Flow cytometry of renal oncocytoma: common occurrence of deoxyribonucleic acid polyploidy and aneuploidy

A non-diploid DNA status is linked to poor prognosis in renal cell cancer

Purpose

DNA ploidy measurement has earlier been suggested as a potentially powerful prognostic tool in many cancer types, but the role in renal tumors is still unclear.

Methods

To clarify its prognostic impact, we analyzed the DNA content of 1320 kidney tumors, including clear cell, papillary and chromophobe renal cell carcinoma (RCC) as well as renal oncocytoma and compared these data with clinico-pathological parameters and patient prognosis.

Results

A non-diploid DNA content was seen in 37% of 1276 analyzable renal tumors with a striking predominance in chromophobe carcinoma (74.3% of 70 cases). In clear cell carcinoma, a non-diploid DNA content was significantly linked to high-grade (ISUP, Fuhrman, Thoenes; p < 0.0001 each), advanced tumor stage (p = 0.0011), distant metastasis (p < 0.0001), shortened overall survival (p = 0.0010), and earlier recurrence (p < 0.0001). In papillary carcinoma, an aberrant DNA content was significantly linked to high Fuhrman grade (p = 0.0063), distant metastasis (p = 0.0138), shortened overall survival (p = 0.0010), and earlier recurrence (p = 0.0003).

Conclusion

In summary, the results of our study identify a non-diploid DNA content as a predictor of an unfavorable prognosis in clear cell and papillary carcinoma.

Electronic supplementary material

The online version of this article (10.1007/s00345-020-03226-8) contains supplementary material, which is available to authorized users.

Renal Oncocytomas (Typical and Atypical Variants): A Pathologic, Immunohistochemical, Morphometric, and Flow Cytometric Differential Study of 14 Cases With Cytogenetic Support

Eight conventional and six atypical oncocytomas in a series of 147 renal neoplasms were studied. Histopathologic findings revealed exclusively oncocytic cells, but cellular polymorphism was higher in the atypical tumors. Atypical oncocytomas presented focal necrosis, transcapsular invasion, or both. Electron microscopy showed similar findings in all cases. Immunohistochemistry of atypical oncocytomas had higher expression against proliferating cell nuclear antigen and more discontinuous immunostaining against laminin than typical ones. Flow cytometry revealed one or two aneuploid peaks in five typical and two atypical cases, although the latter had a higher proliferative fraction than typical oncocytomas. Cytogenetics of one typical oncocytoma showed a normal diploid karyotype; one atypical case resulted in a diploid karyotype but with endoreduplications in 13% of metaphases, and a second atypical oncocytoma became hypodiploid without structural anomalies. Based on the present results, the proposed distinction between conventional and atypical oncocytomas seems of limited clinical significance.

Pseudotetraploid clone with structural chromosomal rearrangements in a chondromyxoid fibroma: a case report

Chondromyxoid fibroma is a rare benign tumor accounting for 1-2% of primary bone tumors. Most of the patients are young males in the 2nd and 3rd decades of life. Metaphyses of long bones are predominantly affected. The histology of this tumor is well established, but its genetic mechanism remains poorly characterized. To our knowledge, only 22 abnormal cytogenetic analyses have been reported, and all contained diploidy or near-diploidy karyograms as their primary event, and inv(6)(p25)(q13) and rearrangements involving regions 6p23-25, 6q12-15, and 6q23-27 constituted a recurrent observation. In this report, a pseudotetraploidy tumor clone with multiple numerical and structural aberrations involving 6p23 as well as other chromosomal loci was identified in a chondromyxoid fibroma from the metaphysis of the left fibula of an 18-year-old male, which has not been reported. The finding may relate to the atypical-looking large cells often seen in this benign tumor.

Update on oncocytoma

Oncocytoma is the most common benign solid renal tumor, comprising roughly 5% of resected renal masses. Typically discovered incidentally, oncocytoma is generally asymptomatic and very rarely metastasizes; however, multifocal disease and coexistence with renal cell carcinoma can occur. No currently used imaging techniques can reliably distinguish between oncocytoma and malignant lesions; therefore, patients must undergo resection, or in certain circumstances, biopsy, to definitively establish diagnosis. Careful attention to pathologic features and the adjunctive use of immunostains can aid in discriminating oncocytoma from other renal tumors characterized by granular, eosinophilic cytoplasm, especially chromophobe renal cell carcinoma. Nephron-sparing and laparoscopic surgical approaches can be used to treat appropriately selected patients.

