Diagnosis and Management of Immune Checkpoint Inhibitor‐Associated Renal Toxicity: Illustrative Case and Review

Oncologists treating immune checkpoint inhibitor‐exposed patients often face challenging questions during the evaluation of renal immune‐related adverse events. This review addresses frequently asked questions to assist in the recognition and management of these conditions

Abstract

Immune checkpoint inhibitors (ICIs) are monoclonal antibodies directed at negative regulatory components on T cells, such as cytotoxic T lymphocyte‐associated antigen 4, programmed cell death‐1 (PD‐1), and its ligand, programmed cell death ligand‐1. ICIs initate antitumor immunity; however, these agents are associated with immune‐related adverse events (irAEs) that may affect a variety of organs. Renal irAEs most commonly present with asymptomatic acute kidney injury (AKI), which is often detected by routine laboratory testing. The severity of AKI associated with irAEs ranges from mild (grade 1–2) to severe (grade 3–4). It is often challenging to diagnose because this group of patients often have multiple reasons to have AKI (dehydration, sepsis, or nephrotoxic medication exposure). We present an illustrative case of a 60‐year‐old man with metastatic melanoma who presented with AKI during treatment with nivolumab and review the literature to address frequently asked questions concerning the diagnosis and management of renal irAEs in patients with advanced cancer. Importantly, most patients will recover completely, and some may tolerate a rechallenge of ICI therapy, with prompt and effective treatment.

Key Points.

• Renal immune‐related adverse events (irAEs) are less frequently reported than other irAEs; however, it is possible that available data underestimate their true incidence because of missed diagnoses and under‐reporting. Although severe renal irAEs are more easily detected, smaller rises in creatinine may not be appreciated or may be attributed to other causes, because the differential diagnosis of acute kidney injury (AKI) in patients with cancer is broad.

• Baseline creatinine should be established prior to beginning immune checkpoint inhibitorss (ICIs), and it should be monitored with every cycle. If a patient develops AKI, the ICI should be held while the evaluation is pursued. A thorough workup of suspected renal irAEs must exclude other potential causes of AKI such as infection, dehydration, urinary tract obstruction, and nephrotoxin exposure.

• Acute kidney injury after ICI therapy does not appear to be more common in patients with baseline estimated glomerular filtration rate <60 mL per min per 1.73 m. One particular concern, however, is that those with baseline renal disease have less “renal reserve,” and repeated AKI events may push a patient closer to end‐stage renal disease. Thus, clinicians must exert caution when rechallenging patients with pre‐existing renal disease with ICI therapy in the event of a prior AKI from ICI‐related allergic interstitial nephritis.

Introduction

The first immune checkpoint inhibitor (ICI), ipilimumab, was approved by the U.S. Food and Drug Administration (FDA) in 2011 for treatment of unresectable or metastatic melanoma. Since then, six additional agents have been approved, and use has expanded to multiple types of cancers, with improvement in overall survival [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. ICIs act by blocking immune inhibitor‐signaled cytotoxic T lymphocyte‐associated antigen 4 (CTLA‐4), programmed cell death‐1 (PD‐1), and its ligand 1 (PD‐L1), initiating antitumor immune responses. However, by the same mechanisms they may also lead to immune‐related adverse events (irAE) [11]. The overall incidence of irAE affecting any organ is as high as 86% with anti‐CTLA‐4 agents and 82% with PD‐1 and PD‐L1 agents. The incidence of renal toxicity with these agents is <5%, with grade 3–4 acute kidney injury (AKI) having an incidence of <1% [2]. In a recent analysis of 11 clinical trials (comprising 3,695 patients), the overall incidence of renal irAEs, graded by changes in creatinine, was 2.2%. The incidence differs by checkpoint inhibitor class, with combination therapy carrying a higher risk (4.9%) than monotherapy with anti‐CTLA‐4 (2%) or anti‐PD‐1 (1.4%–1.9%) [12]. The spectrum of renal irAEs reported in the literature is broad; however, the most common lesion seen in histopathologic series is allergic interstitial nephritis (AIN) [12], [13], [14], [15].

