Renal medullary carcinoma: molecular, pathological and clinical evidence for treatment with topoisomerase-inhibiting therapy

Authors


Edward M. Schaeffer, The James Buchanan Brady Urologic Institute, The Johns Hopkins Medical Institutions, 145 Marburg, 600 N. Wolfe Street, Baltimore, MD 21287, USA. e-mail: eschaef4@jhmi.edu

Abstract

Study Type – Aetiology (case series)
Level of Evidence 4

OBJECTIVE

To present the molecular rationale and potential clinical benefit of topoisomerase II (TopoII)-inhibiting therapy for renal medullary carcinoma (RMC), a rare but extremely lethal form of kidney cancer that classically afflicts young men with sickle-cell trait. The current therapeutic approach with these aggressive tumours is radical nephrectomy followed by systemic chemotherapy, but the prognosis remains dismal.

MATERIALS AND METHODS

The whole-genome expression was analysed in four RMC tumours. We also report a case of metastatic RMC in which a complete response was achieved for 9 months using a TopoII-inhibiting therapy.

RESULTS

Expanded whole-genome expression analysis showed increases of TopoII in all cases. There was also overall deregulation of DNA remodelling and repair, and an ontological association between RMC and urothelial carcinoma. Using a TopoII-inhibiting agent, there was a complete response for 9 months in a patient with metastatic RMC.

CONCLUSION

This report provides molecular evidence for the rational use of TopoII inhibitors in the treatment of RMC.

Abbreviations
RMC

renal medullary carcinoma

TopoII

topoisomerase II

GO

gene ontology

MVAC

methotrexate, vinblastine, adriamycin and cisplatin.

INTRODUCTION

Renal medullary carcinoma (RMC) is a rare and lethal tumour which typically affects young individuals with sickle-cell disease or trait [1]. Although this disease entity is rare, it is of particular interest due to its highly aggressive phenotype, resulting in locally advanced and metastatic disease at diagnosis in the overwhelming majority of cases [2]. The prognosis for patients with RMC is extremely poor both due to advanced stage at the time of presentation and its resistance to conventional chemotherapeutic agents [3]. With little understanding of the molecular pathways involved in RMC, standard chemotherapy has been used in these patients, often with little success.

Expression profiling [4] and immunohistochemical staining [5] by our research collaborative has been used in patients with RMC, with amplification of the enzyme DNA topoisomerase II (TopoII) noted. We sought to further explore the role of TopoII expression and the entire DNA remodelling machinery pathway in RMC by using expanded, whole-genome expression analysis of four cases of RMC. These findings provided a potential molecular rationale for the treatment of RMC with TopoII agents and agents that target DNA remodelling pathways.

MATERIALS AND METHODS

Tissues were obtained with Institutional Review Board approved procurement protocols, as previously described [4]. Gene expression profiles from 13 non-diseased kidney samples and 76 RCC tumour samples (two medullary, 11 clear cell, 35 papillary, seven chromophobe, seven oncocytoma and 13 TCC) were produced using the HG-U133 Plus 2.0 GeneChip platform (Affymetrix, Santa Clara, CA, USA) as previously described [6], with the inclusion of two additional cases of RMC and whole-genome profiling of these cases with the Affy U133 genechip. Gene expression values were pre-processed using the RMA method, as implemented in the BioConductor Affy package for the R environment. Before data pre-processing, probe set mappings were updated. Pathways were analysed with PGSEA, a previously described parametric gene-set enrichment method [6]. Gene ontologies (GOs) (http://www.geneontology.org) containing the TOP2A gene were examined in addition to the hypoxia gene signature, which was obtained from a previous report [7]. For each pathway and gene, the values plotted are an average of the samples within each tumour subtype and are shown relative to the average score of the non-medullary tumour tissues. A 99.9% significance threshold (0.01) was used to identify genes that could distinguish significantly between the patient groups vs the random patient groupings.

Samples were immunostained on 5 µm thick, formalin-fixed paraffin sections, using the biotin-avidin system with mouse monoclonal antibodies, respectively, specific for DNA TopoIIα (Vector Laboratories, Burlingame, CA, USA), as described previously.

RESULTS

We analysed the whole genome by microarray in four cases of RMC; the patients’ age range was 20–35 years. All specimens used in the analysis were obtained from snap-frozen renal tumours. Whole-genome microarray analysis showed significant increases in TopoII expression (Fig. 1A). GO pathways that contain TopoII were subsequently segregated, and showed enrichment of several pathways regulating DNA topology, including GO categories such as DNA replication, DNA repair and microtubule reorganization (Fig. 1B). Interestingly, increases in these DNA remodelling pathways ontologically link RMC to urothelial TCC (Fig. 1C) and differentiate it from other classical renal parenchymal tumours.

Figure 1.

Hierarchical clustering of renal tumours: A, Differential expression of TopoII (heatmap key: red indicates increased expression relative to control; blue indicates decreased expression relative to control. Control RNA extracted from normal renal parenchyma). B, Differential expression of DNA topography pathways stratified by GO terms. C, Differential expression of hypoxia pathways stratified by GO terms. MED, RMC; PAP, renal papillary; CC, clear cell; CR, chromophobe; ON, oncocytoma; TCC, upper tract urothelial carcinoma.

