The relationship between tumour site, clinicopathological characteristics and cancer-specific survival in patients undergoing surgery for colorectal cancer

AIM

It is recognised that colorectal cancer may arise from different genomic instability pathways. There is evidence to suggest that colon and rectal cancers exhibit different clinicopathological features. We examined the relationship between tumour site, clinicopathological characteristics and cancer-specific survival in patients undergoing potentially curative resection for colorectal cancer.

METHOD

Four hundred and eleven patients who underwent surgery. Clinicopathological data including components of the Peterson index, Klintrup scores, haemoglobin and the modified Glasgow Prognostic Score (mGPS) were studied.

RESULTS

There were 134 (33%) right sided, 125 (30%) left sided and 152 (37%) rectal tumours. Emergency presentation (P < 0.001), anaemia (P < 0.001), higher mGPS (P < 0.001), advanced T stage (P < 0.001), poor differentiation (P < 0.001) and older age (P < 0.05) were more commonly observed in right sided cancer. The mean follow-up was 94 months (minimum 36 months) and 114 patients died of cancer. There was no difference between tumour site and survival (P = 0.427). On multivariate analysis older age (P = 0.015), lymph node ratio (P < 0.001), mGPS (P = 0.028), Peterson Index (P < 0.001) and Klintrup score (P = 0.008) were independently related to cancer-specific survival. Klintrup score was only associated with poor cancer-specific survival in rectal cancer (P = 0.009).

CONCLUSION

The study suggests that colorectal cancer is a group of heterogeneous tumours with different clinicopathological features. Despite this, there was no difference between tumour site and survival. The prognostic role of clinicopathological factors in tumours arising from different genomic instability pathways requires further study.

Extraction of DNA from microdissected archival tissues

Many modern analytical methods require little material, which has made feasible biochemical and molecular analyses of small tissue fragments, even individual cells, by microdissection of histological sections (1,2). Polymerase chain reaction (PCR) can potentially be applied to the analysis of single DNA molecules, as in the analysis of single haploid cells, such as spermatozoa (3). This sensitivity requires careful attention to technique and proper controls to avoid false positive or other spurious results.

The relationships between cellular components of the peritumoural inflammatory response, clinicopathological characteristics and survival in patients with primary operable colorectal cancer

Background:

The host inflammatory response is an important determinant of cancer outcome. We examined different methods of assessing the local inflammatory response in colorectal tumours and explored relationships with both clinicopathological characteristics and survival.

Methods:

Cohort study of patients (n=130) with primary operable colorectal cancer and mature follow-up. Local inflammatory response at the invasive margin was assessed with: (1) a semi-quantitative assessment of peritumoural inflammation using Klintrup–Makinen (K–M) grading and (2) an assessment of individual immune cell infiltration (lymphocytes, plasma cells, neutrophils, macrophages and eosinophils).

Results:

The peritumoural inflammatory response was K–M low grade in 48% and high grade in 52%. Inflammatory cells were primarily macrophages, lymphocytes and neutrophils with relatively few plasma cells or eosinophils. On univariate analysis, K–M grade, lymphocyte infiltration and plasma cell infiltration were associated with cancer-specific survival. On multivariate analysis, only systemic inflammatory response, TNM (tumour, node and metastases) stage, venous invasion, tumour necrosis and K–M grade were independently associated with cancer-specific survival. There was no relationship between local infiltration of inflammatory cells and a systemic inflammatory response. However, high K–M grade, lymphocyte infiltration and plasma cell infiltration were associated with a number of favourable pathological characteristics, including an absence of venous invasion.

Conclusion:

Infiltration of inflammatory cells in the invasive margin of colorectal tumours is beneficial to survival. The adaptive immune response appears to have a prominent role in the prevention of tumour progression in patients with colorectal cancer.

Prospective study of the role of inflammation in renal cancer

BACKGROUND

The local and systemic inflammatory responses provide prognostic information in cancer. The modified Glasgow Prognostic Score (mGPS) provides additional prognostic information than C-reactive protein (CRP) alone when assessing the systemic inflammation in cancer. The aim of this study was to determine the role of local and systemic inflammation in renal cancer.

