Research ArticleTheranostics

Quantitative Test–Retest Measurement of 68Ga-PSMA-HBED-CC in Tumor and Normal Tissue

Janet H. Pollard, Caleb Raman, Yousef Zakharia, Chad R. Tracy, Kenneth G. Nepple, Tim Ginader, Patrick Breheny and John J. Sunderland
Journal of Nuclear Medicine August 2020, 61 (8) 1145-1152; DOI: https://doi.org/10.2967/jnumed.119.236083

Article Figures & Data

Figures

  • FIGURE 1.
    FIGURE 1.

    68Ga-PSMA-HBED-CC CC PET/CT for subject 2 for scan 1 (baseline) and scan 2, performed 2 d later, showing segmentation of normal tissues (salivary glands and spleen) and sites of abnormal uptake in 2 left pelvic lymph nodes (1 and 2) and left prostate bed (3).

  • FIGURE 2.
    FIGURE 2.

    68Ga-PSMA-HBED-CC CC PET/CT for 2 subjects demonstrating different distribution of metastatic lesions. (A) Predominantly soft-tissue metastatic prostate cancer is seen with visceral metastases in lungs and liver and nodal metastases in neck, chest, abdomen, and pelvis. Percutaneous nephrostomy tubing is also visible. (B) Extensive metastatic prostate cancer is seen involving only skeleton.

  • FIGURE 3.
    FIGURE 3.

    Correlation between test–retest measures of 68Ga-PSMA-HBED-CC uptake. Plots show SUVmean of normal salivary glands and spleen (A and B) and SUVmax for bone and nodal lesions (C and D).

  • FIGURE 4.
    FIGURE 4.

    Comparison of SUV RC point estimates; 95% CIs (y-axis) show substantial overlap, indicating no significant difference in repeatability for different categories of measurements (x-axis).

  • FIGURE 5.
    FIGURE 5.

    Bland–Altman scatterplots show difference between test–retest SUV measurements and average lesion or organ intensity. Plots suggest no association between mean lesion or organ intensity (x-axis) and test–retest relative percentage differences (primary y-axis) based on Kendall τ analysis. Both relative percentage difference (●, primary vertical axis) and absolute difference (○, right vertical axis) are plotted together for salivary glands (A), spleen (B), bone lesions (C), and lymph nodes (D). Horizontal lines represent upper limits of agreement (+1.96 SD), lower limits of agreement (−1.96 SD), and bias (or mean difference). Kendall τ coefficient and P values were calculated for relative percentage difference data and are included in field of graphs.

Tables

  • TABLE 1

    Subject Characteristics

    SubjectPSA within ≤90 d (ng/cm3)Gleason score at diagnosisTotal lesions measuredProstate or prostate bedBonesNodesViscera
    10.157 (4 + 5)—————
    24.356 (3 + 3)31—2—
    3104.59 (4 + 5)15—15——
    40.149 (4 + 5)1019——
    50.669 (5 + 4)4121—
    60.229 (5 + 4)1—1——
    756.3Presumptive diagnosis12—12——
    895.57 (4 + 3)23—2183
    9276.39 (4 + 5)15—15——
    100.04Presumptive diagnosis—————
    110.649 (4 + 5)5—5——
    122.8Lymph node biopsy12183—
    1340.110 (5 + 5)131102—
    1419.77 (3 + 4)2—2——
    152.5Bone biopsy11———
    1654.19 (5 + 4)15—15——
    172.59 (5 + 4)211——
    182.59 (5 + 4)3—21—

    • PSA = prostate-specific membrane antigen.