One patient in the series, a 35-year-old African-American man with sickle-cell trait, had presented with gross haematuria, with subsequent imaging showing a large right renal mass radiographically localized to the kidney. He had a laparoscopic radical nephrectomy and hilar lymph-node dissection. Pathological analysis showed node-positive RMC. At 2 weeks after surgery the patient presented with parieto-occipital skin deposits and mucosal-based lesions in the mouth, anterior to the front teeth. CT showed mediastinal adenopathy, numerous lung lesions, a brain lesion and an intracardiac mass consistent with metastatic disease. He had seven cycles of chemotherapy with carboplatin, gemcitabine and paclitaxel, with initial reductions in the size of both his pulmonary lesions and brain metastases. However, after 6 months he complained of increasing headaches and a repeat MRI showed enlargement of the brain lesion. Soon thereafter he developed low extremity pain and was found to have an arterial thrombus. He had an emergency thrombectomy of his common femoral artery, and pathological analysis showed tumour cells within the thrombus. After the recovery period he then had brain radiotherapy and concomitant salvage chemotherapy with a regimen consisting of adriamycin (50 mg/m2) and gemcitabine (1000 mg/m2). He had an excellent response, with resolution of the hilar mass, lungs nodules and mucocutaneous lesions. This complete response lasted 9 months. Subsequently, the patient developed recurrences in the lung and mediastinum. Despite treatment with bortezomib the disease progressed and he ultimately died 25 months after his initial diagnosis.

This patient’s tumour showed elevated TopoII levels, as shown in Fig. 1. This increase in expression of TopoII at the gene level also correlated with elevations in TopoII protein by immunohistochemical staining (Fig. 2).

Figure 2.

(a) Haematoxylin and eosin-stained RMC; (b) immunohistochemistry with anti-TopoII antibody shows intense nuclear staining of RMC tumour cells.

DISCUSSION

RMC is one of the rarest and most aggressive forms of renal cancer. First described in 1995, this subtype affects young patients, typically males with sickle-cell disease or trait [1]. At the time of diagnosis RMC is usually advanced, with most patients presenting with metastatic disease [2]. Reviewing the available survival data, Simpson et al.[3] found a mean survival of 19 weeks from the time of initial diagnosis. In comparison to contemporary cases of RCC, which are most often discovered incidentally, RMC typically presents with symptoms including flank pain, haematuria, a palpable mass or frank symptoms of metastatic disease [8]. There also appears to be an association between RMC and right-sided tumours [3,8].

To date, the prognosis for patients diagnosed with RMC remains extremely poor, in large part due to its presentation at an advanced stage and resistance to conventional chemotherapy agents [3]. Radical nephrectomy has been used in these patients, but is limited because most patients already have metastatic disease at the time of surgery. In a few patients who have undergone radical nephrectomy without metastatic disease, surgery appears to prolong survival [9,10]. Unfortunately, up to 95% of patients have metastatic disease at the time of diagnosis [3]. Various chemotherapy regimens have also proven to have little effect on long-term survival [2,3,8]. Recently the use of high-dose intensity methotrexate, vinblastine, adriamycin and cisplatin (MVAC) was reported to achieve partial responses and prolong survival in a few patients with RMC [11]. Thalidomide was also reported to achieve a greater than expected response (52 week survival from diagnosis) in one patient with metastatic bone lesions at the time of presentation [3]. A complete response using the protease inhibitor bortezomib has also been reported in one patient [12]. Despite these anecdotal successes, the overall response rate for patients with metastatic RMC treated with chemotherapy remains poor.

Several investigators have attempted to further characterize the genetic changes associated with RMC. Simpson et al.[3] found the presence of BCR and ABL gene amplification in patients with RMC, but evidence of a BCR-ABL translocation was not detected. Our collaborative research efforts on this disease first began with gene-expression profiling on two patients with RMC, where we noted TopoII amplification [4]. More recently, our group has used immunohistochemistry to show TopoII expression in 11 of 13 RMCs [5]. This report of whole-genome expression analysis of four cases of RMC is the largest expression analysis of RMCs to date. This work suggests that therapy directed against TopoII with inhibitors such as adriamycin or etoposide might confer better chemosensitivity. In addition, the whole-genome analysis also showed associations between RMC and whole classes of DNA remodelling pathways, including microtubule reorganisation. This expression analysis might also expand the scope of agents used against RMC.

The increases in DNA topography pathways differentiate RMC from other classical renal tumours which contain elevations in ‘hypoxia pathways’, and make use of some of the emerging new agents for other renal parenchymal tumours less intellectually appealing. Interestingly, elevations in these DNA remodelling pathways ontologically link RMC to urothelial carcinoma, suggesting that the development of chemotherapy regimens against RMC should potentially mirror those of urothelial carcinoma. Together, these molecular, pathological and clinical data provide comprehensive evidence for directed treatment of RMC using TopoII-inhibitor based chemotherapy, and further implicate the entire DNA remodelling process as a key RMC-centred process.

TopoII-inhibitor-containing regimens have previously been used for RMC [13,14]; indeed, there are case reports using MVAC, a regimen containing adriamycin, a TopoII-inhibitor. Interestingly, when Strouse et al.[13] reviewed treatment strategies for RMC they noted that patients receiving MVAC had better mean (sd) survival than those treated with alternative agents, of 8 (2) vs 5 (1.4) months. The present study helps to elucidate the underlying explanation for these survival differences. This also might explain the positive findings with high-dose intensity MVAC discussed previously [11]. In addition, the present molecular profiling might also help to direct the use of new, emerging TopoII-targeting agents for patients with RMC [15]. Other recent studies suggested improvements in survival using platinum-based chemotherapy and bortezomib [12,16]. The precise cause of these improved outcomes remain unknown; nonetheless, the present report provides persuasive evidence that directed treatment for RMC should contain an agent targeting DNA TopoII in an effort to improve survival in these unfortunate cases.

In conclusion, RMC remains an often fatal disease with an extremely poor prognosis at presentation. Molecular profiling provides a rational basis for treatment with targeted therapeutic agents. Future clinical studies with TopoII inhibitors in this patient population are warranted to further define the role of such agents.

CONFLICT OF INTEREST

None declared.

Ancillary