Abstract
Objectives:
Imaging appearances of immune checkpoint inhibitor-related nephritis have not yet been described. The primary objective of this study is to describe the appearances of immunotherapy-related nephritis on computerized tomography (CT) and positron emission tomography (PET). The secondary objectives are to investigate the association of radiologic features with clinical outcomes.
Methods:
CT and PET-CT scans before the initiation of immunotherapy (baseline), at nephritis and after resolution of pathology-proven nephritis cases were reviewed. Total kidney volume, renal parenchymal SUVmax, renal pelvis SUVmax and blood pool SUVmean were obtained.
Results:
Thirty-four patients were included. The total kidney volume was significantly higher at nephritis compared to baseline (464.7±96.8 mL vs. 371.7±187.7 mL; p<0.001). Fifteen patients (44.1%) had >30% increase in total kidney volume, which was associated with significantly higher renal toxicity grade (p=0.007), higher peak creatinine level (p=0.004) and more aggressive medical treatment (p=0.011). New/increasing perinephric fat stranding was noted in 10 patients (29.4%) at nephritis. Among 8 patients with contrast-enhanced CT at nephritis, one (12.5%) developed bilateral wedge-shaped hypoenhancing cortical. On PET-CT, the renal parenchymal SUVmax-to-blood pool ratio was significantly higher at nephritis compared to baseline (2.13 vs. 1.68; p=0.035). The renal pelvis SUVmax-to-blood pool SUVmean ratio was significantly lower at nephritis compered to baseline (3.47 vs. 8.22; p=0.011).
Conclusions:
Bilateral increase in kidney size, new/increasing perinephric stranding and bilateral wedge-shaped hypoenhancing cortical foci can occur in immunotherapy-related nephritis. On PET-CT, diffuse increase in radiotracer uptake throughout the renal cortex and decrease in radiotracer activity in the renal pelvis can be seen.
Keywords: Immunotherapy, Nephropathy, Toxicity, Radiology, Radiographic
Introduction
Novel immune checkpoint inhibitor antineoplastic agents (e.g. ipilimumab, pembrolizumab, nivolumab, atezolizumab) have a rapidly expanding role in the management of several malignant disorders due to their better efficacy, tolerability and less toxic side effect profile compared to conventional cytotoxic agents. However, these agents have been associated with immune related phenomena and toxicities in nearly all organ systems which are termed immune related adverse events (irAE)[1].
Immune checkpoint inhibitor-related nephritis is a relatively uncommon condition occurring in about 2.2–3.5% of patients with a higher incidence in patients receiving a combination of immune checkpoint inhibitors as opposed to monotherapy[2; 3]. Renal toxicity occurs between 6–56 weeks after initiation of immune checkpoint inhibitors[4]. Patients are usually asymptomatic but non-specific signs of renal impairment (e.g. dysuria, oliguria, edema, anorexia) have been described[5]. This entity is suspected upon recognition of worsening renal function tests after other causes of renal dysfunction have been excluded. However, tissue sampling is necessary to definitely establish the diagnosis[5–7]. Acute tubulointerstitial nephritis is the most common histopathologic pattern identified in patients with immune checkpoint inhibitor-related nephritis[4].
To our knowledge, radiologic appearances of patients with biopsy proven immune checkpoint inhibitor-related nephritis have not been described[8–12]. The primary objective of this study is to describe the imaging characteristics of pathology proven cases of immune checkpoint inhibitor-related nephritis. The secondary objective is to investigate whether abnormalities seen on radiologic imaging are associated with other clinical outcomes such as disease severity, type of therapy, long-term persistence/resolution of kidney dysfunction and overall mortality.
Materials and Methods
After institutional review board approval, the electronic medical records were screened for patients with biopsy proven immune checkpoint inhibitor-related nephritis between February 2017 and April 2021. Patients’ demographic data, type of malignancy, baseline kidney function tests, immune checkpoint inhibitor regimen, number of immune checkpoint inhibitor cycles, concurrent chemotherapy regimen, concurrent nephrotoxic medications, clinical risk factors for kidney dysfunction (hypertension, diabetes mellitus, coronary artery disease and hyperlipidemia), date of acute kidney injury, peak serum creatinine level, histopathologic pattern on biopsy, interstitial fibrosis and tubular atrophy (IFTA) score on biopsy, management of immunotherapy related nephritis and persistence/resolution of kidney dysfunction after management were collected form the electronic medical records. Tumor response until the time of data collection and all-cause mortality were also recorded.
The toxicity grade of acute kidney injury was graded as follows: up to 1.5-fold increase in serum creatinine above baseline is considered grade 1, 1.5–3.0-fold increase is grade 2, 3–6 fold increase is grade 3 and >6 fold increase is grade 4. Response of renal function was categorized as follows: complete renal response in cases when the final serum creatinine was within 0.3mg/dL of its baseline value, partial renal response if serum creatinine increased >0.3mg/dL above baseline and no response was defined as progression to kidney failure necessitating kidney replacement therapy (i.e. dialysis)[2; 4].
All abdominal computerized tomography (CT), chest CT, abdominal magnetic resonance imaging (MRI), ultrasound and positron emission tomography-computerized tomography (PET-CT) studies were initially screened by a fellowship trained radiologist with 1 year of experience starting from the last imaging study before initiation of immune checkpoint inhibitor up to the most recent available scan. The radiologist was aware of the date of treatment initiation and final tissue diagnosis, but he was blinded to the date at which immune checkpoint inhibitor-related nephritis was identified. On initial screening, an increase in kidney size, new/increase in pre-existing perinephric fat stranding, appearance of cortical hypoenhancing soft tissue lesions, diffuse increase in F18-flourodeoxyglucose (F18-FDG) uptake throughout the renal parenchyma and a decrease in F18-FDG excretion in the urinary collecting system were identified and these findings corresponded with the onset of acute kidney injury. Patients without available pre-treatment imaging studies and/or imaging studies at the time of nephritis were excluded. Patients with or without post-treatment imaging studies were included in the study.
Subsequently two fellowship trained abdominal radiologists with 21 and 1 year of experience reviewed the scans and collected data from pre-nephritis (baseline) scans, scans at the time of acute kidney injury (i.e., between the onset of increase in serum creatinine and before treatment initiation) and from scans post-nephritis (defined as return of kidney function tests to baseline or reaching a new plateau after >3 months of treatment). Consensus between the two radiologists was reached on all measurements.
Several dedicated CT and PET-CT scanners from various vendors were used to obtain images. CT slice thickness varied between 2.5–5.0mm. Multiplanar reconstruction was used to generate oblique coronal and oblique axial images along the long axis of the kidney on which the craniocaudal length, width and thickness of the kidney were measured. The total kidney volume was calculated by summing up the volume of both kidneys[13]. The renal volume was calculated using the following formula[14]:
The presence of new/increasing perinephric stranding and hypoenhancing cortical soft tissue lesions were recorded from CT scans.