METHODS

The cohort consisted of 79 patients who had undergone potential curative resection. Systemic inflammation, mGPS, was constructed by measuring preoperative CRP and albumin concentrations and the Klintrup-Makinen score was evaluated histologically for the local inflammatory response. Pathological parameters such as T stage, grade and tumour necrosis were also assessed. The local inflammatory response was assessed by examining all inflammatory cells at the tumour edge on diagnostic haematoxylin and eosin slides.

RESULTS

On univariate analysis, T stage (p < 0.001), grade (p = 0.044) and mGPS (p < 0.001) were significant predictors of cancer-specific survival. On multivariate analysis, mGPS (hazard ratio 8.64, 95% confidence interval 3.5-21.29, p < 0.001) was the only significant independent predictor of cancer-specific survival.

CONCLUSION

A preoperative systemic inflammatory response as measured by the mGPS is an independent predictor of poor cancer-specific survival in renal cancer in patients undergoing potential curative resection.

Is the presence or absence of tumour necrosis a significant predictor of survival in renal cell cancer?

OBJECTIVE

Currently when renal cancer pathology is assessed the presence or absence of necrosis is simply reported. It has been suggested that a presence or absence response ignores heterogeneity and a classification based on the extent of necrosis involvement would aid prognostic value in cancer-specific survival. The aim of this study was to determine whether a quantitative assessment of tumour necrosis would provide additional prognostic information.

METHODS

We studied the pathological features and cancer-specific survival of 47 patients with renal cancer undergoing surgery with curative intent. A quantitative assessment of tumour necrosis was compared to the presence or absence of necrosis.

RESULTS

Tumour necrosis was present in 27 of 47 cases. A simple assessment of the presence or absence was not associated with cancer-specific survival (p = 0.052). When assessed quantitatively, tumour necrosis was associated with decreased cancer-specific survival (p < 0.001). A 2-tiered assessment, <25% and >25% involvement of necrosis, was further utilised and shown to predict cancer-specific survival (p < 0.001). On multivariate analysis, using this 2-tiered assessment of <25% and >25% involvement of necrosis was retained as a significant independent factor for cancer-specific survival (HR 11.84, 95% CI 3.81-36.75, p < 0.001).

CONCLUSION

A simple assessment of the presence/absence of tumour necrosis is reported to be a prognostic factor in renal cell cancer. In this study, the presence/absence was not shown to be a significant prognostic marker of cancer-specific survival. However, a more accurate quantitative assessment of tumour necrosis, whereby a 2-tiered response is still utilised, but basing this on <25% and >25% involvement of necrosis was statistically significant and independent in predicting cancer-specific survival.

Still waiting for predictive biomarkers in Barrett's oesophagus

Barrett's oesophagus is important as a precursor of oesophageal adenocarcinoma via a metaplasia-dysplasia-carcinoma sequence. It is often detected on upper gastrointestinal endoscopy. In the absence of glandular dysplasia the risk of progression to cancer is low but ascertainment of dysplasia is not always straightforward. Sparse mucosal sampling may miss dysplasia, or reactive changes may be overinterpreted due to inter and intraobserver variation. Low-grade and even high-grade dysplasia do not necessarily progress, provided prevalent cancer has been rigorously excluded. This indeterminacy motivates an ongoing search for clinically useful predictive biomarkers. Although many genetic and epigenetic abnormalities have been associated with neoplastic progression in Barrett's mucosa no molecular tests have as yet been accepted into routine pathology practice. Challenges of assay definition remain and many marker studies lack statistical power or have other methodological flaws. Even where strong evidence of clinically relevant predictive value does exist (in the case of ploidy analysis by flow or image cytometry) adoption has been minimal, likely reflecting technological and possible reimbursement obstacles. Well designed multicentre studies are likely to be required to translate improved knowledge of Barrett's carcinogenesis into clinically significant progress on predictive testing, and will require a degree of cooperation not so far widely seen in the field.

