Analysis of RNA-binding protein IMP3 to predict metastasis and prognosis of renal-cell carcinoma: a retrospective study

Abstract

Background: Distant metastasis is the main cause of death from renal-cell carcinoma, and the metastatic potential of tumours is often unpredictable. We aimed to investigate whether IMP3, an oncofetal RNA-binding protein, can be used as a biomarker to predict metastasis and prognosis of renal-cell carcinoma.

Methods: We studied 501 primary and metastatic renal-cell tumours. 371 patients with localised primary tumours were further investigated by use of survival analysis. We assessed IMP3 expression in tumour tissues by immunohistochemistry, and IMP3 mRNA and protein expression in selected tissues by quantitative real-time PCR and western blot analysis.

Findings: Compared with non-metastatic renal-cell tumours, IMP3 expression was greatly increased not only in metastatic tumours but also in a subset of primary tumours that were likely to subsequently develop metastases. Patients with primary localised tumours that did not express IMP3 had a longer metastasis-free survival and overall survival than did those with tumours expressing IMP3 (p<0.0001). Patients with IMP3-positive localised tumours had a much lower 5-year metastasis-free survival than did those with IMP3-negative tumours (for stage I tumours, 44% vs 98%, hazard ratio 17.18 [95% CI 7.82-37.78]; stage II, 41% vs 94%, 10.14 [3.46-29.68]; stage III, 16% vs 62%, 4.04 [2.23-7.31]). IMP3 expression was also associated with reduced 5-year overall survival (stage I, 32% vs 89%, 6.44 [3.63-11.42]; stage II, 41% vs 88%, 6.93 [2.63-18.27]; stage III, 14% vs 58%, 3.46 [1.98-6.05]). Multivariable analysis of IMP3 status (positive vs negative) in primary tumours showed hazard ratios of 5.84 (95% CI 3.60-9.49) for metastasis-free survival and 4.01 (2.66-6.05) for overall survival (both p<0.0001), which were much higher than hazard ratios associated with other independent risk factors.

Interpretation: IMP3 is an independent prognostic marker that can be used at initial diagnosis of renal-cell carcinoma to identify patients who have a high potential to develop metastasis and who might benefit from early systemic treatment.