stroszczynski C,
    11. Hellwig D,
    12. Grosse J
    : In vivo confirmation of altered hepatic glucose metabolism in patients with liver fibrosis/cirrhosis by 18F-FDG PET/CT. EJNMMI Res 8: 98, 2018. DOI: 10.1186/s13550-018-0452-y
    OpenUrlCrossRefPubMed
  17. ↵
    1. Nishie A,
    2. Yoshimitsu K,
    3. Nakayama T,
    4. Hatakenaka M,
    5. Irie H,
    6. Shioyama Y,
    7. Matsuura S,
    8. Nishihara Y,
    9. Kobayashi K,
    10. Honda H
    : In vitro imaging of human monocytic cellular activity using superparamagnetic iron oxide. Comput Med Imaging Graph 31(8): 638-642, 2007. DOI: 10.1016/j.compmedimag.2007.07.004
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Prediction of Early-stage Liver Fibrosis Using FDG-PET/CT
RIN SHINZATO, AKIHIRO NISHIE, TOMOKO TAMAKI, NAOKI WADA, MITSUHISA TAKATSUKI, GYO IIDA
Anticancer Research Sep 2023, 43 (9) 4221-4227; DOI: 10.21873/anticanres.16614
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords