Karyomegalic interstitial nephritis (KIN) is a rare, slowly progressive tubulointerstitial kidney disease characterized by enlarged, hyperchromatic pleomorphic nuclei in the renal tubular epithelium with development of chronic interstitial nephritis. First described in 1974, Mihatsch subsequently coined the term KIN in a description of cases with systemic karyomegaly and chronic interstitial nephritis [1, 2]. KIN was found to be associated with mutations in the FAN1 (Fanconi anemia-associated nuclease 1) gene, which is involved in the interstrand cross-link repair DNA damage response in an autosomal recessive pattern, with cell cycle arrest and reduced tubular cell mitotic activity [3,4,5,6]. Subsequently, acquired forms of KIN were identified associated with medications including ifosfamide, immune checkpoint inhibitors and pemetrexed with cisplatin or carboplatin as possibly contributing factors, and in ochratoxin A toxicity [7,8,9,10,11,12,13]. Prevailing theories of mechanisms causing genetic and drug-induced KIN include disruption or uncoupling of the cellular proliferation process with cell cycle arrest, and increase in reactive oxygen species (ROS) often involving mitochondrial dysfunction, which are believed to work synergistically to induce karyomegalic cellular changes [4, 7, 10, 14]. However, the overall pathogenesis is uncertain, including necessary and sufficient conditions for KIN development.

In cases of drug-induced KIN, there has been no case report to date in which genetic studies were performed to identify a possible genetic predisposition for disease. Such genetic-environmental double-hit mechanisms occur in kidney diseases; for example, high interferon states and APOL1 mutations in collapsing glomerulopathy or alternate complement pathway activation and CFH mutations in C3 glomerulopathies [15, 16]. For KIN, Fan1−/− mice are more susceptible to cisplatin toxicity and fail to recover from cisplatin-induced AKI in contrast to wild-type littermates, suggesting that FAN1 mutations may cause sensitivity to certain medications [4]. Therefore, it may be of clinical importance to know if an underlying genetic predisposition is required for the development of drug-induced KIN.

Herein we present a patient who developed KIN in the setting of Hodgkin lymphoma treated with several chemotherapeutic agents, including carboplatin, ifosfamide, and brentuximab vedotin (brentuximab). While carboplatin and ifosfamide are both heavily implicated in KIN formation, brentuximab previously has not been reported as a cause of KIN. Additionally, genetic testing was performed in this case for detection of FAN1 mutations.

A 49-year-old female was diagnosed with anaplastic high-grade Hodgkin’s lymphoma of the stomach in 2018, with metastasis to the liver, large intestine, adrenal gland, peritoneum, and periaortic and splenic hilar lymph nodes. She initially was treated with 4 cycles of ABVD chemotherapy (Doxorubicin, Bleomycin, Vinblastine, Dacarbazine), which was complicated by neutropenic fevers. She then was changed to ICE (Ifosfamide, Carboplatin, Etoposide) therapy for residual gastric and intra-abdominal nodal disease. After 3 cycles of ICE, she was hospitalized for myelosuppression with pancytopenia, nausea and vomiting, and acute kidney injury with a serum creatinine of 1.9 mg/dL. She was hydrated and the serum creatinine improved to 1.09 mg/dL. She then was given 2 doses of brentuximab 3 weeks apart, at which point her creatinine steadily increased and estimated glomerular filtration rate (eGFR) dropped below 30 ml/min/1.72m2, prompting discontinuation of brentuximab.

She was seen in nephrology clinic 2 months after the last dose of brentuximab. She had no significant family history of kidney disease and denied smoking or alcohol use, or exposure to ochratoxin A, heavy metals, lithium, or non-steroidal anti-inflammatory drugs. Toxicological work-up was negative. Serological tests for ANA, ANCA, anti-dsDNA, HIV, HBV, and HCV were negative. Serum complement C3 and C4 were within normal range.

At the time of initial consultation, the serum creatinine was 2.75 mg/dL, BUN was 15 mg/dL and eGFR was 19 ml/min/1.72m2. Other laboratory tests included sodium 132 mmol/L, potassium 3.2 mmol/L, phosphorus 1.8 mg/dL, albumin 3.2 g/dL and glucose 108 mg/dL. Urine analysis showed 3 + protein, 4 + glucose, 1 + ketones and no cells or blood. The urine protein to creatinine ratio was 2.8. A renal ultrasound showed normal sized kidneys with mild increase in cortical echogenicity.

A renal biopsy was performed 5 months after last dose of ICE and 2 months after the last dose of brentuximab. There were 16 glomeruli, none of which were globally or segmentally sclerotic. There were no significant glomerular findings by light or electron microscopy and the immunofluorescence was negative. Acute tubular injury was present in most proximal tubules. The tubular epithelial cells focally had enlarged, hyperchromatic, irregularly shaped nuclei with cytoplasmic attenuation and frequent denudation of tubular basement membranes, more prominent in the proximal tubules. (Fig. 1a,b) There was mild to moderate predominantly evolving tubular atrophy with interstitial fibrosis, along with interstitial edema and a focal interstitial lymphocyte-predominant inflammatory infiltrate with minor tubulitis. (Fig. 1c) Arteries showed moderate to severe arteriosclerosis. Polyomavirus SV40 and CMV immunohistochemistry (IHC) stains were negative. Ki-67 IHC showed a very low proliferative index in the tubular epithelial cells. (Fig. 1d). IHC for CD30 was negative. No viral inclusions were found ultrastructurally.

Kidney biopsy showing. A) Tubular injury with flattened epithelial cells and focal interstitial inflammation. The glomeruli are unremarkable or modestly enlarged.(Jones methenamine silver × 100). B Tubular epithelial cells with enlarged, hyperchromatic nuclei and flattened attenuated cytoplasm (arrows) with foci of denuded tubular basement membranes. Note the unremarkable distal tubules (arrowheads). (Hematoxylin and Eosin X 400). C Tubular cell injury with focal karyomegaly and interstitial inflammation. (Periodic acid Schiff × 200). D Ki67 immunohistochemical stain showing a low proliferative index in the tubular epithelial cells. (× 200)

Genetic assessment using Natera Renasight found no mutations in the FAN1 gene. A variant of unknown significance was found in the PRODH gene, which encodes for a mitochondrial protein involved in proline degradation, which may generate ROS. No other mutations were identified that might impact renal tubular epithelium or result in karyomegaly.

One month after kidney biopsy, the patient had an interval increase in lymphoma activity and was started on pembrolizumab. Three months after renal biopsy, her serum creatinine peaked at 6.17 mg/dL, then started to improve and remained stable with an eGFR of 15–20 ml/min/1.72m2 without the need for renal replacement therapy over the ensuing 4 years (Fig. 2). No steroids were given due to concerns about interference with lymphoma immunotherapy. She remains on pembrolizumab with stable lymphoma and stable kidney function.

Timeline of the patient’s serum creatinine levels with respect to medication exposures and kidney biopsy