Aberrant expression of TopBP1 in breast cancer

AIMS

The TopBP1 protein includes eight BRCT domains (originally identified in BRCA1) and has homology with BRCA1 over the carboxyl terminal half of the protein. The aim of this study was to determine whether TopBP1 is aberrantly expressed in breast cancer.

METHODS AND RESULTS

Sixty-one breast carcinomas from an unselected consecutive patient cohort were studied along with 12 samples of breast tissue from cosmetic breast reduction surgery; these were analysed immunohistochemically for TopBP1 expression using a rabbit polyclonal antibody. This antibody was validated in immunoprecipitation and immunofluorescence experiments. Immunohistochemical analysis demonstrated that TopBP1 was expressed almost exclusively in the nuclei of the normal breast epithelium. However, in a significant number of breast carcinomas TopBP1 was aberrantly expressed, as it was detected in the cytoplasm and nucleus of some tumours and exclusively in the cytoplasm of others. In two out of 61 carcinomas investigated, no TopBP1 expression was detected.

CONCLUSIONS

For the first time this report demonstrates aberrant expression of the TopBP1 protein in breast carcinoma. We propose TOPBP1 as a breast cancer susceptibility gene.

Metastatic recurrence of early-stage colorectal cancer is linked to loss of heterozygosity on chromosomes 4 and 14q

OBJECTIVE

To investigate the prognostic value for loss of heterozygosity (LOH) of chromosomes 4 and 14q in early-stage colorectal cancer (CRC).

METHODS

A total of 70, largely microsatellite stable, tumours and their corresponding normal mucosa were subjected to microdissection and analysed for LOH at chromosomes 4 and 14q by using 13 highly polymorphic microsatellite markers. LOH was correlated with the survival of the patients, using univariate, multivariate and Kaplan-Meier's survival curves.

RESULT

LOH at D4S2935, D4S1579 and D4S1595 on chromosome 4 was significantly associated with metastatic recurrence of early-stage CRC. For chromosome arm 14q, two minimal regions of deletion were associated with metastatic recurrence and mapped to neighbouring markers D14S275/D14S49 at 14q12-13 and D14S65/D14S250 at 14q32. High-level loss (loss of five to eight of the informative microsatellite markers) on both chromosomes 4 and 14q, to be an independent prognostic indicator in early-stage CRC was shown by multivariate analysis.

CONCLUSION

Determining the LOH of chromosomes 4 and 14q and their extent in primary tumours of patients with early-stage CRC may constitute a molecular signature of metastatic recurrence. This may be achieved if new finding sheds light on the treatment of this subgroup of patients that have been largely ignored.

An elevated C-reactive protein concentration, prior to surgery, predicts poor cancer-specific survival in patients undergoing resection for gastro-oesophageal cancer

There is increasing evidence that the presence of an ongoing systemic inflammatory response is associated with poor outcome in patients undergoing resection for a variety of tumours. The aim of the present study was to examine the relationship between clinico-pathological status, preoperative C-reactive protein concentration and cancer-specific survival in patients undergoing resection for gastro-oesophageal cancer. One hundred and twenty patients attending the upper gastrointestinal surgical unit in the Royal Infirmary, Glasgow, who were selected for potentially curative surgery, were included in the study. Laboratory measurements of haemoglobin, white cell, lymphocyte and platelet counts, albumin and C-reactive protein were carried out at the time of diagnosis. All patients underwent en-bloc resection with lymphadenectomy and survived at least 30 days following surgery. On multivariate analysis, only the positive to total lymph node ratio (hazard ratio (HR) 2.02, 95% confidence interval (CI) 1.44–2.84, P<0.001) and preoperative C-reactive protein concentration (HR 3.53, 95% CI 1.88–6.64, P<0.001) were independent predictors of cancer-specific survival. The patient group with no evidence of a preoperative systemic inflammatory response (C-reactive protein ⩽10 mg l⁻¹) had a median survival of 79 months compared with 19 months in the elevated systemic inflammatory response group (P<0.001). The results of the present study indicate that in patients selected to undergo potentially curative resection for gastro-oesophageal cancer, the presence of an elevated preoperative C-reactive protein concentration is an independent predictor of poor cancer-specific survival.