On PET-CT scans, the SUVmax of renal parenchyma and renal pelvis were recorded. To obtain the SUVmax of the renal parenchyma, the region of interest (ROI) was placed on the renal cortex avoiding the renal medulla and adjacent pelvicalyceal system. The ROI was placed on the renal pelvis to obtain the renal pelvis SUVmax. The highest SUVmax from either kidney were used as the parenchymal SUVmax and renal pelvis SUVmax and it was included in the calculations. The SUVmean of blood pool was obtained by placing ROIs at the lumen of the abdominal aorta. The ratios of renal parenchyma SUVmax-to-blood pool SUVmean and the renal pelvis SUVmax-to-blood pool SUVmean were calculated. An ROI volume of 1.5 cm3 was used for making the SUV measurements.
Statistical analysis was performed using SPSS software version 24 (IBM). Descriptive statistics were used to summarize the data. Wilcoxon ranked test was used to analyze the differences in total kidney volume, renal parenchyma SUVmax, renal pelvis SUVmax, ratio of renal parenchyma SUVmax-to-blood pool SUVmean and ratio of renal pelvis SUVmax-to-blood pool SUVmean on baseline, nephritis and post-treatment scans. In order to investigate the association of imaging findings and clinical parameters, the cohort was dichotomized according to the presence of >30% increase in total kidney volume, >40% increase in renal parenchymal SUVmax-to-blood pool SUVmean ratio and >40% decrease in renal pelvis SUVmax-to-blood pool SUVmean between the baseline scans and scans at nephritis. Independent T-test was used for continuous variable analysis, Pearson chi squared test and Fisher’s exact test were used for categorical data analysis and Kurksal-Wallis test was used for ordinal data analysis.
Results
A total of 38 patients with biopsy proven immune checkpoint inhibitor-related nephritis were identified between February 2017 and April 2021. Two patients were excluded due to absence of imaging studies at the time of nephritis and 2 patients were excluded due to severe hydronephrosis secondary to pelvic malignancy which could confound the interpretation. Thirty-four patients were included in the final analysis all of whom had CT imaging at baseline and at nephritis; but 25 of whom had CT scans after nephritis (figure 1).
Figure 1:
Study flow chart. CT: computerized tomography; PET-CT: Positron Emission Tomography-Computerized Tomography.
The median age of the study cohort was 68 years (range 37–81) and 24 (70.6%) were male. Eleven patients had malignant melanoma, 8 had lung cancer, 4 had urothelial carcinoma, 2 had renal cell carcinoma, 2 had thyroid cancer, 2 had Hodgkin’s lymphoma, 1 had non-Hodgkin’s lymphoma, 1 had smoldering myeloma, 1 had chronic lymphocytic leukemia, 1 had pancreatic ductal adenocarcinoma and 1 patient had both renal cell carcinoma and chronic myelogenous leukemia.
CT findings
CT images at nephritis were available for all 34 patients. Eight patients (23.5%) had contrast-enhanced CT images at the time of nephritis and non-contrast enhanced CT images were available for the remaining patients.
The total kidney volume was significantly higher at nephritis compared to baseline (464.7±96.8 mL vs. 371.7±187.7mL; p<0.001) (figure 2). The total kidney volume at post-nephritis was significantly lower than its baseline value (304.2±106.9 mL vs. 371.7±187.7 mL; p<0.001) and less than the total kidney volume at nephritis (p<0.001) (table 1).
Figure 2:
Axial CT images of a 66-year-old male with malignant melanoma who developed immune checkpoint inhibitor related nephritis after 3 cycles of a pembrolizumab, ipilimumab and nivolumab. Axial contrast enhanced CT image at baseline (a) and at nephritis (b). His total kidney volume increased from 435 mL at baseline to 726 mL at nephritis (67% increase in total kidney volume). Note the bilateral renal enlargement and new perinephric stranding (arrowheads).
Table 1:
CT and PET-CT parameters before, during and after immune checkpoint inhibitor related nephritis.
| Mean | Standard deviation | Range | baseline vs. nephritis | p-value baseline vs. post nephritis | nephritis vs. post-nephritis | |
|---|---|---|---|---|---|---|
| Total kidney volume (n=34) | ||||||
| Baseline (mL) | 371.7 | 96.8 | 224.8–620.9 | <0.001 | <0.001 | <0.001 |
| During nephritis (mL) | 464.7 | 187.7 | 248.3–1089.5 | |||
| Post nephritis (mL)* | 304.2 | 106.9 | 191.4–629.1 | |||
|
| ||||||
| Renal parenchyma SUVmax (n=14) | ||||||
| Baseline | 3.4 | 0.9 | 1.7–5.0 | 0.051 | 0.734 | 0.138 |
| During nephritis | 4.4 | 1.3 | 3.2–7.1 | |||
| Post nephritis** | 3.2 | 1.0 | 1.7–5.2 | |||
|
| ||||||
| Urine SUVmax (n=14) | ||||||
| Baseline | 16.4 | 14.3 | 3.2–54.1 | 0.009 | 0.038 | 0.515 |
| During nephritis | 7.4 | 4.0 | 1.3–15.3 | |||
| Post nephritis** | 9.1 | 9.5 | 2.3–32.1 | |||
|
| ||||||
| Blood pool SUVmean (n=14) | ||||||
| Baseline | 2.1 | 0.5 | 1.1–2.9 | 0.502 | 0.523 | 1.000 |
| During nephritis | 2.1 | 0.4 | 1.6–2.8 | |||
| Post nephritis** | 2.0 | 0.4 | 1.4–2.8 | |||
|
| ||||||
| Renal parenchyma SUVmax-to-blood pool SUVmean ratio (n=14) | ||||||
| Baseline | 1.68 | 0.31 | 1.13–2.19 | 0.035 | 0.859 | 0.260 |
| During nephritis | 2.13 | 0.62 | 1.48–3.94 | |||
| Post nephritis** | 1.66 | 0.60 | 0.89–3.06 | |||
|
| ||||||
| Renal pelvis SUVmax-to-blood pool SUVmean ratio (n=14) | ||||||
| Baseline | 8.22 | 6.76 | 1.10–25.76 | 0.011 | 0.051 | 0.441 |
| During nephritis | 3.47 | 1.83 | 0.68–7.29 | |||
| Post nephritis** | 4.78 | 5.52 | 1.11–18.88 | |||
Follow up imaging post-nephritis was not available for 9 patients; hence calculations were made for the remaining 25 patients.
Follow up PET-CT post-nephritis was not available for 5 patients; hence calculations were made for the remaining 9 patients.
