Estrogen Receptor Expression in DICER1-related Lesions is Associated With the Presence of Cystic Components

DICER1 tumor predisposition syndrome results from pathogenic variants in DICER1 and is associated with a variety of benign and malignant lesions, typically involving kidney, lung, and female reproductive system. Over 70% of sarcomas in DICER1 tumor predisposition syndrome occur in females. Notably, pediatric cystic nephroma (pCN), a classic DICER1 tumor predisposition syndrome lesion, shows estrogen receptor (ER) expression in stromal cells. There are also renal, hepatic, and pancreatic lesions unassociated with DICER1 tumor predisposition syndrome that have an adult female predominance and are characterized/defined by ER-positive stromal cells. Except for pCN, the expression of ER in DICER1-associated lesions remains uninvestigated. In the present study, ER expression was assessed by immunohistochemistry in 89 cases of DICER1-related lesions and 44 lesions lacking DICER1 pathogenic variants. Expression was seen in stromal cells in pCN and pleuropulmonary blastoma (PPB) types I and Ir, whereas anaplastic sarcoma of kidney and PPB types II and III were typically negative, as were other solid tumors of non-Müllerian origin. ER expression was unrelated to the sex or age of the patient. Expression of ER showed an inverse relationship to preferentially expressed antigen in melanoma (PRAME) expression; as lesions progressed from cystic to solid (pCN/anaplastic sarcoma of kidney, and PPB types I to III), ER expression was lost and (PRAME) expression increased. Thus, in DICER1 tumor predisposition syndrome, there is no evidence that non-Müllerian tumors are hormonally driven and antiestrogen therapy is not predicted to be beneficial. Lesions not associated with DICER1 pathogenic variants also showed ER-positive stromal cells, including cystic pulmonary airway malformations, cystic renal dysplasia, and simple renal cysts in adult kidneys. ER expression in stromal cells is not a feature of DICER1 perturbation but rather is related to the presence of cystic components.

PRAME protein expression in DICER1-related tumours

DICER1 syndrome is an autosomal dominant tumour predisposition syndrome usually affecting persons under 30 years of age. Many of the associated benign and malignant lesions occur almost exclusively in DICER1 syndrome. One such tumour, pituitary blastoma (pitB), overexpresses PRAME 500x above control levels. PRAME (PReferentially expressed Antigen in MElanoma) is expressed in malignancies that are not DICER1-related (e.g. melanoma). To address whether PRAME expression is part of the DICER1 phenotype, or simply a feature of pitB, a series of 75 DICER1-mutated specimens and 33 non-mutated specimens was surveyed using immunohistochemistry for PRAME, together with EZH2, which complexes with PRAME. In DICER1-mutated specimens, positive staining for PRAME was only seen in malignant tumours; 7 of 11 histological types and 34/62 individual tumours were positive, while non-tumourous lesions were always negative. Pleuropulmonary blastoma (PPB) showed a continuum in staining, with type I lesions being PRAME negative (n = 7) but all type II and type III lesions PRAME positive (n = 7). Similarly, cystic nephroma (CN) was negative (n = 8), with anaplastic sarcoma of the kidney being positive (n = 2). However, one atypical CN with mesenchymal cell proliferation was PRAME-positive. Embryonal rhabdomyosarcoma (RMS) with DICER1 pathogenic variants (PVs) was positive for PRAME (5/6), but the same tumour type without DICER1 PVs was also positive (9/15). Staining for EZH2 corresponded to that seen with PRAME, validating the latter. This study leads us to conclude that (1) PRAME expression occurs in two-thirds of DICER1-related malignancies; (2) PRAME may be a marker for the progression that certain DICER1-related lesions are thought to undergo, such as PPB and CN; and (3) PRAME expression in some tumours, such as RMS, appears to be an intrinsic feature of the tumour, rather than specifically related to DICER1 PVs. Therapy directed against PRAME may offer novel treatment options in patients with the DICER1 syndrome.

Targeted RNA sequencing in the routine clinical detection of fusion genes in salivary gland tumors

Salivary gland tumors represent a diverse group of neoplasms that occasionally pose a diagnostic challenge for pathologists, particularly with limited sampling. Gene fusions, which may reflect genetic drivers, are increasingly recognized in a subset of these neoplasms, and can be leveraged for diagnostic purposes. We performed a retrospective analysis on a cohort of 80 benign and malignant salivary gland tumors, enriched for subtypes known to harbor recurrent fusion events, to validate the diagnostic use of a targeted RNA sequencing assay to detect fusion transcripts. Testing identified fusion genes in 71% (24/34) of pleomorphic adenoma and carcinoma-ex-pleomorphic adenoma, with 56% of cases showing rearrangement of PLAG1 and 15% HMGA2. In addition to confirming known partners for these genes, novel PLAG1 fusion partners were identified, including DSTN, NTF3, and MEG3; CNOT2 was identified as a novel fusion partner for HMGA2. In adenoid cystic carcinoma, 95% of cases (19/20) were positive for a fusion event. MYB was rearranged in 60% (12/20), MYBL1 in 30% (6/20), and NFIB in 5% (1/20); two tumors exhibited novel fusion products, including NFIB-TBPL1 and MYBL1-VCPIP1. Fusion genes were identified in 64% (9/14) of cases of mucoepidermoid carcinoma; MAML2 was confirmed to partner with either CRTC1 (43%) or CRTC3 (21%). One salivary duct carcinoma was found to harbor a novel RAPGEF6-ACSL6 fusion gene. Finally, as anticipated, gene fusions were not detected in any of the five acinic cell carcinomas included in the cohort. In summary, targeted RNA sequencing represents a diagnostically useful ancillary technique for identifying a variety of existing, and novel, fusion transcripts in the classification of salivary gland neoplasms.

Pediatric fibromyxoid soft tissue tumor with PLAG1 fusion: A novel entity?

The classification of undifferentiated soft tissue tumors continues to evolve with the expanded application of molecular analysis in clinical practice. We report three cases of a unique soft tissue tumor in young children (5 months to 2 years old) displaying a purely fibromyxoid histology, with positive staining for desmin and CD34. In two cases, RNA sequencing detected a YWHAZ-PLAG1 gene fusion, while in the third case, a previously unreported EEF1A1-PLAG1 fusion was identified. PLAG1 fusions have been reported in several pathologic entities including pleomorphic adenoma, myoepithelial tumors of skin and soft tissue, and lipoblastoma, the latter occurring preferentially in young children. In these tumors, expression of a full length PLAG1 protein comes under the control of the constitutively active promoter of the partner gene in the fusion, and the current cases conform to that model. Overexpression of PLAG1 was confirmed by diffusely positive immunostaining for PLAG1 in all three cases. Our findings raise the possibility of a novel fibromyxoid neoplasm in childhood associated with these rare PLAG1 fusion variants. The only other report of a PLAG1-YWHAZ fusion occurred in a pediatric tumor diagnosed as a "fibroblastic lipoblastoma." This finding raises the possibility of a relationship with our three cases, even though our cases lacked any fat component. Further studies with regard to a shared pathogenesis are required.

Abstract A01: Novel childhood genitourinary manifestations of DICER1 syndrome

Pathogenic germline variants in DICER1, a gene encoding an RNase key in microRNA-mediated silencing, cause DICER1 syndrome. This genetic disorder predisposes to the development of a wide array of mainly childhood-onset conditions including genitourinary tumors such as cystic nephroma (CN), anaplastic sarcoma of the kidney, Wilms’ tumor (WT), Sertoli-Leydig cell tumor (SLCT), and cervical embryonal rhabdomyosarcoma (ERMS). As the spectrum of clinical manifestations is not yet fully defined, we sought to explore the involvement of DICER1 mutations in pediatric genitourinary lesions not previously well studied. A series of 31 formalin-fixed, paraffin-embedded tumor samples including 15 paratesticular ERMS (ptERMS), one unclassifiable ovarian sex cord-stromal tumor, one fallopian tube ERMS (ftERMS), six cystic partially differentiated nephroblastomas (CPDN), two cystic WT, two CN, and four rare renal lesions were collected. Median age at diagnosis was 4 years. Tumor DNA was screened for DICER1 variants using Sanger sequencing or a Fluidigm array. Sanger sequencing on tumor DNA was used to validate the variants, and on normal DNA, to determine the germline status. Seven samples harbored biallelic DICER1 mutations and in 6/7 cases, the loss-of-function variant was confirmed to be of germline origin. The paratesticular tumors were initially diagnosed as ERMS, but pathology review reclassified one as an ectomesenchymoma and another as an undifferentiated low-grade myxoid sarcoma. No DICER1 mutation was identified in any of the ptERMS, but the myxoid sarcoma had DICER1 mutations. Interestingly, the patient with the paratesticular myxoid sarcoma also developed a CN. We identified DICER1 mutations in a tumor originally classified as an ovarian WT. Subsequent pathologic evaluation of the lesion led to its reclassification as a retiform SLCT with rhabdomyosarcomatous elements. A typical hotspot and a potential splicing mutation were detected in the ftERMS. Also, two atypical cystic kidney lesions harbored DICER1 mutations: one composed of blastema-type cells expressing nuclear WT1 diagnosed as a CPDN and another unusual multicystic renal lesion considered to be a CN that had moved into a proliferative phase but lacked blastema and anaplastic foci. The patient with the latter renal lesion had previously developed a unilateral CN with a distinct hotspot mutation. As expected, the two classical CN had DICER1 mutations. Based on these results, patients diagnosed with rare genitourinary tumors (except for ptERMS), may have DICER1 syndrome, especially if these individuals have a personal or family history of DICER1-associated lesions. In these cases, surveillance for DICER1-related conditions is advised given that the risk of tumor development is highest in early childhood. In summary, specialist pathology review and DICER1 testing in this setting will lead to improved diagnosis with potential implications for clinical care.

