⁶⁸Ga-Dotatate hybrid PET/MR imaging for non-invasive early detection of heart transplant rejection

Endomyocardial biopsy (EMB) is the gold standard for diagnosing heart transplant rejection, but it is limited by sampling error and has associated morbidity. Cardiac magnetic resonance (CMR) and ¹⁸F-FDG-PET have been introduced as a complementary approach to EMB in the detection of cardiac allograft rejection;¹⁸F-FDG specificity, however, is limited by the uptake of normal myocardium and requires specific patient preparation, with failure of adequate myocardial suppression potentially leading to ambiguous results¹. ⁶⁸Ga-Dotatate, a somatostatin receptor type 2-binding radiotracer, has emerged as an alternative to¹⁸F-FDG for the evaluation of cardiovascular inflammation due to its specific binding to inflammatory cells². In addition, somatostatin receptor imaging has shown promising results in the early identification of acute allograft rejection in a rat model using PET ³ and for rejection prediction in human using SPECT⁴. Because PET imaging has higher effective resolution and sensitivity compared to SPECT, it might improve rejection prediction further. On this basis, we aimed to evaluate whether cardiac ⁶⁸Ga-Dotatate uptake could be a marker of rejection in transplant patients, especially for biopsy-negative cases where rejection is clinically suspected. The study was approved by the Institutional Review Board (HS-17–00371). Imaging protocols are included in Supplemental Material.

⁶⁸Ga-Dotatate-PET/MR was performed in 4 heart transplant patients, 3 with suspected acute allograft rejection presenting with acute heart failure (AHF), and 1 control (Table S1). Patient 1 was admitted for AHF; EMB showed grade 3R (International Society for Heart and Lung Transplantation) acute cellular rejection (ACR) with associated antibody-mediated rejection grade 1(AMR1). She was treated with high-dose steroids and cytolytic agents in accordance with local protocols. Subsequent EMB showed no AMR and ACR(1R). She again presented 10 days later, with evidence of 2:1 atrioventricular (AV) block. A ⁶⁸Ga-Dotatate-PET/MR showed intense uptake in the basal anteroseptum, consistent with the observed conduction abnormality (Figure-1A). Blinded biopsy of the RV septum was read as ACR0 and AMR1, while an imaging-guided EMB (4 days after PET/MR) showed significant mononuclear cell infiltrates in specimens obtained from the basal septum, correlating with the area of ⁶⁸Ga-Dotatate uptake (ACR-1R, AMR1). Average Standardized Uptake Value (SUVmean:0.64±0.13) and Target-to-Background Ratio (TBRmean:1.4±0.3) were higher than the control (SUVmean:0.62±0.06, TBRmean:1.1±0.1). Notably, T2 values on CMR were similarly increased in the basal anteroseptum, colocalizing with ⁶⁸Ga-Dotatate uptake, but average T2 values were normal. Patient 2 was admitted for AHF with EMB showing ACR(1R). After completing treatment for rejection, he underwent ⁶⁸Ga-Dotatate-PET/MR and EMB; both the scan and EMB were negative, which was concordant with his clinical improvement. Six weeks later, he had recurrent AHF symptoms and elevated troponin. A second ⁶⁸Ga-Dotatate-PET/MR revealed uptake in the left lateral wall and interatrial septum up to the AV junction (Figure-2). Following the scan, he developed AV block, which was consistent with the area of inflammation detected by ⁶⁸Ga-Dotatate. Notably, a conventional EMB of the interventricular septum, performed the same day as the PET/MR, was negative. SUVmean and TBRmean were higher in the first scan (visually negative) compared to control (SUVmean:1.16±0.10 vs 0.62±0.06, TBRmean:1.3±0.1 vs 1.1±0.1) but lower than in the second scan (visually positive) (SUVmean:1.20±0.05; TBRmean:1.7±0.1). Concordantly, T2 values were higher in the second than in the first scan; however, they were diffusely increased, without localizing to a specific region of the myocardium. Patient 3 received treatment for ACR(3R), and a negative biopsy was obtained after completion of treatment. 3 weeks later, a ⁶⁸Ga-Dotatate-PET/MR was performed, showing focal uptake in the basal inferior wall. SUVmean and TBRmean were higher than the control (SUVmean:1.48±0.13 vs 0.62±0.06, TBRmean:1.5±0.1 vs 1.1±0.1; Figure-1B). A subsequent EMB (17 days later) was positive for ACR(2R). Increased T2 values in the inferior wall correlated with ⁶⁸Ga-Dotatate findings. Coronary angiography did not show significant cardiac allograft vasculopathy in any of the patients. Patient 4 (control), asymptomatic, underwent a routine EMB with no signs of rejection. ⁶⁸Ga-Dotatate-PET/MR (6 days later) showed no uptake; however, abnormal T2 values were found in anterolateral and apical segments (Figure-1C).