In order to investigate the association of total kidney volume with other clinical and demographic data the patients were dichotomized according to the increase in total kidney volume using a cutoff value of 30% (table 2).
Table 2:
Patient characteristics according to change in total kidney volume at nephritis compared to baseline.
| All patients (n=34) | <30% increase in total kidney volume (n=19) | >30% increase in total kidney volume (n=15) | p-value | |
|---|---|---|---|---|
| Mean age ± standard deviation, (year) | 64.4 ± 12.3 | 64.3 ± 13.7 | 64.5 ± 10.8 | 0.972 |
| Gender | 0.128 | |||
| Male | 24 (70.6) | 11 (57.9) | 13 (86.7) | |
| Female | 10 (29.4) | 8 (42.1) | 2 (13.3) | |
|
| ||||
| Nephrologic risk factors | ||||
| Hypertension | 21 (61.7) | 10 (52.6) | 11 (73.3) | 0.217 |
| Hyperlipidemia | 10 (29.4) | 6 (31.6) | 4 (26.7) | 1.000 |
| Diabetes Mellitus | 8 (23.5) | 4 (21.1) | 4 (26.7) | 1.000 |
| Coronary artery disease | 1 (2.9) | 0 (0.0) | 1 (6.7) | 0.441 |
| NSAIDs | 2 (5.8) | 2 (10.5) | 0 (0.0) | 0.492 |
| Proton pump inhibitors | 11 (32.4) | 7 (36.8) | 4 (26.7) | 0.715 |
| Concurrent non-CPI chemotherapy | 15 (44.1) | 9 (47.4) | 6 (40.0) | 0.667 |
|
| ||||
| CPI immunotherapy | 0.370 | |||
| Monotherapy | 28 (82.3) | 17 (89.5) | 11 (73.3) | |
| Multidrug regimen | 6 (17.6) | 2 (10.5) | 4 (26.7) | |
|
| ||||
| Specific CPI drug | ||||
| Pembrolizumab | 11 (32.4) | 6 (31.6) | 5 (33.3) | 1.000 |
| Nivolumab | 19 (55.9) | 10 (52.6) | 9 (60.0) | 0.667 |
| Ipilimumab | 6 (17.6) | 2 (10.5) | 4 (26.7) | 0.370 |
| Atezolizumab | 3 (8.8) | 2 (10.5) | 1 (6.7) | 1.000 |
| Durvalumab | 2 (5.8) | 1 (5.3) | 1 (6.7) | 1.000 |
|
| ||||
| Number of CPI cycles | 10.1 ± 20.6 | 14.7 ± 26.8 | 4.3 ± 3.4 | 0.148 |
| Duration between CPI initiation to nephritis (days) | 191.1 ± 187.9 | 240.6 ± 232.3 | 128.3 ± 79.0 | 0.061 |
|
| ||||
| Kidney function | ||||
| Baseline creatinine, mg/dL | 1.13 ± 0.48 | 1.16 ± 0.59 | 1.10 ± 0.30 | 0.737 |
| Peak creatinine, mg/dL* | 3.60 ± 1.94 | 2.79 ± 1.46 | 4.64 ± 2.03 | 0.004 |
| Final creatinine, mg/dL** | 1.61 ± 1.04 | 1.53 ± 0.55 | 1.72 ± 1.59 | 0.659 |
| Baseline-to-peak creatinine difference, mg/dL | 2.47 ± 1.81 | 1.63 ± 1.10 | 3.54 ± 2.01 | 0.003 |
| Baseline-to-peak creatinine change, % | 236.4 ± 189.9 | 152.4 ± 101.8 | 342.8 ± 223.5 | 0.007 |
| Baseline-to-final creatinine difference, mg/dL | 0.67 ± 1.07 | 0.54 ± 0.40 | 0.88 ± 1.75 | 0.473 |
| Baseline-to-final creatinine change, % | 61.16 ± 115.2 | 52.0 ± 51.4 | 76.2 ± 179.7 | 0.593 |
| Dialysis | 5 (14.7) | 1 (5.3) | 4 (26.7) | 0.146 |
| Other immunotherapy related adverse effects | 14 (41.1) | 6 (31.6) | 8 (53.3) | 0.201 |
|
| ||||
| Toxicity grade | 0.007 | |||
| Grade 1 | 1 (2.9) | 1 (5.3) | 0 (0.0) | |
| Grade 2 | 14 (41.2) | 10 (52.6) | 4 (26.7) | |
| Grade 3 | 10 (29.4) | 7 (36.8) | 3 (20.0) | |
| Grade 4 | 9 (26.4) | 1 (5.3) | 8 (53.3) | |
|
| ||||
| Treatment | 0.011 | |||
| CPI cessation only | 5 (14.7) | 5 (26.3) | 0 (0.0) | |
| CPI cessation and steroid only | 27 (79.4) | 14 (73.7) | 13 (86.7) | |
| CPI cessation, steroid and infliximab | 2 (5.8) | 0 (0.0) | 2 (13.3) | |
|
| ||||
| Renal function after nephritis | 0.717 | |||
| Creatinine within 0.3 mg/dL of baseline value | 12 (35.3) | 6 (31.6) | 6 (40.0) | |
| Creatinine >0.3 mg/dL of baseline value | 17 (50.0) | 12 (63.2) | 5 (33.3) | |
| Dialysis | 5 (14.7) | 1 (5.3) | 4 (26.7) | |
|
| ||||
| Biopsy | ||||
| Acute tubular necrosis (ATN) | 25 (73.5) | 13 (68.4) | 12 (80.0) | 0.697 |
| Acute interstitial nephritis (AIN) | 29 (85.3) | 14 (73.7) | 15 (100.0) | 0.053 |
| Both AIN and ATN | 21 (61.8) | 9 (47.4) | 12 (80.0) | 0.079 |
| Interstitial fibrosis and tubular atrophy (IFTA, %) | 19.1 ± 17.0 | 22.5 ± 18.6 | 14.6 ± 14.1 | 0.199 |
|
| ||||
| Cancer response | 0.709 | |||
| Disease progression | 19 (55.8) | 11 (57.9) | 8 (53.3) | |
| Partial response | 3 (8.8) | 2 (10.2) | 1 (6.7) | |
| Remission | 12 (35.3) | 6 (31.6) | 6 (40.0) | |
|
| ||||
| Prognosis | ||||
| Death | 11 (32.4) | 7 (36.8) | 4 (26.7) | 0.715 |
| Duration between nephritis and death (days) | 463.3 ± 479.9 | 250.0 ± 98.6 | 961.0 ± 690.1 | 0.215 |
including patients who eventually progressed to dialysis.
Excluding patients who eventually progressed to dialysis.
