In this cross-sectional study of dialysis patients, we diagnosed RCC in 7 (4.7%) of the 149 participants included in this study using FDG-PET/CT. This value is consistent with previous reports on the prevalence of RCC among dialysis patients (4.2%) [2]. FDG-PET/CT also showed high sensitivity (100%) and specificity (93%) for the diagnosis of RCC. The high negative predictive value (100%) suggested that this examination is sufficient as a screening tool. There was a significant difference in SUV ratios between the RCC-positive and the RCC-negative groups, but there was a large variance among values; therefore, the clinical utility of this finding is limited.

RCC histology in dialysis patients has been reported to have a distinct tendency. In patients on dialysis for < 10 years found to have RCC, clear cell RCCs have been reported to be the most common, just as they are in non-dialysis patients. In this patient population, 17–18% of RCCs were reported to be ACD-RCCs, although in patients on dialysis for ≥ 10 years, a much higher percentage (46%) of RCCs were ACD-RCCs [5, 12, 13]. Papillary RCCs have also been reported to increase with the duration of dialysis [14]. Papillary RCC is hypovascular on contrast-enhanced CT [15], and ACD-RCC shows equivocal or mild enhancement on imaging [13], making their detection among multiple (frequently hemorrhagic) cysts difficult. These characteristics make FDG-PET a potentially useful screening tool for RCC in patients on dialysis for ≥ 10 years.

Magnetic resonance diffusion-weighted imaging (MRI-DWI) may be more widely available than FDG-PET/CT, making it potentially useful for RCC screening. However, infected and hemorrhagic cysts also have restricted diffusion, potentially contributing to a higher false-positive rate. Studies have examined on the efficacy of choline PET for diagnosing RCC, stating its clearer uptake than FDG-PET [16]. The false-positive rate of choline PET may be lower than that of FDG-PET (because it does not accumulate in infection); however, it is less accessible than FDG-PET/CT, making it a less appropriate screening tool.

Even with the decreased FDG uptake in patients on dialysis, residual normal renal parenchyma can complicate the interpretation of low uptake, making it challenging to distinguish between cancerous and physiologic uptake. Therefore, these images should be interpreted by experienced nuclear medicine physicians. The development of a tracer to replace FDG in tumor detection has long been a focus of researchers. Recent advancements include amino acid tracers that reflect the vivid metabolism of cancer cells and tracers targeting ligands specific to cancer cells. For instance, prostate specific membrane antigen (PMSA), used specifically to diagnose prostate cancer, has been reported to accumulate in RCC, showing an affinity for clear cell cancer in particular [17]. However, clear cell cancer is easily identifiable on CT, limiting the utility if PSMA in this context. Papillary carcinoma and ACD-RCC, which enhance poorly on CT, would benefit from a tracer that accumulates in these cancers for early diagnosis. Unfortunately, PMSA reportedly does not accumulate in papillary carcinoma, and its uptake (or lack thereof) in ADC-RCC has not been adequately studied. This situation limits PMSA’s utility in patients on dialysis, in whom these cancers are more prevalent [17]. Another tracer whose utility in diagnosing RCC has been examined is [68 Ga]Ga-fibroblast activation protein inhibitor-04([68 Ga]Ga-FAPI-04). Compared to FDG-PET, [68 Ga]-FAPI-04 has limited physiologic accumulation in the gastrointestinal tract and kidneys and less excretion into urine [18]. Civan et al. [19] reported primary tumors, local recurrence, and most metastatic lesions of RCC showed a higher SUVmax for FAPI compared to FDG, but that lymph node metastases of RCC showed a higher SUVmax for FDG compared to FAPI. Given the severe decrease in renal function (and FDG excretion) in patients on dialysis, FDG is potentially useful in this limited setting. Conversely, decreased excretion of FAPI into urine might make the detection of RCC even easier. Further studies comparing the potential of FAPI-PET for diagnosing RCC in patients on dialysis with that of FDG-PET are warranted.

