Exceptional Response to Cabozantinib in a Patient With Multiply Relapsed Wilms Tumor

INTRODUCTION

Wilms tumors (WTs) account for a majority of pediatric renal tumors and have a high cure rate with surgery, radiation, and chemotherapy.¹ High-risk features in WTs include anaplastic histology, blastemal predominance, bilateral presentation, and loss of heterozygosity of 1p and 16q, and patients who have these features typically have worse outcomes and are treated with more aggressive therapies. Despite the risk-stratified approach to therapy, outcomes remain poor for the subgroup of patients with WTs who relapse.² Better understanding of the biology of WT could help identify risk factors associated with disease relapse. It has been shown that integrative clinical sequencing (ICS) has the potential to enable a better understanding of the biology of pediatric malignancies and to help identify newer targets which can be exploited for novel therapies.³ Cabozantinib and sorafenib are both multi tyrosine kinase inhibitors (TKIs) that target many of the same kinases with some key differences.^4,5 In this article, we review the details of a patient who had refractory WT, a unique genomic profile with prolonged survival, and exceptional responses to these TKIs.

CASE REPORT

In December 2000, a 6-year-old male presented with favorable histology in stage V WT. The patient received first-line therapy with vincristine, actinomycin, and adriamycin, left nephrectomy, whole lung radiation, and abdominal radiation. In November 2005, 3.5 years after completion of therapy, the patient experienced a widespread relapse that was detected on routine surveillance and was treated with cyclophosphamide, etoposide, carboplatin, vincristine, and radiation.

Two years after therapy for the relapse, a second recurrence was detected on imaging; he underwent surgical resection and received treatment consisting of ifosfamide/etoposide, cyclophosphamide/topotecan, and autologous stem cell transplantation. The patient’s stem cell transplantation was complicated by severe BK viral hemorrhagic cystitis infection, which necessitated a prolonged stay in the intensive care unit.

In October 2010, 2 years after transplantation, the patient began experiencing back pain and was found to have a third relapse, which was treated with vincristine, adriamycin, and irinotecan. Approximately 1 year after this therapy was completed, a fourth relapse was treated with eight cycles of sorafenib, which led to stable disease, after which disease progression was detected and the patient began radiation therapy. His disease continued to progress through various experimental therapies spanning 1.5 years which included a trial of Seneca Valley virus, sirolimus, and tivantinib. In July 2013, the patient was then treated with cabozantinib, and he experienced a partial response lasting close to 2 years (Fig 1).

Fig 1.

Treatment responses revealed by computed tomography scan. The patient experienced a 33% reduction in combined target lesion tumor size by Response Evaluation Criteria in Solid Tumors (RECIST).

After his eventual progression on cabozantinib, the patient advanced through several other therapies, including additional radiation, lorvotuzumab mertansine, axitinib, ramucirumab, and pazopanib. Unfortunately, the patient succumbed to his disease just short of 17 years after his original diagnosis.

The patient’s tumor tissue, obtained from a biopsy performed after his fourth relapse, underwent an ICS (paired tumor/normal DNA whole-exome sequencing and tumor transcriptome sequencing) study, which was approved by the institutional review board (UM-IRBMED HUM No.00056496). The patient’s parents provided informed consent and received mandatory pre-enrollment genetic counseling.^3,6 Tumor content of the biopsy was determined to be 80% to 90%. He was found to have somatic MAX (p.R60Q) and MYCN (p.P44L) mutations and AMER1 (WTX) deletion. Both MAX and MYCN mutations are monoallelic events with no evidence of loss of heterozygosity in the tumor. In addition, transcriptome sequencing (RNA sequencing) uncovered various overexpressed tyrosine kinases (RET, MET, and TEK/TIE2) when compared with the rest of our institution’s compendium of patients with WT (Table 1). The patient was also found to have a rare, putative pathogenic HOXB13 germ line nonconservative substitution, p.G84E. A review of the family history showed that the mother was positive for melanoma (age of onset, 37 years), and the paternal grandfather had basal cell cancer (age of onset, 64 years).

Table 1.

Kinase Expression in Multiply Relapsed/Refractory Patient With WT

DISCUSSION

Prognosis for relapsed WT continues to be poor in spite of high-dose intensive therapy. Despite having favorable histology and lacking genetic biomarkers indicating high-risk disease, some relapsed patients continue to be refractory to treatment. Genetic aberrations discovered through ICS can help identify new prognostic markers, can help researchers better understand tumor biology, and can sometimes lead to targeted therapies.³