Mutations of the VHL gene in sporadic renal cell carcinoma: definition of a risk factor for VHL patients to develop an RCC

To investigate the nature of somatic von Hippel-Lindau (VHL) mutations, we analyzed 173 primary sporadic human renal cell carcinomas for mutations of the VHL tumor suppressor gene, using polymerase chain reaction (PCR) and single-strand conformational polymorphism analysis (SSCP) of DNA. We detected abnormal SSCP pattern in 73 samples. After sequencing, we identified microdeletions in 58% of cases, microinsertions in 17%, nonsense mutations in 8%, and missense mutations in 17%. Among these mutations, 50% correspond to new mutations. VHL mutations were found only in the nonpapillary renal cell carcinoma (RCC) subtype, as previously reported. To compare somatic and germline mutations, we used the VHL database, which includes 507 mutations. The study of mutational events revealed a significant difference between somatic and germline mutations with mutations leading to truncated proteins observed in 78% of somatic mutations vs only 37% in germline mutations (P < 0.001). We postulated that a specific pattern of VHL mutations is associated with sporadic RCC. This pattern corresponds to mutations leading mainly to truncated proteins with few specific missense mutations. We then analyzed the occurrence of RCC in VHL families, based on the nature of mutations. We observed RCC in at least one member of the VHL families in 77% of cases with mutations leading to truncated proteins versus 55% in cases with missense mutations (P < 0.05). Thus, mutations resulting in truncated proteins may lead to a higher risk of RCC in VHL patients.

Lack of genetic changes at specific genomic sites separates renal oncocytomas from renal cell carcinomas

Morphological similarities between renal oncocytomas and 'oncocytic' renal cell carcinomas (RCCs) make a differential diagnosis in many cases difficult. A series of 41 renal oncocytomas has been analysed by microsatellite markers from chromosomes 1, 2, 3p, 6q, 8p, 9, 10, 13q, 14q, 17, and 21, alterations of which are known to be involved specifically in non-papillary and chromophobe RCCs. Only eight of the 41 renal oncocytomas showed loss of heterozygosity (LOH). LOH at chromosomes 1 and 14 occurred in four tumours each and at chromosomes 2, 8, and 9 in one tumour each. Combined LOH at chromosomes 1, 9, and 14 and also at chromosomes 1 and 14 occurred in one case each. No LOH was seen at any other genomic sites. The lack of combination of LOH at specific chromosomal sites differentiates renal oncocytomas from other renal cell tumours with overlapping phenotypes. Applying the microsatellite assay described here, the diagnosis can be established within 2 days, from fresh as well as from paraffin-embedded material.