Although renal irAEs occur less frequently than irAEs [16], [17] that affect other organ systems such as the skin, gastrointestinal tract, and endocrinopathies, we suspect that available data underestimate their true incidence because missed diagnoses and under‐reporting. Although severe (grade 3–5) renal irAEs are more easily detected, smaller rises in creatinine may not be appreciated or may be attributed to other causes (chemotherapy, antibiotics, nonsteroid anti‐inflammatory drugs, protein pump inhibitors), becuase the differential diagnosis of AKI in patients with cancer is broad. In our medical center, 13.5% of patients receiving ICI had AKI, defined by a rise in creatine at least 1.5 times baseline, within 6 months of starting ICI therapy [18]. Furthermore, substantial renal injury can develop before a significant rise in the baseline serum creatinine, highlighting the importance of early recognition. A doubling of serum creatinine level (indicating significant renal injury) in patients with a low muscle mass and low baseline creatinine (0.4–0.6 mg/dL) is typically documented as being in the “normal range” in electronic charts and may go unnoticed by providers. Associated symptoms of AKI (such as fatigue, nausea, dysgeusia, or generalized weakness) may be attributed to cancer‐related symptoms. Urinalysis findings such as hematuria or pyuria are generally nonspecific in patients with renal irAEs [12], [19], and an unremarkable urinalysis should not reassure the evaluating clinicians. Finally, while a kidney biopsy is the “gold standard” for diagnosing renal irAEs, biopsy may be difficult to obtain in a timely manner because of the need to stop aspirin or other blood thinners at least 7–10 days before a native kidney biopsy.

For these reasons, oncologists treating ICI‐exposed patients are commonly faced with challenging questions during the evaluation of renal irAEs. In this review, we address frequently asked questions to assist in the recognition and management of these conditions.

Case Vignette

A clinical case timeline is shown in Figure 1.

Figure 1.

Case timeline. Serum creatinine values (solid line) for the patient in the case vignette. The first prednisone taper began the evening after kidney biopsy at 60 mg per day and tapered over 6 weeks. The second prednisone course, which was begun empirically because rise in serum creatinine, began at 10 mg per day and slowly tapered by 2.5 mg per day per month. The patient continued on low‐dose prednisone.

Abbreviations: CT, computed tomography; IV, intravenous.

Mr. A, a 60‐year‐old man, was seen in outpatient oncology clinic for treatment of melanoma originating on the left leg. He underwent a wide local excision and sentinel lymph node biopsy, followed by a groin dissection that demonstrated a total of five (microscopically) positive nodes in the left inguinal area. Adjuvant ipilimumab was started, but the fourth cycle was held because of grade 1 liver function tests. These test results subsequently normalized; however, he developed grade 3 diarrhea and abdominal pain. Esophagogastroduodenoscopy and colonoscopy demonstrated pan‐enteritis, most severe in the duodenum. He was treated with prednisone 60 mg daily but experienced a flare of symptoms while prednisone was being tapered. After two doses of infliximab, his symptoms resolved and he was tapered off steroids; briefly, he showed no evidence of disease.

Six months later, Mr. A had a local recurrence of melanoma in his left inguinal scar with epidermotrophic metastases to his left cheek. He began nivolumab 120 mg intravenous (IV) every two weeks. He tolerated his first two cycles of nivolumab well. However, on presentation for his third cycle, his creatinine was 3.2 mg/dl up from his baseline of 1.0 mg/dL, and he had developed multiple new metastatic pigmented lesions on his left leg. He complained of nausea, abdominal pain, and mild chills but was tolerating normal oral intake and denied fever, diarrhea, or vomiting. His blood pressure (85/40 mmHg) was 20 mmHg lower than his recent baseline.

In addition to his melanoma, he had thyroid cancer (status post‐thyroidectomy and radioactive iodine), a noninvasive bladder cancer, prostate cancer (treated with prostatectomy), hypertension, hypercholesterolemia, erectile dysfunction, and Barrett's esophagus. His medications included omeprazole, lisinopril, acetaminophen, docusate sodium, levothyroxine, hydrochlorothiazide, lorazepam, oxycodone, papaverine‐phentolamine alprostadil, sildenafil, and simvastatin.

