RCC, the most common malignant tumour of the kidney, consists of different subtypes
10. Malignant tumors of the kidney constitute approximately 2-3% of all malignant tumors and tumor-related deaths, with the highest incidence in developed countries. Male/female=2/1. RCC is a tumor of adulthood and accounts for 85% of all primary renal malignancies
11. It is frequently observed after the age of 40. Although its incidence is highest between 60-70 years of age, the mean age at which it is most common is 55 years. However, it can also be observed rarely in childhood
12,13. An annual increase of approximately 2% is observed in the incidence of the disease worldwide and in Europe, while a decline has been observed in Denmark and Sweden for the last two decades
14. Diagnosis rates increase with developing imaging methods and this leads to a relative increase in incidence. However, it is thought that there is no change in the actual incidence of the tumor
13.
Various grading systems have been developed for RCC15. Firstly, in 1971, Skinner et al. created a grading system evaluating only the shape of the nucleus16. Fuhrman et al. made a new proposal in 1982 to classify RCC according to its nuclear grade. The best single indicator of postoperative metastasis is tumour stage and size. FNG, which later became the most widely recognised grading system for RCC, assuming the role of an independent predictor for prognosis with numerous additional studies.6,17. In the Fuhrman study, he divided the patients into three groups based on survival: the good prognostic group (grade 1), the poor prognostic group (grade 4) and the group in between (grades 2 and 3)18.
Most investigators report that nuclear grading has prognostic validity only in classical and pRCC19.
However, the nuclear grading system has been adopted for the strongest ccRCC and is now applied for this type of tumour in the clinical setting 17.
In 2006, Rioux-Leclercq et al.20 tested the accuracy of various Fuhrman grading schemes to predict survival due to RCC. As a result, the nuclear grading system was accepted as an independent predictor of survival in patients with RCC. FNG system should be taken into consideration when evaluating survival in RCC. However, no difference was observed between modified or traditional Fuhrman grading schemes.
Diffusion-weighted imaging is a functional MRI sequence whose image contrast is based on microscopic movements of water and can be obtained in a very short time using mainly echo planar imaging technique. This sequence does not require the use of contrast agents. In conventional MRI, the diffusion motion of water molecules in the tissue contributes very little to the magnetic resonance signal obtained. In DWI, on the other hand, the movement of water molecules in biological tissues can be measured by applying very strong magnetic field gradients to the area to be imaged. In this way, information can be obtained at the cellular level from the tissue under investigation and important contributions can be made to diagnosis and differential diagnosis by determining the signal characteristics of free or restricted water molecules that change with pathologies. In addition, the fact that it allows quantitative evaluation by ADC measurements is an important superiority over other methods21-24.
Indeed, a strong correlation between ADC and grade has been shown in two tumors. These are gliomas and prostatic adenocarcinomas and tumor cellularity is an important determinant. ADC should be considered as a complex variable reflecting tissue characteristics as well as cellularity17.
Magnetic resonance imaging is the only method available today for the assessment of molecular diffusion processes in vivo. The kidney is an organ of interest for ADC study due to its high blood flow rate and its role in water transport25.
The basic scale defining diffusion sensitivity in diffusion-weighted imaging is the b value. It is known that examinations performed with a high b value (1000-1200 sec/mm2) increase the sensitivity to diffusion by minimizing the T2 effect in tissues26.
At high b values (800 or 1000 sec/mm2), images are obtained with more signal loss from water molecules. Tissues with a high degree of diffusion restriction show bright signal areas in the images at high b values, and low signal intensity in the corresponding ADC map. Diffusion MRI can help to characterize renal lesions in a non-invasive manner. In the characterisation of renal masses, diffusion MRI has a promising role to play. Solid RCC is a highly cellular tumour and usually has a high signal intensity compared to normal parenchyma on high b-value images. Conversely, benign cysts and low cellularity masses usually show low diffusion restriction and low signal intensity on high b-value images. However, RCC may have various diffusion MR appearances due to variable degree of cellularity, cystic change, necrosis and presence of hemorrhagic elements27.
In addition, ADC was shown to correlate with cellular diversity in benign and malignant tissue. ADC may therefore play a role in predicting RCC grade. If this is confirmed, ADC may be effective in determining the most appropriate treatment option to use with pre-operative imaging in a given case17.
Recent studies have revealed the potential of ADC in the assessment of various conditions such as pionephrosis, infection, renal ischaemia and diffuse renal disease.25.