Citation Format: Maria V. Apellaniz-Ruiz, Catherine Goudie, Noelle Cullinan, Elvis T. Valera, Krisztina Z. Hanley, Luiz G. Tone, Paula Marrano, Ronald Grant, W. Glenn McCluggage, Gordan M. Vujanic, Paul S. Thorner, William D. Foulkes. Novel childhood genitourinary manifestations of DICER1 syndrome [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr A01.

DICER1 screening in 15 paediatric paratesticular sarcomas unveils an unusual DICER1-associated sarcoma

Individuals with DICER1 syndrome, a genetic disorder caused by pathogenic germline variants in DICER1, are at increased risk of developing a wide array of predominantly childhood onset conditions, including genitourinary sarcomas. However, data on DICER1 involvement in paratesticular sarcomas have not been published. Herein, we analyse a series of 15 paediatric paratesticular sarcomas and describe in detail the case of a male infant with a paratesticular myxoid tumour, considered to be a low‐grade sarcoma, who also manifested a cystic nephroma, a classic DICER1 syndrome phenotype. He harboured a pathogenic germline DICER1 variant and different somatic hot‐spot mutations in each tumour. The paratesticular tumour showed strong and diffuse expression for WT1 and CD10, an unusual immunophenotype in paediatric sarcomas, but typical of tumours of Müllerian origin. The tumour was postulated to arise from the appendix testis, a Müllerian remnant located in the paratestis. Such an origin would be analogous to other DICER1‐associated non‐epithelial gynaecological tumours, thought to arise from Müllerian derivatives. These findings point towards a key role of DICER1 in Müllerian‐derived structures. Supporting this hypothesis is the fact that the other paratesticular sarcomas from the series were either negative or focally positive for WT1 and for CD10, and none had any DICER1 mutations. In summary, we present the first case of a paratesticular sarcoma associated with DICER1 syndrome, emphasising that paratesticular tumours with an unusual histological appearance may suggest an underlying DICER1 mutation, especially in the presence of a personal or family history of DICER1‐associated disease. In this context, DICER1 mutation testing could lead to changes in clinical care including implementation of cancer care surveillance strategies.

Genetic diversity in alveolar soft part sarcoma: A subset contain variant fusion genes, highlighting broader molecular kinship with other MiT family tumors

Alveolar soft part sarcoma (ASPS) is a rare malignancy that, since its initial description, remains a neoplasm of uncertain histogenesis. The disease-defining molecular event characterizing the diagnosis of ASPS is the ASPSCR1-TFE3 fusion gene. Following identification of an index case of ASPS with a novel TFE3 fusion partner, we performed a retrospective review to determine whether this represents an isolated event. We identified two additional cases, for a total of three cases lacking ASPSCR1 partners. The average patient age was 46 years (range, 17-65); two patients were female. The sites of origin included the transverse colon, foot, and dura. Each case exhibited a histomorphology typical of ASPS, and immunohistochemistry was positive for TFE3 in all cases. Routine molecular testing of the index patient demonstrated a HNRNPH3-TFE3 gene fusion; the remaining cases were found to have DVL2-TFE3 or PRCC-TFE3 fusion products. The latter two fusions have previously been identified in renal cell carcinoma; to our knowledge, this is the first report of a HNRNPH3-TFE3 gene fusion. These findings highlight a heretofore underrecognized genetic diversity in ASPS, which appears to more broadly molecularly overlap with that of translocation-associated renal cell carcinoma and PEComa. These results have immediate implications in the diagnosis of ASPS since assays reliant upon ASPSCR1 may yield a false negative result. While these findings further understanding of the molecular pathogenesis of ASPS, issues related to the histogenesis of this unusual neoplasm remain unresolved.

Sarcoma Subgrouping by Detection of Fusion Transcripts Using NanoString nCounter Technology

BACKGROUND

NanoString technology is an innovative barcode-based system that requires less tissue than traditional techniques and can test for multiple fusion transcripts in a single reaction. The objective of this study was to determine the utility of NanoString technology in the detection of sarcoma-specific fusion transcripts in pediatric sarcomas.

DESIGN

Probe pairs for the most common pediatric sarcoma fusion transcripts were designed for the assay. The NanoString assay was used to test 22 specific fusion transcripts in 45 sarcoma samples that had exhibited one of these fusion genes previously by reverse transcription polymerase chain reaction (RT-PCR). A mixture of frozen (n = 18), formalin-fixed, paraffin-embedded (FFPE) tissue (n = 23), and rapid extract template (n = 4) were used for testing.

RESULTS

Each of the 22 transcripts tested was detected in at least one of the 45 tumor samples. The results of the NanoString assay were 100% concordant with the previous RT-PCR results for the tumor samples, and the technique was successful using both FFPE and rapid extract method.

CONCLUSION

Multiplexed interrogation for sarcoma-specific fusion transcripts using NanoString technology is a reliable approach for molecular diagnosis of pediatric sarcomas and works well with FFPE tissues. Future work will involve validating additional sarcoma fusion transcripts as well as determining the optimal workflow for diagnostic purposes.

Fusion of ALK to the melanophilin gene MLPH in pediatric Spitz nevi

Spitzoid neoplasms typically affect young individuals and include Spitz nevus, atypical Spitz tumor, and Spitzoid melanoma. Spitz tumors can exhibit gene fusions involving the receptor tyrosine kinases NTRK1, NTRK3, ALK, ROS1, RET, or MET, or the serine-threonine kinase BRAF. Because most studies have been based on adult cases, we studied ALK fusions in Spitz nevi occurring in pediatric patients. Twenty-seven cases were screened for ALK expression by immunohistochemistry, and 6 positive cases were identified. These cases were studied further using the TruSight RNA Fusion Panel, and in 4 cases, exon 20 of the ALK gene was found to be fused to exon 14 of the MLPH (melanophilin) gene, a gene fusion that has only been reported in a Spitz nevus in an adult. The remaining 2 cases showed no fusion of ALK with any gene. The cases with the MLPH-ALK fusion showed a similar histology to that described for Spitz nevi with ALK fusions, with spindle-shaped and epithelioid melanocytes in fusiform nests with a plexiform growth pattern and infiltrative border. We created a breakapart fluorescence in situ hybridization assay for MLPH, and all 4 cases with the MLPH-ALK fusion were positive, whereas the other 23 cases in the study were negative. Thus, ALK and MLPH were fused only to each other in our series. Melanophilin is part of the melanosome trafficking apparatus together with MYO5a, TPM3, and RAB27a, all constitutively expressed in melanocytes. Kinase fusions involving MYO5A and TPM3 have been reported in Spitz tumors, and our series adds MLPH to this group.

Heterogeneity of MYCN amplification in neuroblastoma at diagnosis, treatment, relapse, and metastasis

Amplification of the MYCN gene in neuroblastoma is associated with a poor prognosis and is considered to remain unchanged in post-treatment specimens and metastases. While heterogeneity of MYCN copy number in tumor cells has been reported, serial samples have only been studied in a limited way, and the biologic relevance of this finding is not well understood. We used in situ hybridization on paraffin sections of 102 specimens from 30 patients with MYCN-amplified neuroblastoma to determine MYCN copy number in the primary tumor, pre- and post-treatment, and in metastatic samples. Nineteen cases (63%) showed diffuse MYCN amplification in all samples tested. Nine cases (30%) showed a reduction in MYCN copy number: five cases with diffuse amplification subsequently showed focal amplification, one case with diffuse MYCN amplification showed MYCN gain after treatment, and three focally amplified cases were non-amplified in later specimens. In two cases (7%), focal amplification became diffuse in subsequent samples. Histology was not predictive of the temporal or spatial pattern of MYCN amplification for a particular tumor. If extent of amplification (focal vs. diffuse) is not considered, 26/30 (87%) of cases were consistently MYCN-amplified. However, our data suggest that MYCN status can be heterogeneous between tumor sites, during tumor progression or following treatment, challenging the notion that MYCN copy number does not change for a particular neuroblastoma. Assessing the biologic significance of MYCN heterogeneity will require larger studies of clinically annotated tumor samples, and will depend on interpreting heterogeneity in MYCN status in combination with other genetic changes. © 2016 Wiley Periodicals, Inc.

TFE3-positive renal cell carcinomas are not always Xp11 translocation carcinomas: Report of a case with a TPM3-ALK translocation

Translocation-associated renal cell carcinoma (RCC) is a distinct subtype of RCC with gene rearrangements of the TFE3 or TFEB loci. The TFE3 gene is located at Xp11 and can fuse to a number of translocation partners, resulting in high nuclear expression of TFE3 protein. TFE3 immunostaining is often used as a surrogate marker for a TFE3 translocation. We report a case of an RCC that expressed TFE3 but showed only gain of TFE3 rather than a translocation. Moreover, this case had a t(1;2) translocation fusing ALK and TMP3, identical to that seen in inflammatory myofibroblastic tumour. There was resulting overexpression of ALK protein in a cytoplasmic and membranous pattern. The patient was not treated with chemotherapy but following regional nodal recurrence, an ALK inhibitor was added and the patient remains alive one year later. There are only rare reports of RCC with an ALK-TMP3 fusion, and these tumours can express TFE3 on some unknown basis not related to a TFE3 translocation. Any RCC positive for TFE3 and lacking a translocation should be tested for ALK expression and translocation. Recognition of this subtype of RCC will allow ALK inhibitor therapy to be added, in the hope of improving patient outcome.