Mr. A was referred for specialty consultation the same day. At the nephrologist's office, his exam was notable for a blood pressure of 103/60 mmHg, pulse rate of 59 beats per minute, and an oxygen saturation of 98% on room air. He was well appearing with moist mucous membranes, anicteric sclera, and clear conjunctiva. His breathing was unlabored, and his jugular venous pressure was not elevated above his clavicle. His heart rate was slightly bradycardic but with a regular rhythm and without murmurs, rubs, or gallops. His lung sounds were clear, his abdomen was soft and nondistended (without any tenderness or organomegaly), he had no edema, and his capillary refill and palpable peripheral pulses were normal. His neurologic exam demonstrated that he was fully oriented and had excellent attention, normal cranial nerve function, and normal strength and sensation in his extremities.

In clinic, he received 3 liters of normal saline, and his repeat creatinine was 2.4 mg/dL. His laboratory testing was notable for blood urea nitrogen of 45 mg/dL. Urinalysis revealed <5 leukocytes and <5 erythrocytes per high powered field, and a microscopic urine sediment had few granular casts. A summary of his laboratory test results is shown in Table 1. A computed tomography scan of the abdomen, obtained without IV contrast enhancement, showed enlarged kidneys (∼12.5 cm) without any evidence of urinary tract obstruction. He was admitted to the hospital for an expedited kidney biopsy. Same‐day preliminary results on his kidney biopsy demonstrated severe AIN (Fig. 2).

Table 1. Laboratory values.

Abbreviations: BUN, blood urea nitrogen; hpf, high power field; RBC, red blood cell; WBC, white blood cell.

Figure 2.

Kidney histopathology for immune checkpoint inhibitors‐allergic interstitial nephritis. Periodic‐acid Schiff stain 400× magnification demonstrating dense interstitial inflammatory infiltrate, tubulitis, and tubular basement membrane rupture (arrows).

That evening, corticosteroids were initiated for ICI‐induced AIN. He received prednisone (starting at 60 mg daily) and was discharged the day after kidney biopsy with a down‐trending creatinine (Fig. 1). His proton pump inhibitor (PPI) therapy was switched to a histamine H²‐receptor antagonist (H² antagonist) given the association of PPIs with AIN. One week later, his serum creatinine was 1.2 mg/dL. His prednisone was tapered by 10 mg every week, and nivolumab was held. He received Talimogene laherparepvec injections into his lesions while he underwent a prednisone taper. When his prednisone dose decreased to 5 mg per day he was rechallenged with nivolumab, approximately 6 weeks after his kidney biopsy. After three more cycles of nivolumab, his creatinine rose to 1.7 mg/dL (Fig. 1). During this time, he also received iodinated contrast and his PPI was restarted. He was hydrated, and prednisone was restarted empirically at 10 mg per day; his PPI was switched back to a H² antagonist. His repeat creatinine returned to his baseline. He remains on nivolumab therapy and low‐dose prednisone with a stable serum creatinine and without evidence of melanoma recurrence.

Diagnosis and Management of Renal irAEs

In General, How Soon After Initiation of ICIs Do Renal irAEs Develop?

A renal irAE should be suspected in any patient who develops new AKI after ICI exposure. Signs and symptoms typically present within 3–6 months of starting therapy, with a median symptom onset of 3 months; however, a range of 3 weeks to 9 months has been reported [12]. Renal irAEs can develop after a single dose of an ICI and up to several months after the end of treatment [12], [19].

Which Renal irAEs Have Been Reported and How Do They Present?

Although the spectrum of reported renal irAEs in patients receiving ICIs is broad, the vast majority of patients have AIN on biopsy; this lesion is characterized by a predominance of CD3+ T lymphocytes in the mixed cellular infiltrate that infiltrates the interstitium of the kidney and is often associated with inflammation of the tubules (tubulitis) [12]. Case reports have noted that other inflammatory kidney diseases can develop in patients treated with ICIs, including cases of antibody‐triggered kidney diseases such as antinuclear cytoplasmic antibody (ANCA) associated disease, lupus nephritis, and podocytopathies [12], [13], [14], [15]. Allograft rejection has been reported in kidney transplant recipients [20].