In a meta-analysis conducted by Lassel et al. in 2013, 17 studies with 764 patients were included and according to this study, ADC values of RCCs were significantly lower than normal parenchyma. Uroepithelial cancers could be differentiated with low ADC values. ADC values differed significantly between RCC and oncocytoma. It was concluded that ADC values may help to differentiate malignant and benign renal tumors and may reduce inappropriate nephrectomies by differentiating oncocytoma from malignant tumors28.
In our study using high b (b 1000) values, the mean ADC values of RCCs (1.38x10-3 sec/mm2) were found to be lower than the mean ADC values of normal renal parenchyma (2.10x10-3 sec/mm2) in accordance with the literature (p=0.002).
In a study by Sasamori et al.29 in 2014, 31 patients with renal mass underwent diffusion imaging with preoperative 50, 500 and 1000 b values. It was observed that the mean ADC values were significantly higher in RCC compared to urothelial carcinoma (p<0.05) and the mean ADC value was lower in angiomyolipoma compared to RCC (p<0.01). In the differentiation of solid renal tumours, ADC values generated by 3T diffusion MRI may be useful.
Paudyal et al.30 evaluated diffusion MRI for the characterisation of renal carcinoma. In this study of 47 cases, ADC values were found to be significantly higher in RCCs compared to urothelial carcinomas (p=0.022). In histologic subtype analysis, a significant difference in ADC values was observed between clear cell (1.59x10-3 sec/mm2) and non-clear cell (6.72x10-3 sec/mm2) RCC (p=0.0004). Similarly, while there was a significant difference between ADC values in RCC lesions with and without metastases (p=0.0004), no difference in intensity was observed in T1 and T2 weighted images.
Maruyama et al.31 evaluated the usefulness of tumor size and ADC/dimension ratio in differentiating low and high grade tumors in a study of 49 cases, 34 of 49 ccRCCs were low grade and 15 were high grade tumours. There were significant differences in ADC values, tumour size and ADC/size ratio between high and low grade tumours (p<0.05). There was also a correlation between tumour size and ADC value (p<0.01). There was statistically significant difference between high-grade ccRCC and low-grade ccRCC by tumor size and ADC/size ratio.
In our study, although a decrease in ADC value was observed with increasing tumor diameter, it was not statistically significant. Again, an increase in nuclear grade was found with increasing tumor diameter, which was not statistically significant. We attributed this to our low number of cases.
Cogley et al.27 examined diffusion MRI of RCC, urothelial carcinoma and renal infections in 2013. ADC values were lower for solid components than for necrotic or cystic areas in complex renal masses. Restricted diffusion areas in complex solid and cystic renal masses were found to be helpful in differentiating complicated cysts, cystic or necrotic areas and RCC with MRI without contrast agent.
Doğanay et al.32 investigated the role and dependability of DW-MRI in the differentiation of malignant and benign renal lesions. In this study, DW-MRI was used to differentiate angiomyolipoma and oncocytoma from RCC and malignant lesions from benign lesions. Higher b values (b 600 and b 1000) were found to have higher sensitivity and specificity values.
Sandrasegaran et al.33 examined the role of DWI in differentiating renal mass subtypes in 42 patients with renal masses. Benign lesions were found to have a higher mean ADC than malignant lesions in this study. ADC measurements have been used for the differentiation of benign cystic lesions from cystic kidney cancer.
Goyal et al.34 in a study of 33 cases with 36 RCCs in 2012, histologic subtype, nuclear grade and cell count were performed for each lesion, and the relationship between ADC values and cell count of different grades and subtypes was investigated. Out of 23 low grade (grade 1 and 2) and 13 high grade (grade 3 and 4) tumors, 32 were clear cell and 4 were nccRCC. It was observed that the grade increased with decreasing ADC values. Mean ADC values were significantly higher for ccRCC (1.62x10-3 sec/mm2) than for nccRCCs (1.04x10-3 sec/mm2) (p=0.005). Only low-grade RCCs and ccRCC were found to have higher ADC values.
In a study by Wang et al.35, diffusion imaging was performed with 3T MRI using 500 and 800 b values in 83 patients with 85 kidney masses. It was found that 49 of the cases were ccRCC, 22 were pRCC and 14 were crRCC. At 500 b values, ADC values of ccRCC were significantly higher than other subtypes (p=0.001). The difference between pRCC and crRCC was not significant (p=0.68). At 800 b values, ccRCC showed the highest mean ADC value of the subtypes and the difference with each of the subtypes was statistically significant (p<0.001). The mean ADC at 800 b values was more effective in differentiating ccRCC and nccRCC (ROC= 0.973). 1.281x10-3 sec/mm2 was the threshold value that allowed differentiation with a sensitivity of 95.9% and specificity of 94.4%.