Renal oncocytoma: a clinicopathologic study of 70 cases

We reviewed 954 primary nonurothelial epithelial renal neoplasms with primary resection at Memorial Sloan-Kettering Cancer Center between the years 1980 and 1995 and classified 70 cases (7%) as renal oncocytomas. The study population was composed of 39 men and 31 women, and the mean age was 65 years (range 25 to 86 years). Fifty-six patients (80%) were asymptomatic at presentation, six (4%) had flank pain, six (4%) presented with a mass, and two (3%) had hematuria. Sixty-one were treated with total or radical nephrectomy, nine with partial nephrectomy. The right kidney was involved in 35 cases (50%), the left kidney in 32 (46%). Three cases (4%) were bilateral. Sixty-one cases (87%) were unifocal, nine (13%) multifocal. All the tumors were well circumscribed but unencapsulated. Forty-five (64%) were described as brown or red, whereas the remainder were variously described as tan to yellow. Central fibrosis or scar was described in 23 cases (33%), and gross areas of hemorrhage or cystic changes in 14 (20%). The mean size was 5.2 cm and median 5.0 cm (range 1.5 cm to 14 cm). Histologically, the tumors were characterized by a mixture of architectural patterns: compact cellular nests and acini embedded in a hyalinized, hypocellular stroma were present in 62 cases (89%), a solid nested architecture in 47 cases (67%), and a variable tubular component in 50 cases (71%). Small papillae, pseudopapillae, and intratubular epithelial tufts were seen in 19 cases (27%). Cytologically, the neoplasms also showed a mixture of cell types, the most common being the classic oncocyte, which consisted of round or polygonal cells with moderate to abundant granular, eosinophilic cytoplasm, and small round nuclei with evenly dispersed granular chromatin. Small basophilic nucleoli were visible in many of these cells in all cases. Thirty-one cases (44%) had a variable number of oncocytic cells with pyknotic nuclei and 20 (30%) contained clusters of small cells with a high nuclear/cytoplasmic ratio and dense hyperchromatic nuclei (so-called oncoblasts). Foci of tubules with clear cells embedded in a hyalinized stroma were present in six cases (9%). Cellular atypia was evident in 42 cases (60%) and was marked in 21 (30%). Eleven cases (16%) exhibited mitotic activity, albeit low. No case had atypical mitoses or necrosis. Twenty-two cases (31%) had areas of calcification within the hyalinized stroma, 12 (17%) had calcospherites, and three (4%) had osseous and myeloid metaplasia. Vascular invasion was present in three cases (4%), and invasion of perinephric fat in 14 (20%). One patient presented with liver metastasis. Fourteen cases (20%) were pT1, 42 (60%) pT2, and 14 (20%) pT3. After a mean follow-up of 58 months (range 1 to 181), 62 patients (89%) were alive with no evidence of tumor, six (9%) had died of other causes, one was alive with stable metastatic disease in the liver 58 months after diagnosis, and one died with metastatic disease to bone and liver. We conclude that renal oncocytomas have a varied morphologic appearance and their pathologic diagnosis should be based on a constellation of architectural and cytologic features. The overwhelming majority of cases behave in a benign fashion, although in rare instances they can metastasize. The presence of atypical morphologic features do not alter the excellent prognosis associated with oncocytomas and do not predict an aggressive clinical course.

Microcalorimetric measurements carried out on isolated tumorous and nontumorous tissue samples from organs in the urogenital tract in comparison to histological and impulse-cytophotometric investigations

In this comparative study, microcalorimetric measurements were carried out on a total of 96 tumorous and nontumorous tissue samples taken from organs of the urogenital tract using a thermal activity monitor (TAM). Changes in the heat emission of the tissue samples were measured at 1-min intervals and graphically displayed as a function of time. The aim of the study was to compare the microcalorimetric results with impulse-cytophotometric and histological findings and provide evidence for the metabolic activity of tumorous and nontumorous tissue. In order to obtain the variation in metabolic activity, the maxima (Pmax) of the curves were determined as a value of the maximum thermal power of a tissue sample, the mean values (P) were determined by the mean thermal power and the contour integrals (W) were defined by the behavior of the energy reserves and their mobilization. The first part of the study was carried out to investigate whether tumorous and nontumorous tissue samples differ in general according to their metabolic activity. We discovered, using the parameters described above, that in general tumorous tissue exhibited a higher metabolic activity than nontumorous tissue samples. For example, both W and P in tumorous prostate tissue samples were eightfold higher and the (Pmax) value was 8.4-fold higher than in normal tissue. Additional investigations on testicle and kidney tissues were performed to find a possible correlation between microcalorimetric results and histological grading. We found that an increasing malignancy correlated with a higher metabolic activity of the tissue. Based upon these results we were able to differentiate the various histological gradings of these tumorous tissues by microcalorimetric measurements. The results show it is possible to differentiate between normal and tumorous tissue samples by microcalorimetric measurement based on the distinctly higher metabolic activity of malignant tissue. Furthermore, microcalorimetry allows a differentiation and classification of tissue samples into their histological grading. With the help of microcalorimetry, it might be possible in future to detect and record the metabolic processes of isolated tissue structures and changes in these activities as a result of medical intervention such as cytostatic treatment.