  • TABLE 2

    Repeatability Data (SUVmax)

    Organ or lesion typewCV (%)Symmetric RC*Asymmetric RC
    LRC (%)URC (%)
    Bone lesions, total (n = 99)11.7± 32.5 [28.5, 37.8]−28.0+38.8
    Location
     Extremities (n = 20)10.7±29.6 [22.5, 43.3]−25.8+34.7
     Pelvis (n = 25)9.5±26.3 [20.6, 36.6]−23.2+30.3
     Skull (n = 9)15.2±42.1 [28.4, 80.6]−34.8+53.3
     Thorax (n = 23)12.7±35.1 [27.1, 49.7]−29.9+42.6
     Vertebrae (n = 22)13.1±36.2 [27.9, 51.8]−30.7+44.2
    Lesion volume, method 1
     <1.5 cm3 (n = 33)8.0±22.1 [17.8, 29.3]−19.9+24.8
     1.5–8.0 cm3 (n = 49)8.7±24.0 [20.1, 30.0]−21.4+27.3
     >8.0 cm3 (n = 17)7.1±19.6 [14.6, 29.8]−17.8+21.7
    Lesion volume, method 2
     <1.0 cm3 (n = 19)7.5±20.9 [15.8, 30.9]−18.9+23.3
     ≥1.0 cm3 (n = 80)8.3±22.9 [19.9, 27.2]−20.5+25.9
     Nodal lesions (n = 27)13.7±37.9 [29.8, 51.9]−31.9+46.8
    Size, long axis
     <1.0 cm (n = 7)9.4±26.1 [16.8, 57.4]−23.1+30.0
     1.0–1.5 cm (n = 15)14.7±40.8 [29.8, 64.3]−33.9+51.3
     >1.5 cm (n = 5)13.3±36.7 [22.0, 100.0]−31.0+44.9
    Prostate or prostate bed (n = 7)10.9±30.3 [19.5, 66.7]−26.3+35.7
    Visceral lesions (n = 3)6.1±16.9 [8.8,100.0]−15.6+18.5

    • ↵* Data are percentages followed by 95% CIs in brackets.

    • LRC = lower RC; URC = upper RC’

  • TABLE 3

    Repeatability Data (SUVmean)

    Organ or lesion typewCV (%)Symmetric RC*
    Salivary glands, total (n = 72)8.9±24.6 [21.1, 29.4]
    Parotid glands (n = 36)9.6±26.5 [21.5, 34.5]
    Submandibular glands (n = 36)8.2±22.8 [18.5, 29.8]
    Spleen (n = 18)10.7±29.6 [22.2, 44.4]
    Bone lesions, total (n = 99)11.2±30.9 [27.2, 36.0]
    Location
     Extremities (n = 20)13.6±37.8 [28.7, 55.2]
     Pelvis (n = 25)9.6±26.7 [20.8, 37.1]
     Skull (n = 9)13.4±37.0 [25.0, 70.9]
     Thorax (n = 23)11.0±30.4 [23.5, 43.1]
     Vertebrae (n = 22)9.7±26.7 [20.7, 38.4]
    Lesion volume, method 1
     <1.5 cm3 (n = 33)6.3±17.4 [14.0, 23.0]
     1.5–8.0 cm3 (n = 49)7.4±20.5 [17.1, 25.6]
     >8.0 cm3 (n = 17)6.2±17.1 [12.7, 26.0]
    Lesion volume, method 2
     <1.0 cm3 (n = 19)7.1±19.6 [17.0, 23.2]
     ≥1.0 cm3 (n = 80)6.2±17.1 [12.9, 25.3]
    Nodal lesions (n = 27)13.2±36.6 [28.9, 50.2]
    Size, long axis
     <1.0 cm (n = 7)8.3±23.0 [14.8, 50.7]
     1.0–1.5 cm (n = 15)13.7±37.8 [27.7, 59.7]
     >1.5 cm (n = 5)17.1±47.3 [28.4, 100.0]
    Prostate or prostate bed (n = 7)8.3±22.9 [14.8, 50.4]
    Visceral lesions (n = 3)2.7±7.6 [4.0, 47.8]

    • ↵* Data are percentages followed by 95% CIs in brackets.

    • LRC = lower RC; URC = upper RC.

Additional Files

Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Keywords