In the largest series of patients with ICI‐triggered AIN, the majority (8/13, 61%) had another irAE at the time of diagnosis of AIN [12]. Importantly, two patients developed a renal irAE while on a steroid taper for an extrarenal irAE. The median peak serum creatinine was 4.5 mg/dL (interquartile range, 3.6–7.3); approximately half had pyuria (>5 white blood cells per high power field), and a minority had hematuria (23%). Urine eosinophilia is no longer considered a helpful test in the evaluation of AIN because of its high false positive and false negative rate [21], [22]. Because AIN does not affect the glomerulus, proteinuria is generally modest, with the largest series showing a range of 0.12–0.98 g per day. It is extremely rare to experience other symptoms of the “allergic triad” (rash, fever, eosinophilia) associated with other forms of AIN. Only 10% of patients had eosinophilia, and none had rash or fever. Although enlarged kidneys may be seen in AIN because of the infiltration of leukocytes, this finding lacks the sensitivity and specificity to be helpful diagnostically.

How Is the Diagnosis and Severity of Renal irAEs Determined?

The Common Terminology Criteria for Adverse Events (CTCAE) provides grading criteria for renal irAEs (Table 2). In general, mild AEs are grade 1–2, severe are grade 3–4, and death is grade 5. The CTCAE was created prior to immunotherapy and may not adequately represent the threat that the immunotherapy toxicities pose to the patient owing to their dynamic, potentially rapidly progressive nature.

Table 2. Grading of renal immune‐related adverse events and management by severity.

Modified from CTCAE 5.0.

Of note, in cases of immune checkpoint inhibitor (ICI)‐allergic interstitial nephritis (AIN), creatinine‐based changes will drive immune‐related adverse event severity grading as AIN is typically not associated with heavy proteinuria.

^aRecommended serologies: antinuclear antibody, anti‐double stranded DNA antibody, complement factors C3 and C4, anti‐neutrophil cytoplasmic antibody, Hepatitis B and C titers, HIV titers.

^bRecommendations regarding course of prednisone and taper described in section titled “How do we manage renal complications related to ICIs?”

Abbreviations: CPI, IV, intravenous; SCr, serum creatinine; ULN, upper limit of normal.

What Is the Recommended Workup of irAEs?

Baseline creatinine should be established prior to beginning ICI and it should be monitored with every cycle. If a patient develops AKI, the ICI should be held while the evaluation is pursued. A thorough workup of suspected renal irAEs must exclude other potential causes of AKI such as infection, dehydration, urinary tract obstruction, and nephrotoxin exposure (Table 3). Patients with a persistent grade 2 creatinine elevation or higher, any grade 3 event, or metabolic changes concerning for renal failure should be referred to a nephrologist [23], [24], [25]. Although the case above demonstrates rapid coordination of care that may not be feasible in all cases, expedited evaluation with a nephrologist who is familiar with ICIs and their toxicity is essential.

Table 3. Differential diagnosis of AKI in a patient with cancer.

Abbreviations: ACEi, angiotensin converting enzyme inhibitor; AIN, allergic interstitial nephritis; AKI, acute kidney injury; ARB, angiotensin receptor blockade; ATN, acute tubular necrosis; BUN, blood urea nitrogen; Cr, creatinine; CT, computed tomography; FENa, fractional excretion of sodium; ICI, immune checkpoint inhibitor; IV, intravenous; NSAID, nonsteroidal anti‐inflammatory drug; PPI, proton pump inhibitor.