Abstract 5333: TH-302 potentiates the antitumor activity of topotecan in neuroblastoma and rhabdomyosarcoma preclinical models

Background: There is strong evidence that tumor cells in the hypoxic regions cause drug resistance and tumor relapse. Hypoxia targeting agents may make cytotoxic therapy more effective. In this study, we investigated the efficacy of TH-302, a hypoxia-activated prodrug of bromo-isophosphoramide mustard, both alone and in combination with the cytotoxic agent topotecan, in preclinical models of neuroblastoma (NBL) and rhabdomyosarcoma (RMS).

Materials and methods: A panel of four NBL and three RMS cell lines was tested in vitro under normoxic or hypoxic conditions to assess the effect of TH-302 on cell proliferation, both as a single agent and in combination with topotecan. In vivo activity was also evaluated in different xenograft models. TH-302 (50 mg/kg) and topotecan (1 mg/kg) were administered daily for 2 consecutive weeks, as single agents or in combination. Animal survival was studied with an NBL (SK-N-BE(2)) metastatic tumor model. Cleaved caspase-3 and the hypoxia staining agent pimonidazole were used as apoptotic and hypoxic markers in vivo.

Results: Under normoxic conditions, all tested lines showed an anti-proliferative response to TH-302 with the IC50s ranging from 4.6 to 52μM. When exposed to hypoxia (1% O2) overnight, a dramatic decrease of IC50 was observed, ranging from 0.07 to 2.4μM. Adding topotecan to the TH-302 treatment significantly increased cytotoxicity in all tested tumor cell lines. In both NBL models (SK-N-BE(2) and CHLA-20), single-agent TH-302 treatment delayed tumor growth. Complete tumor regression was observed in the combined treatment group after a two-week treatment period, but all the regressed tumors relapsed after treatment ended. In the CHLA-20 xenograft model, single-agent topotecan also induced complete tumor regression. However, median time to relapse in the combination group was significantly longer than in the topotecan group (43 days vs. 16 days; p<0.01). In the combination group, all relapsed tumors remained responsive to the drug treatment when re-challenged with the original treatment regimen. In the neuroblastoma metastatic model, all three regimens prolonged animal survival, but the combined treatment group showed better survival than single-agent treatments (p<0.01). In a RMS xenograft model, combined treatment was also more effective than single-agent TH-302 or topotecan. Double immunostaining of apoptotic and hypoxic markers showed that TH-302 mostly targeted tumor cells residing at hypoxic regions while topotecan was more effective on tumor cells outside the hypoxic zones. The combined treatment induced tumor cell apoptosis in both normoxic and hypoxic regions.

Conclusions: TH-302 showed antitumor effects in neuroblastoma and rhabdomyosarcoma xenograft models. Compared to single-agent TH-302 or topotecan, the combination treatment significantly improved tumor response, delayed tumor relapse, and enhanced animal survival.

Citation Format: Libo Zhang, Bing Wu, Paula Marrano, Paul Thorner, Sylvain Baruchel. TH-302 potentiates the antitumor activity of topotecan in neuroblastoma and rhabdomyosarcoma preclinical models. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 5333. doi:10.1158/1538-7445.AM2015-5333

BRAF mutations in pediatric metanephric tumors

Metanephric neoplasms of the kidney are uncommon, and some cases are associated with papillary carcinoma. Most cases of metanephric adenoma occur in adults, with fewer than 25 cases reported in children, and metanephric adenofibroma is even less common. The few metanephric tumors studied at the genetic level have not shown the gains of chromosomes 7 and 17 commonly seen in renal cell carcinoma, suggesting that the carcinoma arising in this setting has a separate genetic origin from the adenoma. However, the assumption that this carcinoma has the same chromosome gains as sporadic renal cell carcinoma has never been validated. We studied 4 cases of metanephric tumors in children, including 1 metanephric adenofibroma with papillary carcinoma. The composite tumor was studied by single nucleotide polymorphism array and fluorescence in situ hybridization, with the adenoma and carcinoma components analyzed separately. No copy number alterations were detected in either component. A BRAF V600E mutation has been reported in most cases of metanephric adenoma in adults. We performed BRAF V600E immunostaining and sequencing in our 4 pediatric cases. Three cases had a BRAF V600E mutation including the composite tumor, with both the adenoma and carcinoma components showing the same mutation. This finding provides the first genetic evidence that these 2 tumors are biologically linked. Ten cases each of pediatric renal cell carcinoma and Wilms tumor were immunonegative. Thus, BRAF V600E immunostaining is a helpful marker for pediatric metanephric adenoma, and additional research is required on the possible role of this mutation in the development of renal carcinoma.

The neuroblastoma and ganglion components of nodular ganglioneuroblastoma are genetically similar: evidence against separate clonal origins

Nodular ganglioneuroblastoma is characterized by a macroscopic nodule of neuroblastoma within a ganglioneuromatous component. These two components have been considered to originate from separate clones, with the neuroblastoma clone accounting for the clinical behavior of nodular ganglioneuroblastoma. In order to investigate the clonal origin of the cellular components (neuroblasts, ganglion cells, and Schwann cells) of nodular ganglioneuroblastoma, paraffin-embedded tumor samples from eight cases were analyzed by single nucleotide polymorphism array and in situ hybridization. DNA was extracted separately from neuroblastomatous and ganglioneuromatous areas. By in situ hybridization, MYCN gain (4-10 gene copies/nucleus) was detected in 7/8 neuroblastoma samples. In ganglioneuromatous regions, gains were also detected in ganglion cells but not in Schwann cells. Single-nucleotide polymorphism array studies identified chromosome losses (11q and 14q) and gains (12, 13q, 17q and 18q) in the neuroblastoma component, whereas the ganglioneuromatous component showed fewer or no genetic alterations. There were no unique copy number changes distinguishing nodular ganglioneuroblastoma from other subtypes of neuroblastoma. By in situ hybridization, ganglion cells but not Schwann cells showed the same alterations detected in neuroblasts. Thus, neuroblasts and ganglion cells in nodular ganglioneuroblastoma are genetically related and may arise from the same clone. In contrast, the Schwann cells have a different origin and may be derived from a non-neoplastic neural crest precursor. Our results suggest that the clinical behavior of nodular ganglioneuroblastoma cannot be explained by the presence of separate clones with distinct genetic signatures.

Nab-paclitaxel is an active drug in preclinical model of pediatric solid tumors

PURPOSE

To investigate the antitumor effect of nab-paclitaxel, an albumin-stabilized nanoparticle formulation of paclitaxel, on pediatric solid tumor models.

EXPERIMENTAL DESIGN

A panel of three rhabdomyosarcoma, one osteosarcoma and seven neuroblastoma cell lines were exposed to increasing concentrations of nab-paclitaxel in vitro. Cell viability was evaluated using the Alamar Blue Assay. Antitumor effect was further assessed in vivo in NOD/SCID xenograft and metastatic neuroblastoma mouse models. Tumor sections were analyzed by immunohistochemistry for cleaved caspase-3 and phospho-histone H3. Plasma and intratumoral paclitaxel concentrations were measured by liquid chromatography-mass spectrometry. Ratio of intratumoral and plasma concentration was compared between nab-paclitaxel and paclitaxel treatment groups.

RESULTS

Nab-paclitaxel displayed significant cytotoxicity against most pediatric solid tumor cell lines in vitro in a dose-dependent manner. In vivo, nab-paclitaxel showed antitumor activity in both rhabdomyosarcoma (RH4 and RD) and neuroblastoma [SK-N-BE(2) and CHLA-20] xenograft models. In the SK-N-BE(2) metastatic model, nab-paclitaxel treatment significantly extended animal survival compared with control (P < 0.01). Nab-paclitaxel treatment induced tumor cell-cycle arrest and apoptosis in vivo. In the RH4 model, increased local relapse-free intervals were observed with nab-paclitaxel treatment (37.7 ± 3.2 days) comparing with paclitaxel (13.6 ± 2.07 days). Local relapsed tumors following paclitaxel treatment proved to be paclitaxel-resistant and remained responsive to nab-paclitaxel. Mechanistically, a higher tumor/plasma paclitaxel drug ratio in favor of nab-paclitaxel was observed.

CONCLUSIONS

Nab-paclitaxel showed significant antitumor activity against all pediatric solid tumors associated with an enhanced drug intratumor delivery. Furthermore, testing of nab-paclitaxel in pediatric solid-tumor patient population is under development.

Natural course of low risk neuroblastoma

BACKGROUND

Neuroblastoma is characterized by heterogeneity of histology, biology, and clinical behavior. Most epidemiology studies are based on Western and Japanese populations; there are very few studies on neuroblastoma from Southeast Asia.

PROCEDURE

Cases of Thai children with neuroblastoma were retrospectively reviewed to determine if the epidemiology of the disease differs from Western populations. Sixty-two cases were assembled from two pathology centers in Bangkok. Histologic prognostic category and MYCN copy number were determined.

RESULTS

The median age at diagnosis was 2.9 years. Only 11% of cases presented at less than 1 year of age and 12% cases had low stage disease (1, 2, and 4S). The majority of tumors had unfavorable histology (48/62); this was at least partly due to the higher age at diagnosis for most patients. MYCN amplification was detected in 18/52 (35%) tumors, all in stage 3 or 4 tumors. We assigned patients to high, intermediate and low risk categories using the Children's Oncology Group risk stratification criteria. In contrast to Western studies, the majority of cases (50/59 or 85%) in our series had high risk disease.