Figure 1. 68Ga-Dotatate-PET/MR imaging in rejection patients versus control.

Representative examples of ⁶⁸Ga-Dotatate-PET/MR imaging in heart transplant patients with (A and B) and without rejection (control, C). Panels show fused PET/MR imaging, and T2 mapping polar map is represented for each patient (T2 values are represented in milliseconds; reference values for T2 mapping are detailed in Supplemental Material). A) shows ⁶⁸Ga-Dotatate uptake in the basal anteroseptum in a patient with allograft rejection (arrows), colocalizing with an increased T2 time in basal anteroseptum, but with normal average T2 values; B) shows ⁶⁸Ga-Dotatate uptake in the inferior wall (arrows), colocalizing with abnormal T2 values, in a patient with allograft rejection; C) shows the absence of ⁶⁸Ga-Dotatate uptake in a patient with a heart transplant and no signs of rejection. Please note abnormal T2 values in mid anterolateral wall and apical segments.

Figure 2. Positive ⁶⁸Ga-Dotatate-PET/MR imaging in a biopsy-negative rejection patient.

Representative examples of fused ⁶⁸Ga-Dotatate-PET/MR imaging. The baseline scan was performed after completing treatment for rejection and showed no evident visual ⁶⁸Ga-Dotatate uptake. The recurrence of heart failure symptoms after 6 weeks led to a follow-up PET/MR (follow-up scan), which showed increased ⁶⁸Ga-Dotatate uptake in the interatrial septum reaching for the atrioventricular junction and in the left lateral ventricular wall (arrows). This patient subsequently developed a complete heart block, likely secondary to the involvement of the atrioventricular junction.

These data support the hypotheses that i) acute allograft rejection can be identified by ⁶⁸Ga-Dotatate-PET/MR, ii) the use of this technique surpasses some of the limitations of EMB, and iii) this hybrid technique is superior to CMR alone.

All the patients with suspected rejection exhibited positive ⁶⁸Ga-Dotatate-PET/MR scans. The presence and location of ⁶⁸Ga-Dotatate uptake correlated well with clinical events such as conduction abnormalities. Furthermore, ⁶⁸Ga-Dotatate-PET/MR overcomes a key limitation of EMB by identifying affected areas inaccessible to biopsy. This technique could help increase the diagnostic yield of EMB by identifying areas of inflammation and guiding subsequent biopsy sampling. Lastly, ⁶⁸Ga-Dotatate-PET/MR can aid in the early identification of rejection in biopsy-negative patients, thus allowing prompt treatment initiation, supporting the diagnosis when histologic findings are equivocal and avoiding multiple biopsies.

PET/MR appears superior to stand-alone CMR although rigorous evaluation was limited in this case series. Increased T2 values, a marker of inflammation/edema, have been described in ACR¹, but with only moderate positive predictive value and limited utility in the early post-transplant stages. In contrast, ⁶⁸Ga-Dotatate-PET/MR, with its ability to detect inflammatory cells directly, accurately identified allograft rejection across a range of durations after transplant (0–13 years). Average T2 values were normal in a patient with rejection, while ⁶⁸Ga-Dotatate-PET/MR was positive. Conversely, T2 values were abnormally increased in the control with a negative ⁶⁸Ga-Dotatate scan, implying a low specificity of these parameters in identifying rejection. Regional distribution of the increased T2 values might also not be specific enough to allow image-guided biopsy. Lastly, ⁶⁸Ga-Dotatate-PET/MR does not require gadolinium-based contrast.

SUVmean and TBRmean were higher in the rejection group than in the control, even when visual ⁶⁸Ga-Dotatate uptake was absent. These values may represent low-grade inflammation that contributes to diffuse myocardial edema, as they correlated well with mean T2 values (Pearson coefficient=0.55 and 0.93), as opposed to SUVmax and TBRmax, which seem to identify focal inflammation.

In conclusion, our data support the use of ⁶⁸Ga-Dotatate-PET/MR imaging as a non-invasive and complementary alternative to EMB in the detection of heart transplant rejection. The use of this technique for the diagnosis and prognosis of acute allograft rejection warrants further investigation.

Abbreviations:

ACR

acute cellular rejection

AHF

acute heart failure

AMR

antibody-mediated rejection

EMB

endomyocardial biopsy

¹⁸F-FDG

¹⁸F-Fluorodeoxyglucose

ROI

region of interest