Numerical aberrations of chromosomes 1 and 7 in renal cell carcinomas as detected by interphase cytogenetics

Alcohol-fixed single cell suspensions of 37 renal cell carcinomas (RCCs) were assessed by both flow cytometry (FCM) and the fluorescence in situ hybridization (FISH) technique, using chromosome 1- and chromosome 7-specific centromere DNA probes. DNA diploidy or near-diploidy was observed in 30 of the 37 RCCs and only 12 of these (near-)diploid tumours were disomic for both chromosomes 1 and 7. Numerical aberrations of chromosome 1 and/or chromosome 7 were present in 18 of the 30 (near-)diploid RCCs and five of these cases showed monosomy for chromosome 1 in more than 50 per cent of the tumour cells. A double target FISH, with a centromeric and a telomeric specific probe for 1p36, excluded misinterpretation on the basis of clustering of 1q12, and suggested a complete loss of chromosome 1. All these five (near-)diploid RCCs with monosomy for chromosome 1 were eosinophilic chromophilic cell carcinomas, according to the Thoenes classification of RCC. This observation is of special interest, because it was recently concluded from cytogenetic studies that the diagnosis of chromophilic renal cell carcinoma must be considered as obsolete. Monosomy for chromosome 1 seems to be a non-random numerical aberration of (near-)diploid eosinophilic chromophilic cell carcinomas, and a gain of one or more chromosomes 1 appeared to be a common phenomenon in RCCs, especially in the DNA aneuploid tumours. As these chromosomal abnormalities were not found in the earlier classical cytogenetic studies, we conclude that in situ hybridization techniques are required in addition to chromosome banding techniques to obtain a complete characterization of the chromosome imbalances in RCCs.

Renal oncocytoma: clinical and biological correlates

We treated 31 patients with renal oncocytoma. Renal cell carcinoma was found existing separately within the same or contralateral kidney in 10 patients (32%). Followup of 29 patients revealed 24 alive with no evidence of disease and 1 alive with recurrent oncocytoma, while 1 with coexistent tumors died of progressive renal cell carcinoma and 3 tumor-free patients died of unrelated diseases. Quantitative deoxyribonucleic acid analysis was performed on cell suspensions of fresh tumor by flow cytometry or by image cytometry on touch preparations from frozen tissue in 16 patients with renal oncocytoma. Ploidy analysis revealed all oncocytomas to be diploid. Frozen tissue immunohistology was performed using murine monoclonal antibody against human HLA-A, B and C (class I) antigens with the avidin-biotin peroxidase technique in 11 patients with renal oncocytoma. Ten oncocytomas did not express these self-recognition antigens and 1 was only weakly positive for antigen expression. In contrast, renal cell carcinomas strongly expressed HLA class I antigens. The high incidence of coexistence of renal oncocytoma and renal cell carcinoma has important clinical implications. Loss of HLA class I antigen expression by renal oncocytomas may provide an additional method for differentiating this lesion from renal cell carcinoma.

Cytogenetic analysis of six renal oncocytomas and a chromophobe cell renal carcinoma. Evidence that -Y, -1 may be a characteristic anomaly in renal oncocytomas

Renal oncocytomas are benign tumors whose morphologic features may sometimes be confused with those of certain low-grade malignant neoplasms of the kidney, e.g., chromophobe cell and granular cell variants of renal carcinoma. The presence of a specific genetic abnormality might help differentiate these tumors. Because very few cytogenetic studies of renal oncocytomas have been published, we investigated a consecutive series of six such tumors. We also performed chromosome analysis on a chromophobe cell carcinoma because cytogenetic analyses of this tumor have not been previously reported. Tumor cell metaphases were analyzed after mechanical and enzyme disaggregation, in situ culture, and robotic harvesting. Clonal abnormalities were present in five of the six oncocytomas, and loss of chromosome 1 with loss of the Y chromosome occurred in two. Review of the literature disclosed four other renal oncocytomas with the 44,X,-Y,-1 karyotype. In the chromophobe cell carcinoma, we noted an abnormal clone with a del(11)(p12p15.1); similar anomalies were not observed in the renal oncocytomas. We conclude that renal oncocytomas have clonal chromosome abnormalities and that a subgroup of these tumors may be specifically associated with loss of chromosomes 1 and Y. Because this is a small series, further investigation may help establish whether cytogenetic studies can provide diagnostic and pathogenic information about renal oncocytomas.