In this study, the negative predictive value of FDG-PET/CT for RCC was considered sufficiently high (100%) for screening examination. However, the positive predictive value was low (46.6%) and the false-positive rate was high. Of the eight false-positive participants in our study, one had a renal mass and uptake in the renal vein and paraaortic, mediastinal, and bilateral supraclavicular lymph nodes. The renal mass was biopsied under CT guidance; however, we could not diagnose a malignant tumor, and all lesions spontaneously regressed on clinical follow-up. Although rare, there have been reports of spontaneously regressing RCC [20, 21], and this may have been such a case. Biopsy may have upregulated damage to tumor vascularity [22]. Another case was suspicious for RCC and surgically resected; however, pathology showed atypical epithelium. This lesion was considered to be a precancerous lesion.

FDG accumulation reflects an increase in glucose metabolism, and FDG accumulates not only in tumors but also in infections. The false positives in our study may have also been because of FDG accumulation in infectious lesions. In a report on renal infection of ADPKD or multiple renal cysts diagnosed by FDG-PET, nine of 10 patients exhibited symptoms [23]. None of the false-positive participants in our study complained of fever or abdominal pain; however, we did not perform additional examinations, such as laboratory studies, and the possibility of mild infections remains. These false positives may have been prevented by more detailed patient interviews and additional laboratory studies.

There may be another mechanism for false positives, although only in cases involving ADPKD. In ADPKD, the normal parenchyma remaining between each of the innumerable cysts will have FDG uptake, potentially increasing false positives. In fact, of the six false-positive participants in this study, three had ADPKD. Furthermore, 16% (24/149) of the participants had ADPKD, and these participants may be more likely to show false-positive results than participants with chronic kidney disease of other causes.

Patients on dialysis have minimal FDG excretion into urine, potentially increasing radiation dose. In this study, the protocol standard of our institution was used, and the issue of radiation dose was not explored. Further studies exploring radiation dose optimization would be beneficial to reduce unnecessary risk.

This study had several limitations. First, selection bias was unavoidable. Participants were recruited by displaying leaflets in area hospitals; however, dialysis physicians may have encouraged participation in patients they thought to be at a higher risk of RCC, inadvertently recruiting a higher number of participants with RCC than actually found in patients on dialysis. We did not have a protocol for additional evaluation or clinical follow-up after the FDG-PET/CT examination, leaving these to the discretion of the referring physician and the patient. This situation led to the second and third limitations. The second limitation is that we cannot ascertain if the follow-up period in this study was sufficient. This study was prospective, and recruiting 150 participants took time, leading to a limited observation period in some participants. Third, proving true negatives is challenging. All participants lost to follow-up were in the FDG-PET negative group, raising the possibility that false negatives were among these participants. Among the 107 FDG-negative participants and the eight false-positive participants, only one RCC was detected. This finding is inconsistent with the known RCC prevalence of 4.2% among patients on dialysis. In clinical practice, US is likely the primary screening tool for RCC. Renal US of patients on dialysis may be challenging due to multiple cysts and calcifications, potentially resulting in the oversight of some RCCs. This study intentionally avoided a rigid follow-up protocol, leaving subsequent clinical decisions, including modality and frequency of imaging for follow-up, to referring physicians. Depending on availability, some hospitals performed annual ultrasound, while others performed contrast-enhanced CT or non-enhanced MRI. The flexible protocol allowed participants to be followed-up at the hospitals where they were recruited, making it easier for them to participate in the study. We hope for future large-scale, long-term studies with robust protocols which include specific timelines and imaging modality (or modalities) for follow-up.

In conclusion, we conducted a cross-sectional study of FDG-PET/CT on dialysis patients and found RCC in 4.7% of the participants. This prevalence was obviously higher than that in the general population. This method exploits the lack of physiological FDG accumulation in the urinary tract, and this study showed the potential of this modality in detecting RCC in this patient population. Additional large-scale studies are required to prove its efficacy as a screening tool.