The median change in total kidney volume was 32%. This value was rounded to the nearest ten and a cutoff value of 30% was selected to dichotomize the cohort into two groups. Among 34 patients, 15 (44.1%) had >30% increase in total kidney volume. No statistically significant difference was noted in demographic characteristics, nephrologic risk factors, type of immune checkpoint inhibitor therapy or baseline creatinine value between the two groups. The duration between CPI initiation and nephritis was approximately half as long in patients with >30% increase in total kidney volume but this difference did not reach statistical significance (128.3±79.0 days vs. 240.6±232.3 days; p=0.061). Patients with >30% increase in total kidney volume had significantly higher toxicity grade and higher peak creatinine (p<0.007 and p<0.004). However, there was a significant difference in the type of clinical management employed in each group (p=0.011) with more aggressive medical therapy being utilized in patients with >30% increase in total kidney volume (13.3% were managed by a combination of CPI cessation, steroids and infliximab as opposed to none receiving this combination in patients with <30% change in total kidney volume; none of the patients with >30% increase in total kidney volume was managed by CPI cessation only as opposed to 26.3% in the <30% group) (table 2). Permanent kidney replacement therapy (dialysis) was necessary in 4 (26.7%) patients with >30% increase in total kidney volume as opposed to 1 (5.3%) in the <30% group but this did not reach statistical significance (p=0.146). There was no significant difference between the two groups in terms of post-nephritis serum creatinine levels (excluding dialysis patients), occurrence of other immunotherapy related adverse events, histopathologic characteristics on kidney biopsy, overall cancer response, death or time-to-death (table 2).
New/increasing perinephric fat stranding was noted in 10 patients (29.4%) at nephritis. Among 15 patients with >30% increase in total kidney volume 9 had new/increasing perinephric stranding (60.0%). The remaining patient with new perinephric stranding had 15.0% increase in total kidney volume.
Among 8 patients with contrast-enhanced CT at nephritis, 4 had >30% increase in total kidney volume, one of whom additionally had ill-defined wedge-shaped hypoenhancing cortical foci. On post-nephritis follow up scans, scarring and cortical thinning developed at the site of these hypoenhancing foci (figure 3) and his serum creatinine was 0.93mg/dL at baseline, it peaked at 3.18 mg/dL during nephritis and it plateaued at 2.25 mg/dL after nephritis. On the contrary to this patient, focal scarring was not identified on post contrast studies in any of the other patients after nephritis.
Figure 3:
Axial CT images of a 54-year-old male with lung adenocarcinoma receiving a pembrolizumab at baseline (a), at nephritis (b), one month after nephritis (c) and at 4 months after nephritis (d). He developed immunotherapy related nephritis after 5 cycles of immunotherapy. His total kidney volume increased from 394 mL at baseline to 612 mL at nephritis (55% increase in total kidney volume). Note the wedge shaped hypoenhancing cortical foci seen at nephritis (b) and 1 month post nephritis CT scans (c) (arrowheads). Note the focal cortical thinning and scarring which developed at the sites of these lesions 4 months after management (d) (arrows).
PET-CT findings
PET-CT images were available for 14 patients at baseline and at nephritis; 9 of whom also had post-nephritis PET-CT images (figures 4).
Figure 4:
Maximum intensity projection (MIP) images and fused coronal reformatted F18-FDG PET-CT images at baseline (a) and at nephritis (b) of a 76-year-old male with lung cancer. Note the bilateral increase in radiotracer uptake throughout the renal parenchyma at nephritis (SUVmax 6.3) compared to baseline (SUVmax 1.7) (arrows). Also note the decrease of FDG activity in the renal pelvis and collecting system at nephritis (SUVmax: 6.3) compared to baseline scan (SUVmax: 17.3) (dotted arrows).
The renal parenchyma SUVmax was higher at nephritis compared to baseline PET-CT but it did not reach statistical significance (4.4 vs. 3.4; p=0.051). The renal pelvis SUVmax was significantly lower at nephritis compared to baseline (7.4 vs. 16.4; p=0.009). The renal pelvis SUVmax was significantly lower on post-nephritis PET-CT compared to baseline (9.1 vs. 16.4; p=0.038). There was no statistically significant difference in SUVmean values of the blood pool between PET-CT scans at baseline, nephritis and post-nephritis. The SUVmax values of the renal parenchyma and renal pelvis were divided by the blood pool SUVmean to obtain the renal parenchyma SUVmax-to-blood pool SUVmean ratio and the renal pelvis SUVmax-to-blood pool SUVmean ratio. The renal parenchyma SUVmax-to-blood pool SUVmean ratio was significantly higher at nephritis compared to baseline PET-CT (2.13 vs. 1.68; p=0.035). The renal pelvis SUVmax-to-blood pool SUVmean ratio was significantly lower at nephritis compered to baseline scan (3.47 vs. 8.22; p=0.011). The renal pelvis SUVmax-to-blood pool SUVmean ratio was lower on post-nephritis scan compared to baseline but it did not reach statistical significance (4.78 vs. 8.22; p=0.051) (table 1).
The patients were dichotomized according to the increase in renal parenchyma SUVmax-to-blood pool SUVmean ratio and the decrease in renal pelvis SUVmax-to-blood pool SUVmean ratio between baseline scans and PET-CT at nephritis. The median change in renal pelvis SUVmax-to-blood pool ratio SUVmean ratio was 40% and this value was selected as the cutoff value to dichotomize the PET-CT parameters (table 3). The duration between nephritis and death was significantly longer in patients with >40% decrease in renal pelvis SUVmax-to-blood pool SUVmean ratio (1281 days vs. 211.3 days; p=0.001). No statistically significant difference was present among all other clinical and demographic parameters (table 3).
Table 3:
Patient characteristics according to change in kidney parenchyma SUVmax-to-blood pool SUVmean ratio and renal pelvis SUVmax-to-blood pool ratio SUVmean at nephritis compared to baseline.