Bilateral renal oncocytomas: management and nuclear analysis

Only 16 cases of bilateral renal oncocytomas have been previously reported in the English language literature, 9 of which were treated with bilateral surgical resections. We report a case treated with bilateral partial nephrectomy. Although pathological examination revealed grade I oncocytomas on both sides, flow cytometry demonstrated tumor DNA content to be aneuploid on one side and diploid on the other, the first cellular differences reported in a case of bilateral renal oncocytomas. The management of suspected renal oncocytomas is discussed, with emphasis on the need for complete excision in most patients. The results of nuclear studies of these tumors, including DNA flow cytometry, cytogenetics, and molecular analysis of both nuclear and mitochondrial DNA, are reviewed.

Characterization of conservatively resected renal tumors using automated image analysis DNA cytometry

The DNA histograms of 57 conservatively resected renal tumors were studied using automated image analysis DNA cytometry (Leytas II). Forty-nine of the analyzed tumors were renal cell carcinomas, six were oncocytomas, one was an angiomyolipoma, and one was a renal cell adenoma. On the basis of their DNA histograms, diploid, tetraploid, and aneuploid tumors could be distinguished. Aneuploid tumors could be subtyped further according to the DNA content of the stem cell line as hyperdiploid, hypertriploid, or hypertetraploid. Eight of the tumors were characterized by a combination of diploid and hypertriploid stem cell lines. During a mean follow-up of 5 years, only the two patients with a pure hypertriploid tumor died of distant metastases. These results indicate that automated DNA image analysis cytometry is able to differentiate among several types of renal tumors with obviously different prognoses.

Flow cytometric analysis of cellular deoxyribonucleic acid content of nine renal oncocytomas

The cellular deoxyribonucleic acid content of formalin-fixed, paraffin-embedded tissue from nine renal oncocytomas was determined using an Epics-C flow cytometer. The tissues were all from surgical specimens, and all neoplasms met stringent light and electron microscopic criteria for the diagnosis of oncocytoma. Thirteen samples of pure tumor, 5 of tumor and normal kidney, and 3 of normal kidney alone were analyzed. All oncocytomas had single, sharp G1 peaks with no evidence of aneuploidy, and this peak was similar to that of normal renal parenchyma. This was true even for three oncocytomas that had multiple major chromosomal anomalies on karyotyping. These data support the concept that renal oncocytomas are euploid neoplasms, consistent with their benign biologic behavior. The apparent discrepancy between the results of flow cytometry and karyotyping is discussed.

Renal oncocytomas--an Australian experience

A review was made of 24 cases of renal oncocytoma seen between 1978 and 1989. There was considerable overlap between the clinical presentation of renal oncocytomas and renal carcinomas. Although pre-operative radiological, cytological and pathological investigations may suggest the presence of an oncocytoma, these studies cannot make a definitive diagnosis. We recommend that these tumours be treated as potential renal carcinomas until post-operative microscopic evaluation proves otherwise.

Renal oncocytoma: long-term follow-up and flow cytometric DNA analysis

We report a retrospective study on the clinicopathologic features and flow cytometric DNA analysis of ten renal oncocytomas compared with a control group of ten randomly selected renal cell carcinomas. Among the oncocytoma patients, no recurrences or metastases were noted over an average follow-up of 6.7 years (range = 6 months to 16 years). Reproducible, high-quality DNA histograms were obtained on the paraffin-embedded specimens by using our modified flow cytometric procedure. One aneuploid (10%) and two hyperdiploid tumors (20%) were found in the oncocytoma group. There was no correlation between these abnormal DNA histographic patterns and survival or tumor stages. On the contrary, a good correlation was found between tumor grades and DNA ploidy in the controls. We conclude that renal oncocytoma is a clinically benign tumor, yet it may exhibit varying degrees of flow cytometric DNA abnormalities, which have no predictive value on survival and probably reflect the characteristics of oncocytes rather than its malignant potential.