It is challenging to differentiate between cases of AIN versus acute tubular necrosis (ATN), which may occur commonly in patients with cancer in the setting of sepsis, hypotension, or use of nephrotoxins. Urine sediment findings have repeatedly shown insufficient precision to distinguish ATN from AIN. Granular casts, erythrocytes, and erythrocyte casts are commonly found in the urine of patients with AIN, thus rendering the urine sediment unreliable as a diagnostic marker [26], [27]. This poses a dilemma for patients with advanced malignancies who are receiving nephrotoxic chemotherapies and immunotherapy. Platinum‐based chemotherapies (such as cisplatin and carboplatin) carry a significant independent risk of nephrotoxicity, with incidence rates of around 30%–40% for cisplatin and half that for carboplatin [28], [29]. In the KEYNOTE‐0189 trial, addition of pembrolizumab increased the risk of AKI when it was added to pemetrexed and a platinum‐based drug (5.2% vs. 0.5%) [30]. The results of this study need to be interpreted with caution becuase the event rate in the non‐ICI group was lower than in other studies with platinum‐based drugs [28], [29], [31], [32]. In addition to coadministration of nephrotoxic chemotherapies, the use of other potential nephrotoxins is common in patients with malignancies. Nonsteroidal antiinflammatory drugs (NSAIDs) may be particularly likely to cause AKI when coadministered with angiotensin converting enzyme inhibitor, angiotensin receptor blocker, or a diuretic. Finally, intra‐abdominal burden of malignant diseases may lead to urinary tract obstruction, and men with prostate enlargement may develop sudden onset outlet obstruction either in response to opiod medications or in the setting of a urinary tract infection. Table 3 summarizes recommended labs, imaging, and other diagnostic tests that may help identify other causes of AKI in a patient with cancer.

Current guidelines state explicitly that a kidney biopsy is not necessary before initiation of steroids in cases of suspected ICI‐induced renal irAEs, particularly in patients in whom other causes of AKI can be confidently eliminated. In these cases, a therapeutic trial of steroids is warranted. However, in our opinion, provided it is safe, we recommend strongly considering a kidney biopsy to ensure that renal irAE is the cause of AKI in patients with multiple potential causes of AKI.

It is important to weigh the risks and benefits of a kidney biopsy. A kidney biopsy carries a 1% risk of bleeding transfusion due to bleeding [33], and the risk may be higher in patients with cancer. Additionally, some patients have absolute or relative contraindications to kidney biopsy, such as having a solitary functioning kidney, morbid obesity, inability to hold antiplatelet agents, or blood thinners [34]. For those who use aspirin or NSAIDs regularly, these agents must be held for 7–10 days before a kidney biopsy. Anticoagulation should also be held for 1 week if possible. Bridging anticoagulation should be considered in high‐risk patients [35]. For a patient with a rapidly rising serum creatinine, delay of treatment for even one week may cause lasting damage. For these reasons, if other causes of AKI have been excluded (Table 3), it is appropriate to begin empiric therapy with prednisone (Table 2).

However, in contemplating the risks, we also feel strongly that treating clinicians must weigh the risks of inaccurate diagnosis. Disrupting ICI therapy in a patient with advanced malignancy is not a decision to take lightly. Although early studies suggested that overall survival rates and time‐to‐treatment failure do not appear to be affected by the presence of a irAE or the need for systemic steroids [36], emerging data suggest that decreased survival is noted when high dose glucocorticoids are used (vs. low dose for hypophysitis) [37]. Thus, it is crucial that these maneuvers (holding ICI therapy and starting high dose steroids) only be deployed for patients in whom the treating oncologist and nephrologist agree, confidently, that AIN is present. Increasingly, as coadminstration of ICIs with nephrotoxic chemotherapies becomes more common, we suspect that a kidney biopsy will provide diagnostic clarity in a treatment decision that is so frought with both the potential to derail effective cancer therapy and put patients at risk for adverse side effects.

How Do We Manage Renal Complications Related to ICIs?