Somatic mutations in MAX (p.R60Q) and MYCN (p.P44L) were identified in our multiply relapsed patient with unquestionably refractory disease. The MYC proto-oncogene has been implicated in the pathogenesis of many human tumors and was found to be overexpressed and/or activated in more than half of human cancers.^7,8 MYC coordinates changes in the tumor microenvironment, including the activation of angiogenesis.⁹ In patients with WT, MYCN is associated with the high-risk histology features of anaplasia and blastemal predominance as well as poorer relapse-free survival and overall survival. Multiple mechanisms of MYCN dysregulation have been found, including P44L somatic mutation and copy-number gain and loss of function of FBXW7.¹⁰ MAX (MYC-associated Factor X) is a transcription factor of the basic helix-loop-helix leucine zipper (bHLHZ) family which is able to form a homodimer or heterodimer with other family members, including MXD1 (MAD), MXI1, and MYC. MAX (p.R60Q) has been reported in one other patient with WT¹¹ and is the most common MAX gene mutation reported in the Catalogue of Somatic Mutations in Cancer (COSMIC) database (v82), which suggests that it may promote MAX oncogenic activity (Fig 2).^12,13 In silico modeling of the mutation has shown that MAX (p.R60Q) abolishes MAX homodimerization and DNA binding; however, its effects on MAX heterodimerization with C-MYC or other MYC family members is unknown.¹³ Regardless of the patient’s not having other histologic high-risk features such as anaplasia, he had refractory disease. This suggests the potential for a poor prognostic implication of combined MAX and MYCN mutations in this case, but further functional studies are required to validate this interaction.

Fig 2.

Illustration of protein changes in MAX with nonsynonymous alterations identified in this patient and in COSMIC v82. Each circle represents a single case and types of cancers are distinguished by color. HLH, helix-loop-helix domain.

Cabozantinib is an oral small molecule inhibitor of numerous tyrosine kinase receptors with activity toward VEGFR2 (KDR) and MET (hepatocyte growth factor receptor). It also targets important mediators of tumor cell survival, metastasis, and tumor angiogenesis, including RET (rearranged during transfection), KIT (mast/stem cell growth factor receptor), AXL (anexelekto), TIE2 (angiopoietin receptor), and FLT3 (Fms-like tyrosine kinase).^4,14 Cabozantinib has been approved by the US Food and Drug Administration to treat advanced renal cell carcinoma (aRCC) and has shown efficacy against medullary thyroid cancer.¹⁵ Sorafenib targets many of the same kinases as cabozantinib and also targets CRAF, BRAF, VEGFR3 (FLT4), and PDGFR-β.¹⁶

Because of the patient’s exceptional response to cabozantinib, prolonged stable disease while receiving sorafenib, and seemingly no genetic mutations or amplifications that could be targeted, we looked at the expression of various kinases that are inhibited by both agents using RNA sequencing (Table 1). It is known that RET plays a role in the development of the kidney, and expression of high levels of RET in WT has been reported; therefore, this overexpression may play a role in the development of the disease.¹⁷ In addition, the patient’s tumor was likely highly vascularized as a result of his somatic MYCN mutation.⁹ We hypothesize that inhibition of the VEGF signaling pathway enacted by both sorafenib and cabozantinib helps devascularize the tumor, leading to sustained response. Previous research in aRCC substantiates our experience, which shows cabozantinib to be superior to other multi-TKIs such as sorafenib and axitinib, leading to longer progression-free survival.¹⁸ Further evidence in aRCC suggests that increased expression of MET and AXL are associated with a poor prognosis and that their inhibition may help overcome resistance to VEGF pathway inhibition.^19,20

The rare HOXB13 germ line variant uncovered in this patient has been inconsistently reported to increase cancer risk. In hepatocellular carcinoma, overexpression of HOXB13 has been associated with tumor angiogenesis and disease progression.²¹ There are no reported cases of this germ line mutation in WT, but it seems to be associated with an increased risk of prostate cancer, particularly in whites.^22,23 Thus, this finding led to genetic counseling for the patient and his family.

To our knowledge, this is the first report of a patient with WT who had aberrations in both MAX and MYCN genes; in fact, if the MYC/MAX interaction plays a significant prognostic role in WT, agents are being developed that could potentially target this interaction.²⁴ Generally, overexpression is not considered a targetable finding in precision oncology, but if there are no other tangible aberrations, targeting overexpression may be reasonable. This article also highlights the importance of reviewing patient’s responses to targeted therapy in association with their molecular sequencing findings in an attempt to gain new insight into potential genetic targets for biologic agents.

AUTHOR CONTRIBUTIONS

Conception and design: Bailey Anderson, Rama Jasty-Rao, Trisha Paul, Rajen J. Mody

Provision of study materials or patients: Rama Jasty-Rao

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.

Bailey Anderson

No relationship to disclose

Rama Jasty-Rao

No relationship to disclose

Yi-Mi Wu

No relationship to disclose

Trisha Paul

No relationship to disclose

Dan Robinson

No relationship to disclose

Rajen J. Mody

No relationship to disclose