Aneuploidy in pancreatic insulinomas does not predict malignancy

Nuclear deoxyribonucleic acid (DNA) ploidy studies with paraffin embedded archival material from 14 pancreatic insulinomas were performed by flow cytometry. Clinical follow-up (2 to 17 years; mean, 8 years) was obtained for all patients. Half of the tumors had a normal DNA histogram and half exhibited an abnormal DNA profile consistent with DNA aneuploidy. Six of the seven patients with aneuploid tumors are alive and disease free (2 to 5 years postresection), and one is alive with metastatic disease. Of the seven patients with tumors showing normal DNA profiles, five are alive and disease free, one is dead of disease, and one is alive with metastatic disease. These data suggest that DNA ploidy analysis is unlikely to provide useful prognostic information for patients with insulinomas.

Flow cytometric evaluation of multicystic dysplastic kidneys

The most appropriate management of the multicystic dysplastic kidney remains controversial. At issue is the long-term risk of the development of malignancy in the multicystic dysplastic kidney. The association between renal dysplasia and neoplasia has not been confirmed, with only 6 cases of malignancy reported. Nephroblastomatosis, a probable precursor of Wilms tumor, has been found in 5 to 7% of the cases of multicystic dysplastic kidney when specifically sought. In an attempt to determine whether a relationship exists between renal dysplasia and neoplasia in terms of abnormalities of cellular deoxyribonucleic acid content we performed flow cytometric evaluation on 30 formalin fixed, paraffin embedded archival specimens of multicystic dysplastic kidneys. None of the kidneys had evidence of malignancy. Nuclear deoxyribonucleic acid ploidy studies were performed on single dissociated nuclei prepared by the technique of McLemore and associates and stained with propidium iodide. All specimens demonstrated a diploid pattern of deoxyribonucleic acid, including 3 specimens with nephroblastomatosis or extensive papillary growth, and no specimen demonstrated a tetraploid or aneuploid pattern. The mean G0/G1 fraction was 85.94% (standard deviation 4.59) and the mean S/G2/M fraction was 12.54% (standard deviation 4.72). These findings do not support or negate the potential for neoplasm associated with multicystic dysplastic kidney, since a diploid deoxyribonucleic acid pattern does not eliminate the possibility of the future development of malignancy.

Molecular analysis of the short arm of chromosome 3 in five renal oncocytomas

Renal oncocytomas were tested for loss of alleles at loci on the short arm of chromosome 3, a genetic change characteristic of human renal cell carcinoma. Five renal oncocytomas did not show loss of alleles at loci on 3p supporting the view that renal oncocytoma is a distinct form of renal neoplasia.

Benign metastasizing giant cell tumors of bone. A DNA flow cytometric study

Approximately 2% of histologically benign giant cell tumors (BGCT) of bone are complicated by lung metastases, which can progress despite their benign histologic appearance. Almost all BGCT studied by DNA flow cytometry (FCM) have been reported to be diploid. However, the very few cases with lung metastases previously analyzed were all aneuploid. To assess the usefulness of DNA FCM in predicting the metastatic potential of BGCT, seven metastasizing BGCT were studied by DNA FCM using paraffin-embedded tissue. Five were purely diploid, one was tetraploid, and one was aneuploid. The primary and the metastasis showed the same DNA distribution in all but the tetraploid case, in which the metastasis was purely diploid. A single patient, who was in the diploid group, had unresectable tumor in the lungs; she remains alive with stable disease at 30 months. The other six patients, who underwent complete resections of their lung metastases, are free of disease. These results suggest that DNA FCM is not a sensitive method for predicting the metastatic potential of BGCT since most metastasizing cases appear to be diploid.

Flow cytometry using paraffin-embedded tissue: five years on

The use of paraffin-embedded tissue for flow cytometry is reviewed. A number of technical modifications of the original 1983 method have been described, aimed at improving the accuracy of DNA measurements by minimizing cell debris or reducing coefficients of variation, and at simplifying sample preparation. Over 100 clinical studies have now been reported, mainly assessing the effect of DNA index on prognosis, and those published up until mid-1988 are summarized in an appendix. More recently there have been developments in the use of monoclonal antibodies to measure oncogene products or proliferation markers in addition to DNA content. Detailed clinical evaluation and standardization of these more sophisticated methods is still some way ahead, but as was the case with DNA index, the use of archival material from patients whose outcome is already known should speed this process.