| <40% change in kidney parenchyma SUVmax-to-blood pool SUVmean ratio (n=10) | >40% increase in kidney parenchyma SUVmax-to-blood pool SUVmean ratio (n=4) | p-value | <40% change in renal pelvis SUVmax-to-blood pool ratio SUVmean ratio (n=7) | >40% decrease in renal pelvis SUVmax-to-blood pool ratio SUVmean ratio (n=7) | p-value | |
|---|---|---|---|---|---|---|
| Mean age ± standard deviation, (year) | 58.8 ± 12.8 | 56.3 ± 13.8 | 0.748 | 59.0 ± 11.6 | 57.1 ± 14.5 | 0.795 |
| Gender | 1.000 | 0.592 | ||||
| Male | 4 (40.0) | 2 (50.0) | 2 (28.6) | 4 (57.1) | ||
| Female | 6 (60.0) | 2 (50.0) | 5 (71.4) | 3 (42.9) | ||
|
| ||||||
| Nephrologic risk factors | ||||||
| Hypertension | 5 (50.0) | 3 (75.0) | 0.580 | 2 (28.6) | 6 (85.7) | 0.103 |
| Hyperlipidemia | 2 (20.0) | 2 (50.0) | 0.520 | 2 (28.6) | 2 (28.6) | 1.000 |
| Diabetes Mellitus | 1 (10.0) | 1 (25.0) | 0.505 | 1 (14.3) | 1 (14.3) | 1.000 |
| Coronary artery disease | 0 (0.0) | 1 (25.0) | 0.286 | 0 (0.0) | 1 (14.3) | 1.000 |
| NSAIDs | 1 (10.0) | 0 (0.0) | 1.000 | 1 (14.3) | 0 (0.0) | 1.000 |
| Proton pump inhibitors | 2 (20.0) | 0 (0.0) | 1.000 | 0 (0.0) | 2 (28.6) | 0.462 |
| Concurrent non-CPI chemotherapy | 5 (50.0) | 1 (25.0) | 0.580 | 3 (42.9) | 3 (42.9) | 1.000 |
|
| ||||||
| CPI immunotherapy | 0.286 | 1.000 | ||||
| Monotherapy | 10 (100.0) | 3 (75.0) | 6 (85.7) | 7 (100.0) | ||
| Multidrug regimen | 0 (0.0) | 1 (25.0) | 1 (14.3) | 0 (0.0) | ||
|
| ||||||
| Specific CPI drug | ||||||
| Pembrolizumab | 6 (60.0) | 1 (25.0) | 0.559 | 4 (57.1) | 3 (42.9) | 1.000 |
| Nivolumab | 4 (40.0) | 2 (50.0) | 1.000 | 3 (42.9) | 3 (42.9) | 1.000 |
| Ipilimumab | 0 (0.0) | 1 (25.0) | 0.286 | 1 (14.3) | 0 (0.0) | 1.000 |
| Atezolizumab | 0 (0.0) | 0 (0.0) | - | 0 (0.0) | 0 (0.0) | - |
| Durvalumab | 0 (0.0) | 1 (25.0) | 0.286 | 0 (0.0) | 1 (14.3) | 1.000 |
|
| ||||||
| Number of CPI cycles | 7.0 ± 4.5 | 6 ± 2.4 | 0.685 | 6.0 ± 1.7 | 7.4 ± 5.4 | 0.519 |
| Duration between CPI initiation to nephritis (days) | 129.3 ± 78.9 | 153.5 ± 85.6 | 0.621 | 158.3 ± 87.9 | 114.1 ± 66.5 | 0.310 |
|
| ||||||
| Kidney function | ||||||
| Baseline creatinine, mg/dL | 0.91 ± 0.13 | 1.26 ± 0.46 | 0.219 | 0.95 ± 0.14 | 1.06 ± 0.41 | 0.536 |
| Peak creatinine, mg/dL* | 3.19 ± 1.36 | 3.38 ± 1.93 | 0.831 | 2.62 ± 1.29 | 3.86 ± 1.45 | 0.116 |
| Final creatinine, mg/dL** | 2.01 ± 1.60 | 1.41 ± 0.49 | 0.540 | 1.51 ± 0.47 | 2.30 ± 2.04 | 0.337 |
| Baseline-to-peak creatinine difference, mg/dL | 2.28 ± 1.36 | 2.12 ± 1.50 | 0.850 | 1.66 ± 1.28 | 2.80 ± 1.24 | 0.115 |
| Baseline-to-peak creatinine change, % | 256.4 ± 154.9 | 155.7 ± 62.1 | 0.108 | 175.6 ± 128.6 | 279.7 ± 141.1 | 0.175 |
| Baseline-to-final creatinine difference, mg/dL | 1.11 ± 1.60 | 0.33 ± 0.26 | 0.431 | 0.55 ± 0.46 | 1.37 ± 2.05 | 0.337 |
| Baseline-to-final creatinine change, % | 124.7 ± 177.6 | 31.8 ± 23.2 | 0.400 | 59.6 ± 49.4 | 154.2 ± 228.7 | 0.306 |
| Dialysis | 0 (0.0) | 1 (25.0) | 0.286 | 0 (0.0) | 1 (14.3) | 1.000 |
| Other immunotherapy related adverse effects | 6 (60.0) | 1 (25.0) | 0.559 | 3 (42.9) | 4 (57.1) | 1.000 |
|
| ||||||
| Toxicity grade | 0.442 | 0.061 | ||||
| Grade 1 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | ||
| Grade 2 | 4 (40.0) | 0 (0.0) | 5 (71.4) | 2 (28.6) | ||
| Grade 3 | 4 (40.0) | 3 (75.0) | 2 (28.6) | 2 (28.6) | ||
| Grade 4 | 2 (20.0) | 1 (25.0) | 0 (0.0) | 3 (42.9) | ||
|
| ||||||
| Treatment | 0.115 | 0.091 | ||||
| CPI cessation only | 2 (20.0) | 0 (0.0) | 2 (28.6) | 0 (0.0) | ||
| CPI cessation and steroid only | 8 (80.0) | 3 (75.0) | 5 (71.4) | 6 (85.7) | ||
| CPI cessation, steroid and infliximab | 0 (0.0) | 1 (25.0) | 0 (0.0) | 1 (14.3) | ||
|
| ||||||
| Renal function after nephritis | 0.593 | 0.935 | ||||
| Creatinine within 0.3 mg/dL of baseline value | 2 (20.0) | 1 (25.0) | 1 (14.3) | 1 (14.3) | ||
| Creatinine >0.3 mg/dL of baseline value | 8 (80.0) | 2 (50.0) | 6 (85.7) | 5 (71.4) | ||
| Dialysis | 0 (0.0) | 1 (25.0) | 0 (0.0) | 1 (14.3) | ||
|
| ||||||
| Biopsy | ||||||
| Acute tubular necrosis (ATN) | 8 (80.0) | 2 (50.0) | 0.520 | 4 (57.1) | 6 (85.7) | 0.559 |
| Acute interstitial nephritis (AIN) | 7 (70.0) | 4 (100.0) | 0.505 | 5 (71.4) | 6 (85.7) | 1.000 |
| Both AIN and ATN | 6 (60.0) | 2 (50.0) | 1.000 | 3 (42.9) | 5 (71.4) | 0.592 |
| Interstitial fibrosis and tubular atrophy (IFTA, %) | 20.0 ± 22.6 | 11.3 ± 13.2 | 0.492 | 22.1 ± 24.8 | 11.7 ± 12.5 | 0.371 |
|
| ||||||
| Cancer response | 0.212 | 0.295 | ||||
| Disease progression | 8 (80.0) | 2 (50.0) | 6 (85.7) | 4 (57.1) | ||
| Partial response | 1 (10.0) | 0 (0.0) | 0 (0.0) | 1 (14.3) | ||
| Remission | 1 (10.0) | 2 (50.0) | 1 (14.3) | 2 (28.6) | ||
|
| ||||||
| Prognosis | ||||||
| Death | 5 (50.0) | 0 (0.0) | 0.221 | 4 (57.1) | 1 (14.3) | 0.266 |
| Duration between nephritis and death (days) | 425.2 ± 216.2 | none | N/A | 211.3 ± 79.7 | 1281 | 0.001 |
Including patients who eventually progressed to dialysis
Excluding patients who eventually progressed to dialysis
Discussion
In this study, 44.1% of cases with biopsy proven immune checkpoint inhibitor-related nephritis demonstrated >30% increase in total kidney volume and new/increasing perinephric fat stranding occurred in 29.4% of cases. One of 8 patients with contrast-enhanced CT at nephritis developed bilateral hypoenhancing cortical foci which on follow up healed with focal scarring. An increase in total kidney volume by >30% had significantly higher toxicity grade and they were managed more aggressively. Although 26.7% of these patients required dialysis (as opposed to 5.3%) there was no statistically significant difference in post-nephritis serum creatinine level and rates of dialysis (p=0.659 and p=0.146, respectively). The authors speculate that the difference in clinical management between these two groups may have resulted in the better alleviation of kidney dysfunction after nephritis in patients with >30% increase in total kidney volume causing the absence of statistically significant difference in long-term kidney function.