CONCLUSIONS

Since there is no evidence to date that the biology of neuroblastoma varies by geographic region, the paucity of low risk cases in our study may reflect spontaneous resolution/differentiation of tumors that are not clinically detected. Moreover, a delay in diagnosis of intermediate risk cases could result in higher tumor burden at the time of diagnosis, increasing the proportion of high risk cases observed.

Primary intracranial Ewing sarcoma with an unusually aggressive course: a case report and review of the literature

The occurrence of Ewing sarcoma-peripheral primitive neuroectodermal tumor as a primary intracranial tumor is very rare, with only 29 cases reported in the literature, 19 of which have included molecular studies. We present the clinical, radiologic and pathologic findings of an intracranial Ewing sarcoma in a 22-year-old woman arising from the dura over the right frontal convexity. The patient underwent craniotomy with gross total excision of the tumor. The tumor showed atypical histology and the diagnosis was confirmed by detection of a rearrangement of the EWSR1 gene by fluorescent in situ hybridization and identification of the diagnostic t(11;22)(q24;q12) translocation by reverse transcription-polymerase chain reaction. Additional features were detected in this tumor that are known to be associated with an unfavorable prognosis, including loss of p16 expression and gains of chromosomes 1q and 12. The patient experienced the most rapid downhill course reported to date for intracranial Ewing sarcoma, developing multiple extracranial metastases at 2 months and dying 6 months after the initial operation.

Topographical analysis of telomere length and correlation with genomic instability in whole mount prostatectomies

BACKGROUND

Many critical events in prostatic carcinogenesis appear to relate to the emergence of genomic instability. Characteristic genomic abnormalities such as 8p loss, 8q gain, trisomy 7, and PTEN microdeletions may provide selective advantages to increase neoplastic transformation. Evidence suggests that telomere dysfunction is a plausible mechanism for some of these abnormalities on the basis of the break-fusion-bridge cycle that can lead to manifestations of genomic instability.

METHODS

In this study, we correlate telomere length measured by quantitative FISH in various prostatic histologies with markers of genomic instability and immunohistochemical measures of proliferation and oxidative stress.

RESULTS

We find that telomere shortening is correlated with abnormalities on chromosome 8, but not with trisomy 7 or abnormalities of the PTEN locus. There are associations with C-MYC aberrations in stroma with greater proximity to cancer and a correlation between telomere length in a number of prostatic histologies and the adjacent stroma, suggesting the importance of microenvironmental effects on telomere maintenance in the prostate. This finding was also supported by the finding of the correlation between telomere attrition and the levels of oxidative stress as measured by malondialdehyde staining in HPIN lesions close to cancer.

CONCLUSIONS

Telomere attrition in the prostate gland is associated with particular genomic aberrations that contribute to the genomic instability characteristic of prostatic carcinogenesis. Correlations between various histologies and adjacent stroma telomere length suggest it is also may reveal microenvironmental effects within the prostate gland. Oxidative stress may contribute to telomere attrition in HPIN close to cancer.

Genomic instability and copy-number heterogeneity of chromosome 19q, including the kallikrein locus, in ovarian carcinomas

Many tissue kallikrein (KLK) genes and proteins are candidate diagnostic, prognostic and predictive biomarkers for ovarian cancer (OCa). We previously demonstrated that the KLK locus (19q13.3/4) is subject to copy-number gains and structural rearrangements in a pilot study of cell lines and ovarian cancer primary tissues, shown to overexpress KLK gene family members. To determine the overall frequency of genomic instability and copy-number changes, a retrospective study was conducted using formalin-fixed paraffin embedded (FFPE) tissues. Eighty-one chemotherapy naïve serous OCas were examined using 3-colour fluorescence in situ hybridization (FISH) to identify structural and numerical changes on 19q, including the KLK locus; in addition to immunohistochemistry (IHC) for KLK6, which has been shown to be overexpressed in OCa. The KLK locus was subject to copy-number changes in ∼83% of cases: net gain in 51%, net loss in 30% and amplified in 2%; and found to be chromosomally unstable (p < 0.001). All cases showed a wide range of immuoreactivity for KLK6 by IHC. Although no strong correlation could be found with copy-number, the latter was contributing factor to the observed KLK6 protein overexpression. Moreover, univariate and multivariate analyses showed an association between the net loss of the KLK locus and longer disease-free survival. Interestingly, FISH analyses indicated that chromosome 19q was subjected to structural rearrangement in 62% of cases and was significantly correlated to tumor grade (p < 0.001). We conclude that numerical and structural aberrations of chromosome 19q, affect genes including the KLK gene members, may contribute to ovarian carcinoma progression and aggressiveness.

Abstract 12: Metastatic neuroblastoma cancer stem cells display a mixed phenotype of tumor and niche origin required for survival

Neuroblastoma (NB) is a pediatric tumor of neural crest origin, and is the most common cancer of infancy. 50% of patients have metastases at diagnosis, of which 85% will die after multiple relapses from metastatic disease. We identified tumor-initiating cells (TICs) from bone marrow (BM) metastases of high-risk patients that are propagated as spheres and that have many properties of cancer stem cells, and form NB in mice with as few as 1 cell. To understand patient relapse and disease progression, we compared NB-TICs from BM with those from tumor and brain metastases and SKPs, a normal pediatric stem cell counterpart, by cDNA microarray, flow cytometry, and whole genome shotgun sequencing transcriptome analysis. BM-derived NB-TICs expressed primitive neural crest and neuronal markers as well as hematopoietic markers from primitive, myeloid, and B-cell lineages and contained VDJ gene rearrangements, which are normally associated with B-cell leukemias. Hematopoietic genes were not expressed or expressed at very low levels in tumor-derived sphere lines and a line from an NB brain metastasis, however brain metastasis-derived TICs expressed CD133, while BM-derived NB-TICs did not. Furthermore, we found that shRNA to CD74, which is upregulated in B-cell lymphoma and multiple myeloma and is a therapeutic target for those cancers, induced the rapid death of NB-TICs but not normal SKPs. Interestingly cells co-staining for the hematopoietic markers CD45 or CD74 and the neural neural progenitor marker nestin were found in BM smears of patients with relapsed NB in the bone marrow.

We suggest that metastatic TICs from some tumors adopt resident niche-specific gene expression signatures, which may aid diagnosis and the development of novel treatments. We hypothesize that drugs used for leukemia may be efficacious therapeutics for metastatic NB, and are currently testing this hypothesis in mouse models.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 12.

Abstract 3037: Integrated Genomic, MicroRNA (miRNA) and Proteomic Profiling of Ovarian Carcinoma for Biomarker Discovery

Ovarian cancer (OCa) is the fifth leading cause of cancer-related deaths in North American women and the first due to a gynecologic malignancy. The long-term effectiveness of standard therapy is generally poor and is accompanied by serious side effects. Thus there is a need for developing markers not only for diagnosis and prognosis, but also for predicting therapeutic response. Tumour progression and resistance to therapy is a consequence of the complexity of DNA, RNA and proteins. The search for effective and specific biomarkers should integrate aspects of all these factors. We have previously demonstrated that KLK6 is a promising biomarker for OCa and its observed over-expression is linked to copy-number gains of the 19q13.3/4 locus. Here, we demonstrate by multi-colour FISH analyses that the KLK locus in 81 OCas is subject to high-level of genomic instability (p&lt;0.001); and such instability is significantly co-related to grade (p&lt;0.001). KLK6-specific immunohistochemistry (IHC) showed no strong corelation with KLK6 copy-number, suggesting that other mechanisms, together with copy-number, drive its over-expression. Because 19q contains the highest number of annotated microRNAs (miRNAs) and copy-number instability may affect expression of these miRNAs, we investigated the role of miRNAs in OCa, not only for regulating KLK6, but as biomarkers for OCa. miRNA profiling of OCa cell lines and primary tumours by RT2-PCR showed the differential expression of miRNAs, consistent with other published studies in OCa. Since miRNAs can potentially affect the protein expression of hundreds of genes, the identification of such differentially expressed proteins not only provides putative biomarkers, but may also elucidate pathways for therapeutic intervention. Using Stable Isotope Labelling with Amino Acids in Cell Culture (SILAC) coupled to mass spectrometry for the OVCAR-3 cell line, cultures were labeled separately in light-Arg/Lys and heavy-Arg/Lys. In this control experiment, over 2,800 proteins were identified with 2,465 quantified. Over 94% of these quantified proteins showed a heavy:light ratio between 0.8 and 1.2, making this a robust system for quantitatively distinguishing differentially expressed proteins in the presence of miRNA precursors or inhibitors. Our profiling, as well as others, has shown the loss of expression of let-7 family members and hsa-125a-5p in OCas, and both miRNAs show decreased expression in OVCAR-3. Interestingly, these miRNAs are also predicted to target KLK6. Thus, we hypothesize that these miRNAs will not only affect the expression of KLK6, but also the expression of other target genes. In the future, using SILAC, we will identify the differentially expressed proteins affected upon re-introduction of these miRNAs through the differential labeling of miRNA-transfected OVCAR-3 vs. non-transfected OVCAR-3; thus revealing novel biomarkers for OCa.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 3037.