Conservative surgical treatment of renal cell carcinoma: clinical experience and reappraisal of indications

During a 14-year period 36 patients who presented with renal cell carcinoma underwent conservative surgical treatment. The patients were divided into 3 groups according to treatment indications and condition of the contralateral kidney: group 1 included patients with a solitary kidney or bilateral tumors, group 2 patients had a damaged contralateral kidney and group 3 patients were without abnormalities of the contralateral kidney. Cumulative 6-year survival rates were 58 per cent for group 1, and 90 per cent for groups 2 and 3 combined. The over-all cumulative 6-year survival rate was 74 per cent. Based on these data extension of the indication for conservative surgical treatment seems to be justified in patients who present with low stage tumors and partial or potential damage to the contralateral organ.

Further evidence that renal oncocytoma has malignant potential

We report a case of bilateral renal oncocytoma. The right tumor infiltrated the renal capsule focally, and the left tumor penetrated through the capsule into the surrounding perinephric fat and extended into the left renal vein. Karyotyping of the left oncocytoma showed loss of chromosomes 1 and Y (-1,-Y). These karyotypic anomalies, along with the capsular, perinephric fat and renal vein involvement, indicate that renal oncocytomas, despite similar gross and microscopic pathological characteristics, are a heterogeneous group of tumors, some of which have malignant (local invasiveness) potential.

Flow cytometric analysis of ploidy in solid neoplasms: comparison of fresh tissues with formalin-fixed paraffin-embedded specimens

The DNA content of both fresh and formalin-fixed, paraffin-embedded tissues from 30 solid neoplasms was analyzed by flow cytometry. Cells from a single paraffin block of each tumor were disaggregated according to the method of Hedley et al. Cells were sampled from multiple sites of each fresh specimen. Fresh and deparaffinized cells were stored frozen and prepared for cytometric analysis by the technique of Vindelov et al. Thirteen fresh specimens showed DNA diploidy/tetraploidy and 17 showed DNA aneuploidy. Twenty-six of the 30 paraffin-embedded samples were concordant; aneuploid peaks present in histograms from fresh tissue specimens were absent in the 4 paraffin-embedded specimens that were discordant. For 2 of the discordant cases, a hyperdiploid peak that was absent in the initial paraffin-embedded samples was evident on analysis of cells from additional paraffin blocks, indicating ploidy heterogeneity. Peridiploid aneuploid peaks present in the other 2 fresh samples were not observed in initial or additional deparaffinized specimens. Although fresh tissues are preferred for ploidy examination by flow cytometry, formalin-fixed paraffin-embedded specimens, despite a few inherent pitfalls, are generally useful for retrospective studies that compare the ploidy status of solid neoplasms to therapeutic response and prognosis.

Uterine leiomyosarcomas and benign smooth muscle tumors: usefulness of nuclear DNA patterns studied by flow cytometry

Flow cytometry was used to determine the DNA ploidy pattern of paraffin-embedded archival tissue specimens from 90 surgically resected uterine smooth muscle tumors (49 leiomyosarcomas and 41 leiomyomas). The technique of Hedley was used for preparation of paraffin-embedded tissue into single dissociated nuclei, and the method of Vindeløv was used for staining with propidium iodide. Among the 41 leiomyomas, most tumors (88%) had a DNA diploid pattern; the exceptions were two DNA tetraploid/polyploid and three DNA aneuploid samples. The DNA histograms of the 49 leiomyosarcomas (including 6 epithelioid leiomyosarcomas) were classified as follows: 9 cases (18%) exhibited a DNA diploid pattern, 29 cases (59%) had a DNA tetraploid/polyploid pattern, and 11 cases (23%) had DNA aneuploid peaks. Although DNA ploidy pattern cannot be used diagnostically to distinguish malignant from benign uterine smooth muscle tumors, the nuclear DNA ploidy pattern is an easily measured, objective determination that may have important prognostic significance for patients with uterine leiomyosarcomas.