On PET-CT, there was a statistically significant decrease in FDG activity in the urine within the renal pelvis was noted at nephritis and post-nephritis scans compared to baseline. The renal parenchyma SUVmax was also higher at nephritis compared to baseline. Due to the small number of patients with available PET-CT imaging meaningful association with clinical parameters could not be obtained.
To our knowledge only 2 prior case reports described imaging findings of immune check point inhibitor-related nephritis but neither was verified on biopsy. Forde et al. reported a case of ipilimumab-related acute renal failure in a melanoma patient who required admission to the intensive care unit, hemodialysis and high-dose corticosteroid treatment. Non-contrast CT showed symmetric bilateral renal enlargement which resolved after therapy[15]. Qualls et al. reported a case of immune checkpoint inhibitor-related nephritis on PET-CT in a patient with vulvar melanoma receiving a combination of ipilimumab and nivolumab for two cycles followed by 7 cycles of nivolumab. PET-CT studies obtained before, during and after acute kidney injury showed bilateral multifocal patchy heterogeneous radiotracer uptake in the renal parenchyma with an SUVmax of 4 about 10 days before the diagnosis was made. FDG uptake returned to baseline after treatment[16].
Work-up of acute kidney injury in patients receiving immunotherapy generally consists of imaging to exclude obstructive uropathy and increasing the intravascular volume if pre-renal etiology is considered. After exclusion of pre- and post-renal entities, common causes of renal dysfunction, such as infection and drug toxicity, are considered. However, making accurate radiologic diagnosis of renal dysfunction remains challenging and imaging findings are commonly non-existent or may overlap with different entities.
The differential diagnosis of bilateral nephromegaly and bilateral ill-defined hypoenhancing cortical foci is broad including obstructive uropathy, infections, infiltrative diseases, vascular and autoimmune entities. Although current imaging modalities are unable to pinpoint the diagnosis of immune checkpoint inhibitor-related nephritis, many entities may be excluded based on their radiologic imaging appearances.
Obstructive uropathy related to urinary bladder outlet obstruction or bilateral ureteral obstruction can cause bilateral nephromegaly, perinephric stranding with striated nephrogram sign on delayed phase post-contrast CT images. However, this entity can be easily identified on ultrasound or other cross-sectional imaging modalities by the presence of dilated collecting system.
Pyelonephritis is usually unilateral and may show hypoenhancing wedge-shaped cortical foci or linear bands (also known as striated nephrogram), but in atypical cases it can have a bilateral presentation[17–19]. Unilateral involvement, presence of white blood cells on urinalysis, fever and other signs of sepsis are suggestive of this diagnosis of pyelonephritis. Malakoplakia, HIV-nephropathy and mycotic infections are rare causes of bilateral nephromegaly[20–24].
Renal vein thrombosis is usually unilateral but when bilateral or in cases of inferior vena cava thrombosis congestion can cause engorgement of the renal parenchyma, perinephric stranding, delayed nephrographic progression and slow opacification of the urinary tract[25]. Renal vein thrombosis can be easily identified on contrast enhanced CT or MRI. Other signs of renal vein thrombosis on non-contrast enhanced CT and MRI include dilatation of the renal veins and/or inferior vena cava, loss of normal signal void within the renal veins on T2-weighted images and diffusion restriction. Nevertheless, it should be noted that renal vein thrombosis can occur secondary to glomerulonephritis (especially the membranoproliferative subtype) and nephrotic syndrome[25].
Extramedullary hematopoiesis and malignant infiltration secondary to lymphoma, leukemia or multiple myeloma can be indistinguishable from the imaging patterns identified in the current article[26–32]. Amyloidosis can also cause bilateral renal enlargement with hypoenhancing parenchymal foci. Parenchymal calcifications occurring in 25% of the cases and high density of the renal parenchyma seen in 60% of renal amyloidosis cases can be suggestive of this condition[33].
Sarcoidosis and IgG4-related disease can both present with bilateral renal enlargement and hypoenahancing parenchymal lesions, but concurrent extra-renal manifestations are more common than isolated kidney involvement[34–36]. When these conditions are isolated to the kidney, they may be radiologically indistinguishable from the patterns described for immune checkpoint inhibitor-related nephritis. Glomerulonephritis is a general term that encompasses many histopathologic subtypes and there is a paucity of studies describing imaging features of this entity. No imaging abnormality is identifiable in most cases with glomerulonephritis, but bilateral renal enlargement with preservation of the smooth renal contours and hypoenhancing wedge-shaped foci in the renal cortex may occur[25].