Topographical analysis of telomere length and correlation with genomic instability in whole mount prostatectomies

11107

Background: Many critical events in prostatic carcinogenesis appear to relate to the emergence of chromosomal instability and acquisition of genomic rearrangements. Characteristic abnormalities such as 8p loss, 8q gain, trisomy 7, PTEN microdeletions and TMPRSS2-ERG gene fusions appear to mediate mechanisms to increase neoplastic transformation in prostate cancer. Current evidence suggests that telomere dysfunction is a likely causative factor for some of these abnormalities on the basis of its relationship to mechanisms such as the break-fusion-bridge cycle that can lead to the onset of chromosomal instability. Methods: In this study, we correlated telomere length in various prostatic histologies by quantitative FISH with genomic markers of chromosomal instability by standard FISH and immunohistochemical measures of proliferation in 3 whole mount prostatectomies. Results: After analysing approximately 25,000 cells, we found that telomere shortening was correlated with an increase in the number of cells with abnormalities on chromosome 8, such as an increase in the average number of c-myc signals (r∼0.35, p∼0.02). However, there were no significant correlations with abnormalities such as trisomy 7 or abnormalities of the PTEN locus in any sample. Additional findings included; associations found with the probability of C-MYC aberrations in stroma with greater proximity to cancer (<1,000 um), a correlation between telomere length in a number of prostatic histologies (normal, atrophy, HPIN and cancer) with the adjacent stroma, and a lack of correlation between the Ki67 index of various histologies and their telomere length - all suggesting the importance of microenvironmental effects on telomere maintenance in the prostate. Finally, we also report significant telomere shortening in BPH in 2 cases, a phenomenon that has not been noted previously. Conclusions: This is the first study to directly link a mechanism of chromosomal instability with specific chromosomal abnormalities in prostatic carcinogenesis and also suggests that the microenvironmental milieu is of critical importance in the evolution of in vivo telomere homeostasis.

No significant financial relationships to disclose.

High definition cytogenetics and oligonucleotide aCGH analyses of cisplatin-resistant ovarian cancer cells

Array comparative genomic hybridization (aCGH) is a key platform to assess cancer genomic profiles. Many structural genomic aberrations cannot be detected by aCGH alone. We have applied molecular cytogenetic analyses including spectral karyotyping, multicolor banding, and fluorescence in situ hybridization with aCGH to comprehensively investigate the genomic aberrations associated with cisplatin resistance in A2780 ovarian cancer cells. A2780 is a well-established model of chemotherapeutic resistance with distinct karyotypic abnormalities in the parental and cisplatin-resistant cells. Cytogenetic analysis revealed that two unbalanced translocations, der(8)t(1;8) and der(X)t(X;1), and loss of chromosome 13 were present only in the resistant line. Our aCGH analyses detected imbalances affecting an additional 10.59% of the genome in the cisplatin-resistant cells compared with the parental. DNA copy number changes included deletions at 1p10-p22.1, 8p23.3, and Xq13.1-pter, and a duplication of 8q11.22-q23. Cryptic genomic aberrations associated with concurrent localized changes of specific gene expression included a homozygous deletion of 0.38 Mb at 1p21.3 adjacent to SNX7, and an insertional transposition of 0.85 Mb from 13q12.12 into chromosome 22. This latter rearrangement led to an overexpression of four contiguous genes that flanked one of the breakpoint regions in chromosome 13. Furthermore, 17 genes showed differential expression correlating with genomic gain or loss between the resistant and parent lines, validated by a second expression array platform. These results highlight the integration of comprehensive profiling to determine relationships of genomic aberrations and genes associated with an in vitro drug resistance model in ovarian cancer. This article contains Supplementary Material available at http://www.interscience.wiley.com/jpages/1045-2257/suppmat.

Role of KRAS and EGFR as biomarkers of response to erlotinib in National Cancer Institute of Canada Clinical Trials Group Study BR.21

PURPOSE

To evaluate the effect of KRAS and epidermal growth factor receptor (EGFR) genotype on the response to erlotinib treatment in the BR.21, placebo-controlled trial.

PATIENTS AND METHODS

We analyzed 206 tumors for KRAS mutation, 204 tumors for EGFR mutation, and 159 tumors for EGFR gene copy by fluorescent in situ hybridization (FISH). We reanalyzed EGFR deletion/mutation using two highly sensitive techniques that detect abnormalities in samples with 5% to 10% tumor cellularity. KRAS mutation was analyzed by direct sequencing.

RESULTS

Thirty patients (15%) had KRAS mutations, 34 (17%) had EGFR exon 19 deletion or exon 21 L858R mutations, and 61 (38%) had high EGFR gene copy (FISH positive). Response rates were 10% for wild-type and 5% for mutant KRAS (P = .69), 7% for wild-type and 27% for mutant EGFR (P = .03), and 5% for EGFR FISH-negative and 21% for FISH-positive patients (P = .02). Significant survival benefit from erlotinib therapy was observed for patients with wild-type KRAS (hazard ratio [HR] = 0.69, P = .03) and EGFR FISH positivity (HR = 0.43, P = .004) but not for patients with mutant KRAS (HR = 1.67, P = .31), wild-type EGFR (HR = 0.74, P = .09), mutant EGFR (HR = 0.55, P = .12), and EGFR FISH negativity (HR = 0.80, P = .35). In multivariate analysis, only EGFR FISH-positive status was prognostic for poorer survival (P = .025) and predictive of differential survival benefit from erlotinib (P = .005).

CONCLUSION

EGFR mutations and high copy number are predictive of response to erlotinib. EGFR FISH is the strongest prognostic marker and a significant predictive marker of differential survival benefit from erlotinib.

Loss of RB1 induces non-proliferative retinoma: increasing genomic instability correlates with progression to retinoblastoma

Retinoblastoma clinical observations revealed the role of tumor suppressor genes in human cancer, Knudson's 'two-hit' model of cancer induction. We now demonstrate that loss of both RB1 tumor suppressor gene alleles initiates quiescent RB1(-/-) retinomas with low level genomic instability and high expression of the senescence-associated proteins p16(INK4a) and p130. Although retinomas can remain unchanged throughout life, highly proliferative, clonal and aneuploid retinoblastomas commonly emerge, exhibiting altered gene copy number and expression of oncogenes (MYCN, E2F3, DEK, KIF14 and MDM4) and tumor suppressor genes (CDH11, p75(NTR)) and reduced expression of p16(INK4a) and p130. We suggest that RB1 inactivation in developing retina induces genomic instability, but senescence can block transformation at the stage of retinoma. However, stable retinoma is rarely clinically observed because progressive genomic instability commonly leads to highly proliferative retinoblastoma.

Predictive and Pharmacodynamic Biomarker Studies in Tumor and Skin Tissue Samples of Patients With Recurrent or Metastatic Squamous Cell Carcinoma of the Head and Neck Treated With Erlotinib

Purpose

Pharmacodynamic tissue studies were conducted on a phase I/II trial of erlotinib and cisplatin in patients with recurrent or metastatic head and neck squamous cell carcinoma (HNSCC). Levels of epidermal growth factor receptor (EGFR), downstream signaling components, and markers of angiogenesis and apoptosis were evaluated to determine the relationship between correlative end points and clinical outcomes.

Patients and Methods

Pretreatment and during-treatment tumor and skin biopsies, and archival tumor specimens were evaluated for EGFR, phosphorylated (p) -EGFR, extracellular signal-regulated kinase (ERK), p-ERK, Akt, p-Akt, Ki67, p27, p-nuclear factor kappa B (NFκB), p-signal transducer and activator of transcription 3 (STAT3), and EGFR gene copy number.

Results

On 37 archival samples, response to therapy was evident in two of four (50%) patients with high EGFR gene copy number tumors and in four of 27 (15%) patients with low gene copy number tumors. On nine paired tumor biopsies, elevated pretreatment levels of p27 and p-STAT3 predicted for prolonged time to progression (TTP) and overall survival (OS; P ≤ .03). With treatment, a decrease in p-EGFR, p-NFκB, and p27 correlated with increased TTP, OS, or both TTP and OS, respectively (P ≤ .04). Multidimensional scaling (MDS) models revealed clustering profiles of tumor markers by immunofluorescence could predict response. On 32 paired skin biopsies, suppression of p-EGFR with therapy correlated with increased OS (P = .045).

Conclusion

High EGFR gene copy in tumor specimens may predict which patients have an increased likelihood of response to erlotinib, and decreased p-EGFR level in skin biopsies during therapy may represent a potential surrogate marker for improved clinical outcome. MDS represents a novel way to evaluate the relationships between molecular markers and clinical outcome. Additional biomarker studies with larger sample sizes are required to elucidate HNSCC patients who may benefit from this targeted therapy.

Papillary renal cell carcinoma within a renal oncocytoma: case report of an incidental finding of a tumour within a tumour

The most common renal tumours are clear cell, papillary, chromophobe and collecting duct renal cell carcinomas (RCCs), and benign oncocytomas and angiomyolipomas. Tumours with hybrid features between some of these entities have been recognised; in particular, tumours with features of both chromophobe RCC and oncocytoma. Case reports describing one distinct type of primary renal tumour actually within another are very rare. The incidental finding of a papillary RCC located in an oncocytoma in a nephrectomy specimen from a 75-year-old man is described. Morphological criteria for each tumour type were completely satisfied and fluorescence in situ hybridisation detected the expected number of copies of chromosome 7 in the cells of each tumour type. The cells in the papillary tumour contained three copies, whereas the oncocytoma cells contained only two per nucleus. To our knowledge, this is the first report of a papillary RCC being identified within an oncocytoma.