Primary squamous cell carcinoma of the male urethra: nuclear deoxyribonucleic acid ploidy studied by flow cytometry

Flow cytometry analysis was performed on 30 primary male urethral squamous cell carcinoma specimens. Nuclei were extracted from paraffin-embedded archival material and isolated nuclei were stained with propidium iodide. Bulbomembranous urethral tumors had a higher incidence of abnormal deoxyribonucleic acid ploidy patterns than penile urethral tumors (69 and 29 per cent, respectively). Of the tumors exhibiting a deoxyribonucleic acid diploid pattern and an abnormal (deoxyribonucleic acid tetraploid or aneuploid) histogram 18 and 93 per cent, respectively, showed tumor progression (p less than 0.001). None (0 per cent) of the low grade (grade 1 or 2) tumors with a deoxyribonucleic acid diploid pattern developed local recurrence or distant metastases, whereas 90 per cent of the low grade tumors with an abnormal deoxyribonucleic acid pattern progressed (p less than 0.002). Patients with tumors exhibiting deoxyribonucleic acid diploid ploidy had 5 and 10-year rates free of disease of 85 per cent. In contrast, patients with tumors with abnormal deoxyribonucleic acid ploidy patterns had 5 and 10-year rates of 20 and 0 per cent, respectively (p less than 0.001). Determination of deoxyribonucleic acid ploidy pattern by flow cytometry provides important prognostic information for male patients with primary squamous cell carcinoma of the urethra.

Leiomyosarcomas and benign smooth muscle tumors of the stomach: nuclear DNA patterns studied by flow cytometry

Paraffin-embedded archival tissue samples were used for determination of DNA ploidy by flow cytometry on 117 surgically resected gastric smooth muscle tumors (44 leiomyosarcomas, 53 leiomyomas, and 20 benign leiomyoblastomas). The technique of Hedley was used for preparation of paraffin-embedded tissue into single dissociated nuclei, and the method of Vindeløv was used for staining with propidium iodide. Among the 53 leiomyomas, the DNA ploidy pattern was diploid in most tumors (87%), except for 2 DNA tetraploid/polyploid and 5 DNA aneuploid samples. In comparison, the 20 benign leiomyoblastomas had more frequent abnormal DNA histograms: DNA tetraploidy/polyploidy in 5 (25%) and DNA aneuploidy in 2 (10%). The DNA histograms of the 44 leiomyosarcomas (including 4 epithelioid leiomyosarcomas) were classified as follows: 20 cases (45%) exhibited a DNA diploid pattern, 14 cases (32%) had a DNA tetraploid/polyploid pattern, and 10 cases (23%) had DNA aneuploid peaks. For the patients with leiomyosarcomas, the DNA ploidy pattern was significantly correlated with survival (P less than 0.001), as were tumor grade (P less than 0.001) and tumor size (P less than 0.05). Furthermore, both benign and malignant gastric smooth muscle tumors with DNA tetraploid/polyploid patterns were significantly larger than those with a DNA diploid histogram (P less than 0.05). DNA ploidy pattern cannot be used for diagnosis--that is, to distinguish malignant from benign gastric smooth muscle tumors. For gastric leiomyosarcomas, however, nuclear DNA ploidy pattern is an easily measured objective determination with important prognostic significance.

Well differentiated clear cell renal carcinoma: significance of nuclear deoxyribonucleic acid patterns studied by flow cytometry

Nuclear deoxyribonucleic acid ploidy was determined in 206 samples of well differentiated clear cell renal carcinoma via a previously described technique on paraffin-embedded archival material. Grade 2 tumors had a higher incidence of abnormal deoxyribonucleic acid patterns than grade 1 tumors. Of stage 1 well differentiated clear cell renal tumors 60 per cent had a normal deoxyribonucleic acid histogram pattern, whereas 67 per cent of higher stage tumors had an abnormal pattern. The existence of abnormal nuclear deoxyribonucleic acid histogram patterns in the primary tumor tissue had a significant correlation with subsequent development of metastatic disease, independent of tumor grade and stage. The 10-year actuarial survival rate was 62 per cent for patients with normal deoxyribonucleic acid histograms and 37 per cent for patients with abnormal patterns. These results demonstrate that nuclear deoxyribonucleic acid ploidy measured by flow cytometry is an important variable in the classification and determination of prognosis for patients with clear cell renal carcinoma.