In our study, the average ADCs were higher in ccRCC than in other subgroups, which is consistent with the literature. Contrary to the literature27, a statistically significant difference was observed when comparing subtype mean ADCs individually in our study (p=0.032).
Yu et al.36 In a study consisting of diffusion MR images obtained with 800 b values on 137 RCC patients, ADC values taken from the solid parts of the tumors and the normal parenchyma of the opposite kidney were measured and statistically analyzed. Mean ADC values were significantly lower for RCC than normal renal parenchyma (p<0.001). A significant difference in ADC values was observed between clear cell and non-clear cell RCCs, and a significant difference was also observed between G1 and G2 and between G3 and G4 of RCCs. ADC was found to be useful in the characterization of subtypes and nuclear grades of RCC.
Rosenkrantz et al.17 retrospectively evaluated the usefulness of ADC in differentiating low and high grade ccRCC in 57 ccRCC patients with pathological diagnosis and preoperative DW MRI. A total of 31 low-grade (1 G1 and 30 G2) and 26 high-grade (20 G3 and 6 G4) RCCs were analysed. In both low-grade and high-grade ccRCCs, ADC values at 400 and 800 b were significantly lower in high-grade ccRCCs (p<0.001), demonstrating its effectiveness in detecting high-grade ccRCCs compared with conventional MRI features alone. This study showed that the average lesion size in high-grade ccRCC was larger than in low-grade ccRCC (p<0.006).
In our study, we compared the ADC values of the patients who were grouped according to their histopathologic nuclear grades. Although the literature generally focuses on ccRCC, in our study, subtypes of RCCs were included in different numbers in the groups. As a result of the statistical comparison between the three groups (G1, G2 and G3), a significant difference was found between G1 and G2 in accordance with the study of Yu et al.36. However, we observed no significant difference between other groups. In this study, we compared RCC ADC values with normal kidney ADC values for each group. We found a statistically significant decrease in ADC values of RCC compared to normal kidney parenchyma.
Silva et al.37 tried to differentiate benign and malignant tumors with ADC values in a study of 66 patients with renal tumors. Oncocytoma was found to have the highest ADC value. The ADC value decreased in the order of ccRCC (1.5033±0.1328), crRCC (1.1075±0.1034), pRCC (0.7611±0.0942) and angiomyolipoma. ADC values of RCC subtypes were very close to the values in our study and showed a similar ranking.
Sharma and et.al.38 investigated the utility of chemical shift imaging and DW/ADC maps in differentiating renal tumors. In their study, they measured the ADC values of patients with 15 ccRCC, 1 crRCC, and 3 pRCC as 1.29x10-3mm2, 0.80x10-3mm2, 0.68x10-3mm2, respectively. They found no correlation between ADC values and tumor subtypes. This may be due to the small sample size. In our study, despite a similar sample size, there was a statistically significant difference between the subtypes (p=0.032).
Li and et.al.39 calculated ADC values in a study group consisting of 68 ccRCC (clear cell renal cell carcinoma) and 32 nccRCC patients (21 crRCC and 11 pRCC). The mean ADC values for ccRCC, crRCC, and pRCC were 2.85±1.35x10-3mm2, 1.42±0.78x10-3mm2, 1.34±0.52x10-3mm2, respectively. As in our study, they found a statistically significant difference between ccRCC and nccRCC (p<0.001). However, they were unable to distinguish between crRCC and pRCC using ADC values. In contrast, our study was able to differentiate between the subtypes using ADC values.
The unequal number of cases in the groups, the inhomogeneous distribution of histologic subtypes within the groups, and the absence of a G4 group were the shortcomings of our study. In addition, since DWI was obtained without breath holding, motion artifacts affected the image quality.
In conclusion, preoperative DWI holds promise in contributing to abdominal MRI data and prognostic evaluations due to its ability to detect histologic subtypes of RCCs. This technique offers advantages such as rapid acquisition, ease of use without the need for contrast material, and the potential to differentiate cases based on their nuclear grades. Integrating DWI into clinical practice may enhance our ability to stratify RCC patients prognostically, leveraging its non-invasive nature and capability to provide valuable histological insights. However, we concluded that our findings should be supported by larger and multicenter studies.