Successful management relies on recognition of the condition and management of potential alternative causes (fluid challenge, withholding nephrotoxic medications), followed by close monitoring and withholding of the ICI until the creatinine recovers to grade 1 (<1.5‐fold baseline). To date, no reported large studies have established the therapuetic efficacy of holding the ICI alone. Unless symptoms are very mild (i.e., grade 1), we typically initiate corticosteroids as soon as we are confident of the diagnosis of a renal irAE in anyone with a grade 2 event or higher. In general, we use oral corticosteroids (prednisone) 1 mg/kg per day (up to 60 mg per day) for stage 3–4 disease and would consider IV solumedrol for dialysis‐requiring AKI. Table 2 summarizes the published guidelines for management of renal irAEs [23], [24], [25]. Our experience indicates that most patients treated early improve significantly with steroids alone. Some patients need a steroid‐sparing agent if they are unable to wean off of steroids without recurrence of AKI. Because there are no data to guide selection of this treatment, we consider this on a case‐by‐case basis. Mycophenolate mofetil, an immunomodulator widely used in transplantation and glomerulonephritides, has been used successfully in steroid‐resistant and steroid‐dependent patients with AIN [38]. Some of our patients with renal irAEs have been on infliximab for concomitant refractory colitis; however, this agent is not typically used to treat AIN. Additional studies providing evidence for management of patients refractory to initial steroid treatment are greatly needed.

The duration of steroid treatment and the speed of the steroid taper is not well studied. We typically use prednisone 1 mg/kd per day up to 60 mg per day for the first week. If at that point the serum creatinine is rapidly improving (ideally to within a grade 1), we begin a taper by 10 mg per day each week for a treatment duration of approximately 2 months. We suggest keeping the steroids at a higher dose for another week before starting the taper if renal recovery is slower. In general, we follow creatinine twice per week, until it reaches grade 1, and then weekly while on the steroid taper.

The side effects of prednisone are well known and include mood changes, weight gain, hyperglycemia, hypertension, gastritis, osteoporosis, opportunistic infections, and skin fragility. Therefore, when treating a renal irAE with high doses of steroids, coadministration of calcium and vitamin D supplementation, gastric acid blockade, and pneumocystis jiroveci prophylaxis (PJP) should be considered. Because PPIs and or trimethoprim/sulfamethoxazole have been associated with AIN, our preferred practice is to treat them with H² antagonist and consider atovaquone, acknowledging it is less effective at preventing PJP infection. Coordinating management of steroids with disease‐specific subspecialists is important in patients who have multiple manifestions of irAEs.

A coordinated care model with early access to subspecialists skilled at managing immune‐related toxicities can have a major impact on management and prognosis in these patients, such as seen in our case above. In clinical settings in which such an approach is not feasible, it is reasonable to manage suspected ICI toxicity by holding ICI therapy and empiric steroids in patients without major risk factors for glucorticoid induced adverse effects. Oncologists should consider a referral to a nephrologist in any patient with ≥grade 2 toxicity, in any recurrent nephritis, or in patients with complex presentations in which other etiologies of rising creatinine need to be ruled out. Centers that use electronic consultations should have a system in place for development of a coordinated management plan with the input of the nephrologist and primary oncologist. This may especially be important when a kidney biopsy cannot be expedited or in patients at high risk for complications from a kidney biopsy. In these cases, conservative management and empiric treatment with steroids is supported by current guidelines.

Who Can We Rechallenge with ICIs?

It is possible to rechallenge selected patients who have experienced a renal irAEs with ICI therapy. The decision to treat with a different class of ICI versus continuing the same agent is unclear [39]. Generally speaking, the indication for ICI therapy (adjuvant vs. palliative) affects the decision to rechallenge, with a lower threshold to stop therapy in the adjuvant setting. For patients with renal irAEs, it is reasonable to rechallenge those whose creatinine normalizes rapidly (at least grade 1) on steroids. Those whose creatinine did not normalize, those who took longer to normalize (>30 days), or those with advanced renal disease at baseline would be at higher risk of permanent renal failure with a rechallenge. A rechallenge should be attempted in collaboration with a nephrologist who is familiar with the management of renal irAEs and with extremely close monitoring. Because data are so limited, we routinely engage in a shared decision‐making discussion with the patient ensuring they fully understand the risks and benefits of rechallenging with an ICI versus treatment alternatives. If the decision is made to rechallenge, we establish a new baseline creatinine. We then follow labs weekly for the first several months after starting retreatment. We recommend ongoing careful assessment to detect any rise in creatinine of at least 1.5‐fold from the new baseline. If such an event were to occur, we suggest immediately holding ICI therapy and restarting steroids. We have not pre‐emptively treated patients with prednisone or another therapy to prevent a relapse of AIN when rechallenging with an ICI; however, this is another area in which future research is needed. Large studies of the renal outcomes of retreated patients are needed.