Large scale copy number variation (CNV) at 14q12 is associated with the presence of genomic abnormalities in neoplasia

Background

Advances made in the area of microarray comparative genomic hybridization (aCGH) have enabled the interrogation of the entire genome at a previously unattainable resolution. This has lead to the discovery of a novel class of alternative entities called large-scale copy number variations (CNVs). These CNVs are often found in regions of closely linked sequence homology called duplicons that are thought to facilitate genomic rearrangements in some classes of neoplasia. Recently, it was proposed that duplicons located near the recurrent translocation break points on chromosomes 9 and 22 in chronic myeloid leukemia (CML) may facilitate this tumor-specific translocation. Furthermore, ~15–20% of CML patients also carry a microdeletion on the derivative 9 chromosome (der(9)) and these patients have a poor prognosis. It has been hypothesised that der(9) deletion patients have increased levels of chromosomal instability.

Results

In this study aCGH was performed and identified a CNV (RP11-125A5, hereafter called CNV14q12) that was present as a genomic gain or loss in 10% of control DNA samples derived from cytogenetically normal individuals. CNV14q12 was the same clone identified by Iafrate et al. as a CNV. Real-time polymerase chain reaction (Q-PCR) was used to determine the relative frequency of this CNV in DNA from a series of 16 CML patients (both with and without a der(9) deletion) together with DNA derived from 36 paediatric solid tumors in comparison to the incidence of CNV in control DNA. CNV14q12 was present in ~50% of both tumor and CML DNA, but was found in 72% of CML bearing a der(9) microdeletion. Chi square analysis found a statistically significant difference (p ≤ 0.001) between the incidence of this CNV in cancer and normal DNA and a slightly increased incidence in CML with deletions in comparison to those CML without a detectable deletion.

Conclusion

The increased incidence of CNV14q12 in tumor samples suggests that either acquired or inherited genomic variation of this new class of variation may be associated with onset or progression of neoplasia.

Erlotinib in lung cancer - molecular and clinical predictors of outcome

BACKGROUND

A clinical trial that compared erlotinib with a placebo for non-small-cell lung cancer demonstrated a survival benefit for erlotinib. We used tumor-biopsy samples from participants in this trial to investigate whether responsiveness to erlotinib and its impact on survival were associated with expression by the tumor of epidermal growth factor receptor (EGFR) and EGFR gene amplification and mutations.

METHODS

EGFR expression was evaluated immunohistochemically in non-small-cell lung cancer specimens from 325 of 731 patients in the trial; 197 samples were analyzed for EGFR mutations; and 221 samples were analyzed for the number of EGFR genes.

RESULTS

In univariate analyses, survival was longer in the erlotinib group than in the placebo group when EGFR was expressed (hazard ratio for death, 0.68; P=0.02) or there was a high number of copies of EGFR (hazard ratio, 0.44; P=0.008). In multivariate analyses, adenocarcinoma (P=0.01), never having smoked (P<0.001), and expression of EGFR (P=0.03) were associated with an objective response. In multivariate analysis, survival after treatment with erlotinib was not influenced by the status of EGFR expression, the number of EGFR copies, or EGFR mutation.

CONCLUSIONS

Among patients with non-small-cell lung cancer who receive erlotinib, the presence of an EGFR mutation may increase responsiveness to the agent, but it is not indicative of a survival benefit.

High-resolution mapping of genomic imbalance and identification of gene expression profiles associated with differential chemotherapy response in serous epithelial ovarian cancer

Array comparative genomic hybridization (aCGH) and microarray expression profiling were used to subclassify DNA and RNA alterations associated with differential response to chemotherapy in ovarian cancer. Two to 4 Mb interval arrays were used to map genomic imbalances in 26 sporadic serous ovarian tumors. Cytobands 1p36, 1q42-44, 6p22.1-p21.2, 7q32.1-q34 9q33.3-q34.3, 11p15.2, 13q12.2-q13.1, 13q21.31, 17q11.2, 17q24.2-q25.3, 18q12.2, and 21q21.2-q21.3 were found to be statistically associated with chemotherapy response, and novel regions of loss at 15q11.2-q15.1 and 17q21.32-q21.33 were identified. Gene expression profiles were obtained from a subset of these tumors and identified a group of genes whose differential expression was significantly associated with drug resistance. Within this group, five genes (GAPD, HMGB2, HSC70, GRP58, and HMGB1), previously shown to form a nuclear complex associated with resistance to DNA conformation-altering chemotherapeutic drugs in in vitro systems, may represent a novel class of genes associated with in vivo drug response in ovarian cancer patients. Although RNA expression change indicated only weak DNA copy number dependence, these data illustrate the value of molecular profiling at both the RNA and DNA levels to identify small genomic regions and gene subsets that could be associated with differential chemotherapy response in ovarian cancer.

High-resolution cDNA microarray CGH mapping of genomic imbalances in osteosarcoma using formalin-fixed paraffin-embedded tissue

Formalin-fixed paraffin embedded (FFPE) tumor tissue provides an opportunity to perform retrospective genomic studies of tumors in which chromosomal imbalances are strongly associated with oncogenesis. The application of comparative genomic hybridization (CGH) has led to the rapid accumulation of cytogenetic information on osteosarcoma (OS); however, the limited resolving power of metaphase CGH does not permit precise mapping of imbalances. Array CGH allows quantitative detection and more precise delineation of copy number aberrations in tumors. Unfortunately the high cost and lower density of BACs on available commercial arrays has limited the ability to comprehensively profile copy number changes in tumors such as OS that are recurrently subject to genomic imbalance. In this study a cDNA/EST microarray including 18,980 human cDNAs (which represent all 22 pairs of autosomal chromosomes and chromosome X) was used for CGH analysis of eight OS FFPE. Chromosomes 1, 12, 17, and X harbored the most imbalances. Gain/amplification of X was observed in 4/8 OS, and in keeping with other recent genomic analyses of OS, gain/amplification of 17p11.2 was often accompanied by a distal deletion in the region of the p53 gene. Gain/amplification of the X chromosome was verified using interphase FISH carried out on a subset of OS FFPE sections and OS tissue arrays.

Defective DNA Strand Break Repair after DNA Damage in Prostate Cancer Cells

Together with cell cycle checkpoint control, DNA repair plays a pivotal role in protecting the genome from endogenous and exogenous DNA damage. Although increased genetic instability has been associated with prostate cancer progression, the relative role of DNA double-strand break repair in malignant versus normal prostate epithelial cells is not known. In this study, we determined the RNA and protein expression of a series of DNA double-strand break repair genes in both normal (PrEC-epithelial and PrSC-stromal) and malignant (LNCaP, DU-145, and PC-3) prostate cultures. Expression of genes downstream of ATM after ionizing radiation-induced DNA damage reflected the p53 status of the cell lines. In the malignant prostate cell lines, mRNA and protein levels of the Rad51, Xrcc3, Rad52, and Rad54 genes involved in homologous recombination were elevated approximately 2- to 5-fold in comparison to normal PrEC cells. The XRCC1, DNA polymerase-beta and -delta proteins were also elevated. There were no consistent differences in gene expression relating to the nonhomologous end-joining pathway. Despite increased expression of DNA repair genes, malignant prostate cancer cells had defective repair of DNA breaks, alkali-labile sites, and oxidative base damage. Furthermore, after ionizing radiation and mitomycin C treatment, chromosomal aberration assays confirmed that malignant prostate cells had defective DNA repair. This discordance between expression and function of DNA repair genes in malignant prostate cancer cells supports the hypothesis that prostate tumor progression may reflect aberrant DNA repair. Our findings support the development of novel treatment strategies designed to reinstate normal DNA repair in prostate cancer cells.

Use of whole genome amplification and comparative genomic hybridisation to detect chromosomal copy number alterations in cell line material and tumour tissue

We have established that whole genome amplification (WGA), in conjunction with genomic DNA array comparative genomic hybridisation (gaCGH) allows for the identification of genome-wide copy number abnormalities (CNAs) in DNA extracted from both cell line and patient material. To determine the fidelity and reproducibility of WGA to detect copy number imbalances using gaCGH, well characterized cell line genomic DNA was analysed. The gaCGH data obtained from non-amplified DNA and amplified DNA for the neuroblastoma cell line NUB7 and a paediatric medulloblastoma patient was almost identical. In addition, laser capture microdissection (LCM) of prostate tumour cells and subsequent WGA allowed for the detection of a number of CNAs that may not have been identified if DNA had been extracted in bulk from heterogeneous tissue. The results presented here demonstrate the use of WGA for generating sufficient DNA for gaCGH analysis without the introduction of significant sequence representation bias. The combination of amplification and gaCGH using DNA extracted from archival patient material has the potential for permitting the studying of DNA from small cancerous or pre-cancerous foci, which may help to identify potential genomic markers for early diagnosis.

Defining a 0.5-mb region of genomic gain on chromosome 6p22 in bladder cancer by quantitative-multiplex polymerase chain reaction

Metaphase-based comparative genomic hybridization (CGH) has identified recurrent regions of gain on different chromosomes in bladder cancer, including 6p22. These regions may contain activated oncogenes important in disease progression. Using quantitative multiplex polymerase chain reaction (QM-PCR) to study DNA from 59 bladder tumors, we precisely mapped the focal region of genomic gain on 6p22. The marker STS-X64229 had copy number increases in 38 of 59 (64%) tumors and the flanking markers, RH122450 and A009N14, had copy number gains in 33 of 59 (56%) and 26 of 59 (45%) respectively. Contiguous gain was present for all three markers in 14 of 59 (24%) and for two (RH122450 and STS-X64229) in 25 of 59 (42%). The genomic distance between the markers flanking STS-X64229 is 0.5 megabases, defining the minimal region of gain on 6p22. Locus-specific interphase fluorescence in situ hybridization confirmed the increased copy numbers detected by QM-PCR. Current human genomic mapping data indicates that an oncogene, DEK, is centrally placed within this minimal region. Our findings demonstrate the power of QM-PCR to narrow the regions identified by CGH to facilitate identifying specific candidate oncogenes. This also represents the first study identifying DNA copy number increases for DEK in bladder cancer.