The limitations of the current study include the small study cohort, retrospective design, variety of malignant disorders and differences in immune checkpoint inhibitor regimens. Verification of the findings reported in larger studies is necessary. Cystain C-based estimated glomerular filtration rate (eGFR) gives a better estimate of kidney function, particularly in the setting of muscle wasting occurring in cancer patients. However, serum creatinine is used instead of cystatin C as the standard of care at the author’s institution. Future studies on this topic should use cystatin C-based eGFR to better reflect the extent of kidney dysfunction. Iodinated contrast material can worsen pre-existing renal dysfunction and it is contraindicated in most cases of acute kidney injury. Institutional policies at our center allow administration of iodinated contrast media in patients with mild increase in creatinine. Hence, the 8 patients with available contrast-enhanced CT at nephritis had less severe renal toxicity, which constitutes is a potential source of bias for the findings seen on contrast-enhanced CT. Although almost all patients included in the current study had ultrasound images at the time of nephritis to exclude post-renal causes of acute kidney injury, most did not have prior baseline scans for comparison. Therefore, it was impossible to differentiate whether abnormalities seen on ultrasound were developed secondary to nephritis or were pre-existent findings. In addition, comparing kidney volumes measured on ultrasound with volumes measured on cross-sectional imaging could have introduced additional measurement bias. Hence, the authors chose to compare volumes measured using the same modality (CT). A single patient had abdominal MRI images at the time of nephritis which showed bilateral increase in kidney size. Hence, the imaging characteristics of immunotherapy-related nephritis on MRI remain unknown and future research may consider investigating their radiologic features.
In conclusion, imaging features of immune checkpoint inhibitor-related nephritis include bilateral increase in kidney volume, new/increasing perinephric fat stranding and development of hypoenhancing wedge-shaped cortical foci. A >30% increase in total kidney volume is associated with more severe toxicity grade and more aggressive clinical management. On PET-CT, diffuse increase in FDG uptake throughout the renal parenchyma and decrease in radiotracer activity in the renal pelvis can occur. Novel use of imaging as a supportive diagnostic tool may provide additional guidance for management of immune checkpoint inhibitor-related nephritis and may help improve current guidelines to better optimize patient care and survival[7].
Key points.
CT features of immune checkpoint inhibitor-related nephritis include increase in kidney volume, new/increasing perinephric stranding and bilateral ill-defined wedge-shaped hypoenhancing cortical foci.
FDG-PET features of immune checkpoint inhibitor-related nephritis include increase in FDG uptake throughout the renal cortex and decrease in FDG activity/excretion in the collecting system.
>30% increase in total kidney volume is associated with worse toxicity grade and more aggressive medical management.
Acknowledgements:
The authors would like to thank Erica Goodoff, Senior Scientific Editor in the Research Medical Library at The University of Texas MD Anderson Cancer Center, for editing this article. National Institute of Health K01 grant (N.A.: K01AI163412)
Funding:
The authors state that this work has not received any funding.
Abbreviations
- CT
Computerized Tomography
- eGFR
estimated Glomerular Filtration Rate
- FDG
18F-flourodeoxyglucose
- IFTA
Interstitial fibrosis and tubular atrophy
- irAE
Immune related adverse events
- IV
Intravenous
- MRI
Magnetic Resonance Imaging
- PET-CT
Positron Emission Tomography-Computerized Tomography
- ROI
Region of interest
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Guarantor:
The scientific guarantor of this publication is Prof. Dr. Khaled Elsayes.
Conflicts of interest:
N.A. has received honoraria for serving on a scientific advisory board as a consultant for ChemoCentryx. A.D. has received honoraria from Nektar and he has served as a consultant for Nektar, Memgen and Pfizer.
Statistics and Biometry:
One of the authors has significant statistical expertise.
Informed Consent:
Written informed consent was waived on this retrospective study.
Ethics approval:
Institutional Review Board approval was obtained.
Methodology:
• Retrospective
• Observational study
• Single institution
Contributor Information
Muhammad O. Awiwi, Division of Diagnostic Imaging, Department of Abdominal Imaging, The University of Texas MD Anderson Cancer Center, Houston, 1515 Holcombe Blvd, Houston, 77030, Texas, USA.
Ala Abudayyeh, Division of Internal Medicine, Section of Nephrology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Noha Abdel-Wahab, Division of Internal Medicine, Section of Rheumatology and Clinical Immunology, Department of General Internal Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA; Department of Rheumatology and Rehabilitation, Faculty of Medicine, Assiut University Hospitals, Assiut, Egypt.
Adi Diab, Department of Melanoma Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Migena Gjoni, Department of Internal Medicine, Istanbul University-Cerrahpasa, Istanbul, Turkey.
Guofan Xu, Division of Diagnostic Imaging, Department of Nuclear Medicine and Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Raghu Vikram, Division of Diagnostic Imaging, Department of Abdominal Imaging, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Khaled Elsayes, Division of Diagnostic Imaging, Department of Abdominal Imaging, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
References
- 1.Ramos-Casals M, Brahmer JR, Callahan MK et al. (2020) Immune-related adverse events of checkpoint inhibitors. Nat Rev Dis Primers 6:38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cortazar FB, Marrone KA, Troxell ML et al. (2016) Clinicopathological features of acute kidney injury associated with immune checkpoint inhibitors. Kidney Int 90:638–647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Abdelrahim M, Mamlouk O, Lin H et al. (2021) Incidence, predictors, and survival impact of acute kidney injury in patients with melanoma treated with immune checkpoint inhibitors: a 10-year single-institution analysis. Oncoimmunology 10:1927313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mamlouk O, Selamet U, Machado S et al. (2019) Nephrotoxicity of immune checkpoint inhibitors beyond tubulointerstitial nephritis: single-center experience. J Immunother Cancer 7:2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Puzanov I, Diab A, Abdallah K et al. (2017) Managing toxicities associated with immune checkpoint inhibitors: consensus recommendations from the Society for Immunotherapy of Cancer (SITC) Toxicity Management Working Group. J Immunother Cancer 5:95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Perazella MA, Shirali AC (2020) Immune checkpoint inhibitor nephrotoxicity: what do we know and what should we do? Kidney Int 97:62–74 [DOI] [PubMed] [Google Scholar]
