With limitations of conventional imaging and biopsy, accurate, non-invasive techniques to detect clear-cell renal cell carcinoma in patients with renal masses remain an unmet need. 89Zr-labelled monoclonal antibody ([89Zr]Zr-girentuximab) has high affinity for carbonic anhydrase 9, a tumour antigen highly expressed in clear-cell renal cell carcinoma. We aimed to evaluate [89Zr]Zr-girentuximab PET–CT imaging for detection and characterisation of clear-cell renal cell carcinoma.
Methods
ZIRCON was a prospective, open-label, multicentre, phase 3 trial conducted at 36 research hospitals and practices across nine countries (the USA, Australia, Canada, the UK, Türkiye, Belgium, the Netherlands, Spain, and France). Patients aged 18 years or older with an indeterminate renal mass 7 cm or smaller (cT1) suspicious for clear-cell renal cell carcinoma and scheduled for nephrectomy received a single dose of [89Zr]Zr-girentuximab (37 MBq ±10%; 10 mg girentuximab) intravenously followed by abdominal PET–CT imaging 5 days (±2 days) later. Surgery was performed no later than 90 days after administration of [89Zr]Zr-girentuximab. Blinded central review, conducted by three independent readers, determined the histology from surgical samples. The coprimary endpoints, determined for each individual reader, were the sensitivity and specificity of [89Zr]Zr-girentuximab PET–CT imaging to detect clear-cell renal cell carcinoma, with histopathological confirmation as standard of truth. Analyses were on the full analysis set of patients, defined as patients who had evaluable PET–CT imaging and a confirmed histopathological diagnosis. The trial is registered with ClinicalTrials.gov, NCT03849118, and EUDRA Clinical Trials Register, 2018-002773-21, and is closed to enrolment.
Findings
Between Aug 14, 2019, and July 8, 2022, 371 patients were screened for eligibility, 332 of whom were enrolled. 300 patients received [89Zr]Zr-girentuximab (214 [71%] male and 86 [29%] female). 284 (95%) evaluable patients were included in the primary analysis. The mean sensitivity was 85·5% (95% CI 81·5–89·6) and mean specificity was 87·0% (81·0–93·1). No safety signals were observed. Most adverse events were not or were unlikely to be related to [89Zr]Zr-girentuximab, with most (193 [74%] of 261 events) occurring during or after surgery. The most common grade 3 or worse adverse events were post-procedural haemorrhage (in six [2%] of 261 patients), urinary retention (three [1%]), and hypertension (three [1%]). In 25 (8%) of 300 patients, 52 serious adverse events were reported, of which 51 (98%) occurred after surgery. There were no treatment-related deaths.
Interpretation
Our results suggest that [89Zr]Zr-girentuximab PET–CT has a favourable safety profile and is a highly accurate, non-invasive imaging modality for the detection and characterisation of clear-cell renal cell carcinoma, which has the potential to be practice changing.
Funding
Telix Pharmaceuticals.
ZIRCON was a prospective, open-label, multicentre, phase 3 trial to evaluate [⁸⁹Zr] Zr-girentuximab PET-CT imaging for detection and characterisation of clear-cell renal cell carcinoma. Patients with an indeterminate renal mass 7 cm or smaller (cT1) suspicious for clear-cell renal cell carcinoma and scheduled for nephrectomy received a single dose of [⁸⁹Zr] Zr-girentuximab (37 MBq ±10%; 10 mg girentuximab) intravenously followed by abdominal PET-CT imaging 5 days (±2 days) later. Two hundred and eighty-four (95%) evaluable patients were included in the primary analysis. The mean sensitivity was 85.5% (95% CI 81.5–89.6) and mean specificity was 87.0% (81.0–93.1). Positive predictive value was 92.9% and negative predictive value 75.2% with accuracy of 86%.
The authors deserve commendation on a prospective trial in diagnostics in small renal mass (SRM). Could this trial address the issue of overtreatment in small renal masses (SRM) and potentially eliminate the need for invasive biopsies with the use of [⁸⁹Zr] Zr-girentuximab PET-CT?
It should be noted that [⁸⁹Zr] Zr-girentuximab targets carbonic anhydrase IX, which is overexpressed in clear cell renal cell carcinoma (ccRCC) – the most common subtype of malignant RCC. It does not detect papillary or chromophobe renal cancers. The diagnostic tool demonstrated sensitivity of 85.5% indicating that it misses about 14-15 cases (false negatives) with ccRCC. Specificity of 87% indicated that out of 100 negative scans, it will incorrectly identify 13 (false positives) as having the disease.
Given the high PPV of 92.9%, a positive PET-CT result is very likely to indicate true disease presence, while around 7 out of 100 positive scans may be false positives and therefore 7% amongst ccRCC patients may experience overtreatment with surgery or ablation, if active treatment is offered without additional biopsy or other confirmation. A more concerning result in the trial was the negative predictive value (NPV) of 75.2% indicating that in about 1 out of every 4 cases (25%) with a negative scan, the patient could still have the disease and also other subtypes of RCC. Thus, we cannot be certain that a negative scan means the absence of malignancy. The authors recommended using additional contrast-enhanced CT or multiparametric MRI in negative cases to confirm indolent or benign disease. However, this approach raises concerns about the utility of PET/CT, suggesting it should be considered a primary rather than secondary imaging modality in the cancer diagnostic setting.
Another concern is the cost, which should be evaluated in comparison to other diagnostic tools such as biopsy and emerging radiomics.
Despite limitations, it has great potential to guide clinicians and also for further theranostic purposes.