Advanced cancer genetics in neurosurgical research

RAPID ADVANCES IN the technology used to study nucleic acids have revealed a great deal regarding the underlying biology of cancer. Most cancers arise as a result of chromosomal rearrangements and deoxyribonucleic acid mutations that lead to the activation of proto-oncogenes and loss of function of tumor suppressor genes. There are a number of different molecular routes that lead to these common goals, necessitating several different techniques of mutational analysis. Although many of these techniques can be difficult in practice, most are conceptually simple. We discuss several of the current techniques in cytogenetics and molecular genetics that are widely used in cancer biology laboratories. Understanding the molecular events that lead to cancer should allow the future development of targeted, nontoxic therapeutics similar to modern-day antibiotics. These technologies are being progressively applied in clinical neurosurgery, where they will be used to detect, diagnose, stratify, and treat cancers of the nervous system. High demand from an increasingly educated patient population means that neurosurgeons will need to be familiar with many of these techniques.

High-resolution mapping of amplifications and deletions in pediatric osteosarcoma by use of CGH analysis of cDNA microarrays

Conventional cytogenetic and comparative genomic hybridization (CGH) studies have shown that osteosarcomas (OSs) are characterized by complex structural and numerical chromosomal alterations and gene amplification. In this study, we used high-resolution CGH to investigate recurrent patterns of genomic imbalance by use of DNA derived from nine OS tumors hybridized to a 19,200-clone cDNA microarray. In six OSs, there was copy number gain or amplification of 6p, with a minimal region of gain centering on segment 6p12.1. In seven OSs, the pattern of amplification affecting chromosome arm 8q showed high-level gains of 8q12-21.3 and 8q22-q23, with amplification of the MYC oncogene at 8q24.2. Seven OSs showed copy number gain or amplification of 17p between the loci bounded by GAS7 and PMI (17p11.2-17p12), and three of these tumors also showed small losses at 17p13, including the region containing TP53. An in silico analysis of the distribution of segmental duplications (duplicons) in this region identified a large number of tracts consisting of paralogous sequences mapping to the 17p region, encompassing the region of deletions and amplifications in OS. Interestingly, within this same region there were clusters of duplicons and several genes that are expressed during bone morphogenesis and in OS. In summary, microarray CGH analysis of the chromosomal imbalances of OS confirm the overall pattern observed by use of metaphase CGH and provides a more precise refinement of the boundaries of genomic gains and losses that characterize this tumor.

Quantitative PCR identifies a minimal deleted region of 120 kb extending from the Philadelphia chromosome ABL translocation breakpoint in chronic myeloid leukemia with poor outcome

Fluorescence in situ hybridization (FISH) analysis has shown previously that 10-15% of chronic myeloid leukemias (CML) have hemizygous deletions of variable sizes affecting regions that flank the ABL and BCR translocation breakpoints on the derivative chromosome 9, and these patients have a poor outcome. FISH studies using large commercial genomic probes have previously suggested that haploinsufficiency of sequences flanking either ABL or BCR modify the disease process of CML and lead to an unfavorable prognosis. In this present study, real-time quantitative PCR (Q-PCR) analysis was used to identify and map much smaller hemizygous microdeletions in a subset of CML patients that were not deleted using large genomic FISH probes. Microdeletions were identified by Q-PCR in 25 of 71 patients selected based on less favorable outcome (chronic phase duration of less than 96 months and a survival time of less than 84 months). In contrast, no microdeletion was detected in any of 18 CML samples selected from a group with a more favorable outcome. Detailed mapping of the 25 Q-PCR microdeletions showed that the minimal deleted region extended approximately 120 kb from the 5' end of the ABL gene in the centromeric direction on the derivative chromosome 9, and the region 3' to BCR on chromosome 22 was excluded. Of the four ESTs and/or genes that map to the 120 kb region, the putative tumor suppressor PRDM12 is the strongest candidate gene. The potential role for each sequence in modifying the clinical behavior of CML is presented.

Chromosomal localization of DNA amplifications in neuroblastoma tumors using cDNA microarray comparative genomic hybridization

Conventional comparative genomic hybridization (CGH) profiling of neuroblastomas has identified many genomic aberrations, although the limited resolution has precluded a precise localization of sequences of interest within amplicons. To map high copy number genomic gains in clinically matched stage IV neuroblastomas, CGH analysis using a 19,200-feature cDNA microarray was used. A dedicated (freely available) algorithm was developed for rapid in silico determination of chromosomal localizations of microarray cDNA targets, and for generation of an ideogram-type profile of copy number changes. Using these methodologies, novel gene amplifications undetectable by chromosome CGH were identified, and larger MYCN amplicon sizes (in one tumor up to 6 Mb) than those previously reported in neuroblastoma were identified. The genes HPCAL1, LPIN1/KIAA0188, NAG, and NSE1/LOC151354 were found to be coamplified with MYCN. To determine whether stage IV primary tumors could be further subclassified based on their genomic copy number profiles, hierarchical clustering was performed. Cluster analysis of microarray CGH data identified three groups: 1) no amplifications evident, 2) a small MYCN amplicon as the only detectable imbalance, and 3) a large MYCN amplicon with additional gene amplifications. Application of CGH to cDNA microarray targets will help to determine both the variation of amplicon size and help better define amplification-dependent and independent pathways of progression in neuroblastoma.

Resolution of genotypic heterogeneity in prostate tumors using polymerase chain reaction and comparative genomic hybridization on microdissected carcinoma and prostatic intraepithelial neoplasia foci

Prostate cancer (CaP) is a multifocal heterogenous disease. A major challenge in CaP research is to identify genetic biomarkers that herald aggressive transformation. To investigate the effect of tumor heterogeneity on the analysis of genomic aberration, we compared the results of comparative genomic hybridization (CGH) analysis of DNA extracted from tumor bulk against that of DNA amplified by degenerate oligonucleotide primed polymerase chain reaction (DOP-PCR) from homogeneous cell population obtained by laser capture microdissection of discrete tumor foci. Sampling by microdissection, aberrations were observed in three of three foci of carcinoma involved with prostatic capsule, and in two of three prostatic intraepithelial neoplasia (PIN) foci examined. Carcinoma foci consistently exhibited more extensive aberrations than the PIN samples obtained from the same tumor. Within these samples, the different tumor foci exhibited gain of 8q, whereas PIN showed no consistent aberration. Using bulk extracted DNA, CGH detected aberrations in only 3 of 21 samples investigated, despite the known trisomy 8 status, as revealed by fluorescence in situ hybridization. The results of this study demonstrate that CGH analysis using bulk dissected fresh tissue is insufficiently sensitive to fully detect the chromosomal numerical aberrations in CaP. Given the considerable intratumor genomic heterogeneity, CGH with microdissection and DOP-PCR amplification provides a more complete repertoire of aberrations as well as a better phenotype-genotype correlation in prostate tumors.

Comparative genomic hybridization analysis identifies gains of 1p35 approximately p36 and chromosome 19 in osteosarcoma

Osteosarcomas (OS) are aggressive tumors of the bone and often have a poor prognosis. Conventional cytogenetic analyses of OS have revealed highly complex karyotypes, with numerous abnormalities. In this study, we analyzed 18 untreated OS tumors from 17 patients of the younger incidence age group by comparative genomic hybridization (CGH), 4 tumors by spectral karyotyping (SKY) and fluorescence in situ hybridization (FISH). Comparative genomic hybridization identified frequent copy number changes of the chromosomal region 1p (10/17) and gain of part or all of chromosome 19(8/17). In addition gains were observed at 5p(3/17), 8q(3/17), 16p(3/17), and 17p(5/17); and losses at chromosomes 2q(3/17), 10(4/17) and 13(3/17). High level gains were detected in the 8q23 approximately q24 region in two tumors as well as at 17p in one primary and a metastatic tumor. Minimal regions of gain were present at 1p35 approximately p36.3 (8/17); 5p14 approximately p15.2 (3/17), and 8q22 approximately q24.3 (3/17). SKY analysis demonstrated that OS has a complex pattern of clonal and non-clonal rearrangements and helped confirm the structural basis for the imbalances detected by CGH. Spectral karyotyping confirmed an overall pattern of chromosomal gain affecting 1p in all four tumors. Fluorescence in situ hybridization analysis from these tumors confirmed the gain of the 1p36 region in 2 tumors as determined by CGH analysis as well as the amplification of 8q.

Molecular cytogenetic analysis of glial tumors using spectral karyotyping and comparative genomic hybridization

BACKGROUND

Glial tumors are the most common tumors of the central nervous system, affecting individuals of all ages. Conventional cytogenetics have been unable to identify a consistent chromosomal translocation or rearrangement in this group of tumors; thus, more advanced molecular cytogenetic approaches are required.