- 7.Ala Abudayyeh AD, Wendy Garcia, Jamie S. Lin, et al. (2021) Evaluation and Management of Suspected Immune-Mediated Nephritis Available via https://www.mdanderson.org/content/dam/mdanderson/documents/for-physicians/algorithms/clinical-management/clin-management-nephritis-web-algorithm.pdf. Accessed 01/24/2021 2022
- 8.Anderson MA, Kurra V, Bradley W, Kilcoyne A, Mojtahed A, Lee SI (2021) Abdominal immune-related adverse events: detection on ultrasonography, CT, MRI and 18F-Fluorodeoxyglucose positron emission tomography. Br J Radiol 94:20200663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gandy N, Arshad MA, Wallitt KL, Dubash S, Khan S, Barwick TD (2020) Immunotherapy-related adverse effects on (18)F-FDG PET/CT imaging. Br J Radiol 93:20190832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bronstein Y, Ng CS, Hwu P, Hwu WJ (2011) Radiologic manifestations of immune-related adverse events in patients with metastatic melanoma undergoing anti-CTLA-4 antibody therapy. AJR Am J Roentgenol 197:W992–W1000 [DOI] [PubMed] [Google Scholar]
- 11.Widmann G, Nguyen VA, Plaickner J, Jaschke W (2016) Imaging Features of Toxicities by Immune Checkpoint Inhibitors in Cancer Therapy. Curr Radiol Rep 5:59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kwak JJ, Tirumani SH, Van den Abbeele AD, Koo PJ, Jacene HA (2015) Cancer immunotherapy: imaging assessment of novel treatment response patterns and immune-related adverse events. RadioGraphics 35:424–437 [DOI] [PubMed] [Google Scholar]
- 13.Kim HC, Yang DM, Jin W, Lee SH (2010) Relation between total renal volume and renal function: Usefulness of 3D sonographic measurements with a matrix array transducer. AJR Am J Roentgenol 194:W186–192 [DOI] [PubMed] [Google Scholar]
- 14.Emamian SA, Nielsen MB, Pedersen JF, Ytte L (1993) Kidney dimensions at sonography: correlation with age, sex, and habitus in 665 adult volunteers. AJR Am J Roentgenol 160:83–86 [DOI] [PubMed] [Google Scholar]
- 15.Forde PM, Rock K, Wilson G, O’Byrne KJ (2012) Ipilimumab-induced immune-related renal failure--a case report. Anticancer Res 32:4607–4608 [PubMed] [Google Scholar]
- 16.Qualls D, Seethapathy H, Bates H et al. (2019) Positron emission tomography as an adjuvant diagnostic test in the evaluation of checkpoint inhibitor-associated acute interstitial nephritis. J Immunother Cancer 7:356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vernuccio F, Patti D, Cannella R, Salvaggio G, Midiri M (2020) CT imaging of acute and chronic pyelonephritis: a practical guide for emergency radiologists. Emerg Radiol 27:561–567 [DOI] [PubMed] [Google Scholar]
- 18.Ishikawa I, Saito Y, Onouchi Z et al. (1985) Delayed contrast enhancement in acute focal bacterial nephritis: CT features. J Comput Assist Tomogr 9:894–897 [DOI] [PubMed] [Google Scholar]
- 19.Maturen KE, Blane CE, Strouse PJ (2002) Computed tomographic diagnosis of unsuspected pyelonephritis in children. Can Assoc Radiol J 53:279–283 [PubMed] [Google Scholar]
- 20.Kuhlman JE, Browne D, Shermak M, Hamper UM, Zerhouni EA, Fishman EK (1991) Retroperitoneal and pelvic CT of patients with AIDS: primary and secondary involvement of the genitourinary tract. RadioGraphics 11:473–483 [DOI] [PubMed] [Google Scholar]
- 21.Craig WD, Wagner BJ, Travis MD (2008) Pyelonephritis: radiologic-pathologic review. RadioGraphics 28:255–277; quiz 327–258 [DOI] [PubMed] [Google Scholar]
- 22.Hartman DS, Davis CJ Jr, Lichtenstein JE, Goldman SM (1980) Renal parenchymal malacoplakia. Radiology 136:33–42 [DOI] [PubMed] [Google Scholar]
- 23.Zimina OG, Rezun S, Armao D, Braga L, Semelka RC (2002) Renal malacoplakia: demonstration by MR imaging. Magn Reson Imaging 20:611–614 [DOI] [PubMed] [Google Scholar]
- 24.Ranjan P, Chipde SS, Vashistha S, Kumari N, Kapoor R (2014) Reno-invasive fungal infection presenting as acute renal failure: importance of renal biopsy for early diagnosis. Saudi J Kidney Dis Transpl 25:1282–1284 [DOI] [PubMed] [Google Scholar]
- 25.Ronald Zagoria CB, Raymond Dyer (2015) The Kidney: Diffuse Parenchymal Abnormalities. In: Dyer RZCBR, (ed) Genitourinary Imaging: The Requisites Elsevier, Philadelphia, PA, pp 107–145 [Google Scholar]
- 26.Sherief LM, Azab SF, Zakaria MM et al. (2015) Renal Presentation in Pediatric Acute Leukemia: Report of 2 Cases. Medicine (Baltimore) 94:e1461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hilmes MA, Dillman JR, Mody RJ, Strouse PJ (2008) Pediatric renal leukemia: spectrum of CT imaging findings. Pediatr Radiol 38:424–430 [DOI] [PubMed] [Google Scholar]
- 28.Arora SK, Swarnim S, Hemal A, Bidhuri N (2019) Acute lymphoblastic leukemia presenting as nephromegaly in a child: A rare case report. Turk J Pediatr 61:97–101 [DOI] [PubMed] [Google Scholar]
- 29.Gibbins J, Pankhurst T, Murray J et al. (2005) Extramedullary haematopoiesis in the kidney: a case report and review of literature. Clin Lab Haematol 27:391–394 [DOI] [PubMed] [Google Scholar]
- 30.Ganeshan D, Iyer R, Devine C, Bhosale P, Paulson E (2013) Imaging of primary and secondary renal lymphoma. AJR Am J Roentgenol 201:W712–719 [DOI] [PubMed] [Google Scholar]
- 31.Nalcacioglu H, Yel S, Karadogan M, Mutlu FT, Bastug F, Altuner Torun Y (2018) Abdominal pain in a 5-year-old girl with bilateral nephromegaly: Answers. Pediatr Nephrol 33:423–426 [DOI] [PubMed] [Google Scholar]
- 32.Lei W, Wang H, Di W, Feng M, Li L, Shiren S (2020) Bilateral Primary Renal Lymphoma Presented As Homogenous Renal Enlargement And Acute Interstitial Nephritis. Urol J 17:317–320 [DOI] [PubMed] [Google Scholar]
- 33.Apter S, Zemer D, Terhakopian A et al. (2001) Abdominal CT findings in nephropathic amyloidosis of familial Mediterranean fever. Amyloid 8:58–64 [DOI] [PubMed] [Google Scholar]
- 34.Tsukada H, Shimizu H, Kaname S (2017) Bilateral nephromegaly due to isolated renal sarcoidosis. Clin Exp Nephrol 21:169–170 [DOI] [PubMed] [Google Scholar]
- 35.Ling J, Wang H, Pan W et al. (2020) Clinical and imaging features of IgG4-related kidney disease. Abdom Radiol (NY) 45:1915–1921 [DOI] [PubMed] [Google Scholar]
- 36.Kim B, Kim JH, Byun JH et al. (2014) IgG4-related kidney disease: MRI findings with emphasis on the usefulness of diffusion-weighted imaging. Eur J Radiol 83:1057–1062 [DOI] [PubMed] [Google Scholar]