Are There Any Biomarkers that May Be Useful in Predicting irAEs in These Patients?

Data regarding biomarkers for predicting the occurrence of irAEs are lacking. Studies focusing on ipilimumab related gastrointestinal toxicity have identified clonal CD8 T‐cell expansion, baseline IL‐17 levels, and increased expression of the neutrophil marker CD177 as potential biomarkers for predicting the occurrence of irAEs [40], [41], [42]. Hypothesizing that that the severe inflammatory response induced by ICIs is responsible for irAEs, markers of an acute inflammatory response are likely to be of interest in future studies. More research is required to identify and validate such biomarkers before they can be used in a clinical setting.

Are There Special Concerns for Patients with Baseline Chronic Kidney Disease?

Unlike typical “nephrotoxins,” which exert a toxic effect on the renal tubule that is generally dose dependent, renal irAEs are an immune‐mediated phenomenon. For a nephrotoxin, there is generally speaking a higher incidence of nephrotoxicity in patients with pre‐existing renal disease. We have found that AKI after ICI therapy did not occur more commonly in patients with baseline estimated glomerular filtration rate <60 mL per min per 1.73m² [18]. One particular concern, however, is that those with baseline renal disease have less “renal reserve,” and repeated AKI events may push a patient closer to end‐stage renal disease. Thus, clinicians must exert caution when rechallenging patients with pre‐existing renal disease with ICI therapy in the event of a prior AKI from ICI‐related AIN. Similarly, although those with solitary kidneys are unlikely to have a higher risk of developing a renal irAE, they have less reserve for renal recovery, similar to a patient with pre‐existing renal disease [43].

Patients with a variety of autoimmune conditions are typically excluded from clinical trials; thus, there are no data to guide management in patients with autoimmune kidney diseases, such as ANCA disease or lupus nephritis, but there are data to suggest these diseases may worsen in the setting of ICI therapy [44], [45].

What Is the Prognosis for Patients with Renal irAEs?

Although death has been reported in patients with renal irAEs, most patients show significant recovery with steroid therapy [12], [19]. Of 12 biopsy confirmed cases of AIN reported in one series, 10 patients had partial or complete renal recovery, with renal function typically showing improvement within several days of steroid initiation [12]. Therefore, aggressive workup and management are indicated in this patient population given the potential for long‐term benefit from ICI therapy.

Conclusion

Renal irAEs will become increasingly common in oncology practices as more patients are treated with ICIs and with combination therapies. Clinical presentations of renal irAEs can be nonspecific and may manifest with isolated elevation of the serum creatinine. Effective management requires a coordinated, multidisciplinary effort that is focused on early diagnosis and treatment. Nephrology referral should be considered in patients with grade 2 or higher toxicity, cases of recurrent nephritis, or complex presentations in which other etiologies of rising creatinine need to be ruled out by kidney biopsy. Treatment typically consists of holding the ICI and administering corticosteroids. With this approach, most patients will have resolution of their AKI within weeks. Atlthough retreatment with an ICI can be attempted in patients with renal irAEs who are closely monitored, the risks of this approach need to be defined. Focusing on the development of better diagnostics and better therapies to treat autoimmune‐toxicities while maintaining antitumor immunity is key. Steroid‐sparing strategies are needed [37].

Author Contributions

Conception/design: Meghan E. Sise, Harish Seethapathy, Kerry L. Reynolds

Manuscript writing: Meghan E. Sise, Harish Seethapathy, Kerry L. Reynolds

Final approval of manuscript: Meghan E. Sise, Harish Seethapathy, Kerry L. Reynolds

Disclosures

The authors indicated no financial relationships.