METHODS AND RESULTS

In this study, 16 glial tumors, including two recurrences and six glioma cell lines, were analyzed by spectral karyotyping (SKY) and comparative genomic hybridization (CGH). From 169 rearrangements detected by SKY, chromosomes 1 and 10 were the most frequently affected by translocation (18 of 169 and 16 of 169 rearrangements, respectively). Other frequently altered chromosomes included chromosomes 3 (13 of 169 rearrangements), 5 (ten of 169 rearrangements), 7 (ten of 169 rearrangements ), and 11 (ten of 169 rearrangements). A clustering of centromeric breakpoints was detected in chromosomes 3, 5, 10, 11, 16, 17, and 20. CGH analysis identified consistent gain of part or all of chromosome 7 among the 10 astrocytic tumors (five of ten specimens) in the study group. Analysis of the three gangliogliomas and one ependymoma identified a much simpler pattern of primarily numerical change.

CONCLUSION

Application of improved cytogenetic methods can increase our abilities to progress toward effective strategies of molecular diagnosis and classification of glial tumors.

Acquisition of secondary structural chromosomal changes in pediatric ewing sarcoma is a probable prognostic factor for tumor response and clinical outcome

BACKGROUND

The Ewing sarcoma (ES) group of tumors commonly have the t(11;22)(q24;q12) or other rearrangements involving 22q12. In addition to these consistent aberrations, both numeric and structural aberrations have been reported: namely gains of chromosomes 8 and 12, the unbalanced translocation t(1;16), and deletions at the short arm of chromosome 1.

METHODS

To evaluate the frequency and to study the prognostic implications of some of these aberrations in children, the authors performed a pilot study of 26 ES pediatric patients by classic cytogenetics and/or interphase fluorescence in situ hybridization (FISH) and compared these data with clinical parameters.

RESULTS

Gains of chromosomes 8 and 12 were detected, by interphase FISH, in 48% (10 of 21) and 38% (6 of 16) of the tumors, respectively, and this was not significant with respect to treatment response. Statistical analysis revealed that the presence of additional secondary structural chromosomal aberrations was associated with an unfavorable outcome (P = 0.0034 as an independent prognostic value as an unfavorable marker). Presence of metastasis at diagnosis also was found to be associated with poor outcome (P = 0.0131). Spectral karyotyping analysis was shown to facilitate the detection of more complex structural chromosomal aberrations in a representative ES tumor.

CONCLUSIONS

It is important to determine whether additional structural chromosomal aberrations are present in ES tumors because it appears that a more complex karyotype with multiple chromosomal aberrations is associated with poor outcome in ES.

Molecular cytogenetic analysis of medulloblastomas and supratentorial primitive neuroectodermal tumors by using conventional banding, comparative genomic hybridization, and spectral karyotyping

OBJECT

Medulloblastomas and related primitive neuroectodermal tumors (PNETs) of the central nervous system are malignant, invasive embryonal tumors with predominantly neuronal differentiation that comprise 20% of pediatric brain tumors. Cytogenetic analysis has shown that alterations in chromosome 17, particularly the loss of 17p and the formation of isochromosome 17q, as well as the gain of chromosome 7 are the most common changes among this group of tumors. Comparative genomic hybridization (CGH) studies have largely confirmed these cytogenetic findings and have also identified novel regions of gain, loss, and amplification. The advent of more sophisticated multicolored fluorescence in situ hybridization (FISH) procedures such as spectral karyotyping (SKY) now permits complete recognition of all aberrations including extremely complex rearrangements. The authors report a retrospective analysis of 19 medulloblastoma and five PNET cases studied using combinations of classic banding analysis, FISH, CGH, and SKY to examine comprehensively the chromosomal aberrations present in this tumor group and to attempt to identify common structural rearrangement(s).

METHODS

The CGH data demonstrate gains of chromosomes 17q and 7 in 60% of the tumors studied, which confirms data reported in the current literature. However, the authors have also combined the results of all three molecular cytogenetic assays (Giemsa banding, CGH, and SKY) to reveal the frequency of chromosomal rearrangement (gained, lost, or involved in structural rearrangement).

CONCLUSIONS

The combined results indicate that chromosomes 7 and 17 are the most frequently rearranged chromosomes (10.1% and 8.9%, respectively, in all rearrangements detected). Furthermore, chromosomes 3 (7.8%), 14 (7%), 10 (6.7%), and 22 (6.5%) were also found to be frequently rearranged, followed by chromosomes 6 (6.5%), 13 (6.2%), and 18 (6.2%). Eight (33%) of 24 tumors exhibited high-level gains or gene amplification. Amplification of MYCN was identified in four tumors, whereas amplification of MYCC was identified in one tumor. One tumor exhibited a high-level gain of chromosome 9p. Additionally, desmoplastic medulloblastomas and large-cell medulloblastomas exhibited higher karyotype heterogeneity, amplification, and aneusomy than classic medulloblastomas.

Is the EWS/FLI-1 fusion transcript specific for Ewing sarcoma and peripheral primitive neuroectodermal tumor? A report of four cases showing this transcript in a wider range of tumor types

The presence of t(11;22)(q24;q12) is often considered diagnostic of Ewing sarcoma and peripheral primitive neuroectodermal tumor. We report four cases, all of which possessed this translocation as detected by reverse transcriptase polymerase chain reaction and confirmed by sequencing with or without fluorescent in situ hybridization, but none of which were Ewing sarcoma or peripheral primitive neuroectodermal tumor by histological criteria. Two were polyphenotypic tumors and two were mixed embryonal and alveolar rhabdomyosarcomas. Only one case was positive for MIC2 by immunohistochemistry and only in a rare cell. Two cases (one polyphenotypic tumor and one rhabdomyosarcoma) had double minute chromosomes with > 100 copies of the MDM2 gene. The presence of the t(11;22)(q24;ql2) translocation should probably not be considered diagnostic of Ewing sarcoma and peripheral primitive neuroectodermal tumor in the absence of supporting histological evidence. The presence of this translocation in Ewing sarcoma and peripheral primitive neuroectodermal tumor has been taken as evidence that these two tumors are related. Extending this relationship to include some polyphenotypic tumors and some rhabdomyosarcomas may not be justified unless additional evidence is gathered. Pathologists and oncologists will need to decide whether treatment regimens for tumors are better based on phenotype rather than genotype when these two profiles are seemingly in conflict.

Canine X chromosome-linked hereditary nephritis: a genetic model for human X-linked hereditary nephritis resulting from a single base mutation in the gene encoding the alpha 5 chain of collagen type IV

Many families with X-chromosome linked hereditary nephritis (HN) have mutations in the gene on the X chromosome that codes for the alpha 5 chain of collagen type IV. Canine X-linked HN is an animal model for human X-linked HN. To study the alpha 5(IV) gene in this model, we used the nucleotide sequence published for the human alpha 5(IV) cDNA to construct sets of primers covering approximately 95% of the complete cDNA. cDNA from both affected and normal dog kidneys was amplified by PCR in nine overlapping regions. The nucleotide sequence encoding the noncollagenous domain NC1 hybridized to the human X chromosome and was 93% identical at the DNA level and 97% identical at the protein level to the human alpha 5(IV) NC1 domain, confirming that the canine alpha 5(IV) cDNA had been amplified. Sequence analysis of the alpha 5(IV) cDNA detected a single nucleotide substitution, G-->T, in affected dogs, changing a codon for a conserved glycine residue (GGA) to a stop codon (TGA). When genomic DNA was amplified, the same abnormality was found in exon 35. Using the canine NC1 domain cDNA as a probe for Northern analysis, two transcripts of approximately 8.6 kb and approximately 6.7 kb were identified in both normal and affected male dog kidney RNA. However, the abundance of both transcripts was decreased by a factor of approximately 10 in the affected dog. These results establish at the molecular level that canine X-linked HN is a model for human X-linked HN. This model provides an opportunity to determine the efficacy of new therapies and to investigate the role of the alpha 5(IV) chain in type IV collagen assembly.

Renal disease in carrier female dogs with X-linked hereditary nephritis. Implications for female patients with this disease

Male dogs with X-linked hereditary nephritis (HN) serve as a model for studying male patients with this disease. In the present study, carrier female dogs were found to resemble female patients in showing a broad range of renal dysfunction. Of 37 carrier female dogs studied, all were healthy up to 5 years of age; however, all had proteinuria develop at 2 to 3 months, and focal segmental glomerulosclerosis (FSGS) was detected after 7 months. After 5 years, 4 of 13 dogs remained healthy and showed mild FSGS on renal biopsy; 4 had mild renal dysfunction develop and their kidneys showed extensive FSGS; 5 died prematurely of renal failure with end-stage kidneys. By immunofluorescence, using antibody to the NC1 domain of collagen type IV, segmental staining of glomerular basement membranes (GBM) was seen in all dogs before 3 to 4 years, and lesions of FSGS were negative. Thereafter, a transition to global staining of GBM was noted and lesions of FSGS became positive. Lens capsule and basement membranes in lung and choroid plexus showed discontinuous staining in two young carrier female dogs and continuous staining in one older carrier female dog. By electron microscopy, multilaminar splitting of some GBM was seen up to 4 years, and thereafter, splitting took on a compressed appearance, with the layers becoming apposed though still detectable. The authors conclude that: 1) carrier female dogs with X-linked HN are mosaics for an abnormality in the NC1 domain of GBM and other basement membranes; 2) FSGS develops in all carrier female dogs in glomerular capillary loops that possess an abnormal NC1 domain, and progresses to a variable extent in different dogs; and 3) the abnormality of NC1 in GBM of carrier female dogs appears to diminish with age, but this does not prevent progression of renal disease. Similar conclusions may apply to females with X-linked HN.