IHC for Her2 with CBE356 antibody is a more accurate predictor of Her2 gene amplification by FISH than HercepTest in breast carcinoma

BACKGROUND

Her2 (c-erbB-2/neu) overexpression in breast carcinoma predicts response to the anti-Her2 monoclonal antibody, trastuzumab, and is associated with a poor prognosis. When considering patients for trastuzumab treatment, Her2 protein expression is measured by imunohistochemistry (IHC) and, where staining is equivocal, by fluorescence in situ hybridisation (FISH) detection of Her2 gene amplification.

AIMS

To compare IHC using CBE356 with IHC using the Food and Drug Administration approved HercepTesttrade mark.

METHODS

CBE356 and HercepTest were analysed using 167 FISH characterised breast carcinomas. Immunohistochemical expression of Her2 was measured semiquantitatively. Sensitivity, specificity, predictive values, and overall accuracy were calculated for both IHC methods using gene amplification by FISH as the end point, and IHC and FISH assays were tested in Kaplan-Meier survival analysis.

RESULTS

The accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of CBE356 positive (2+ and 3+) cases were 94%, 89%, 95%, 84%, and 97%, respectively, and of HercepTest positive (2+ and 3+) cases were 91%, 66%, 98%, 92%, and 91%, respectively. A positive result with CBE356, HercepTest, or FISH was associated with significantly decreased overall survival (log rank p = 0.005, p = 0.0017, and p = 0.0005, respectively).

CONCLUSIONS

Positive IHC staining for Her2 using CBE356 is 3% more accurate and 23% more sensitive at predicting Her2 gene amplification by FISH than positive staining with HercepTest. Negative IHC using CBE356 antibody is 6% more likely to represent a truly negative result than negative staining with HercepTest. Overall, CBE356 was a more accurate predictor of Her2 gene amplification by FISH than HercepTest.

Diagnosis of oesophageal cancer by detection of minichromosome maintenance 5 protein in gastric aspirates

Symptomatic oesophageal cancer is usually advanced and the prognosis poor. Lethality of symptomatic oesophageal cancer has motivated screening for these diseases earlier in their evolution, but reliable methods for early diagnosis remain elusive. We have demonstrated that dysregulated expression of minichromosome maintenance (MCM) proteins 2-7 is characteristic of early epithelial carcinogenesis, and that these key DNA replication initiation factors can be used as diagnostic markers for cervical and genito-urinary tract cancer. In this study, we investigated whether minichromosome maintenance protein 5 (Mcm5) can be used to detect oesophageal cancer cells in gastric aspirates. Two monoclonal antibodies raised against His-tagged human Mcm5 were used in a time-resolved immunofluorometric assay to measure Mcm5 levels in cells isolated from gastric aspirates of 40 patients undergoing gastroscopy for suspected or known oesophageal carcinoma or symptoms of dyspepsia. The test discriminated with high specificity and sensitivity between patients with and without oesophageal cancer (85% sensitivity (95% confidence interval (CI)=62-97%), 85% specificity (CI=66-96%)), as demonstrated by the large area under the receiver operating characteristics curve (0.93 (95% CI=0.85-0.99)). Elevated levels of Mcm5 in gastric aspirates are highly predictive of oesophageal cancer. This simple test for oesophageal cancer is readily automated with potential applications in primary diagnosis, surveillance and screening.

A role for BRCA1 in sporadic breast cancer

To test the hypothesis that altered expression of BRCA1 protein may play an important role in sporadic breast cancer development, 50 randomly selected primary breast cancers (frozen sections, 5 years' median follow-up) were immunolabelled with two monoclonal BRCA1 antibodies (MS110 and MS13). MS110 labelling was exclusively nuclear showing no relation to outcome or tumour pathology. Western blotting demonstrated crossreactivity, suggesting antibody nonspecificity. MS13 labelling was predominantly cytoplasmic. Intense labelling predicted decreased overall survival (P=0.012), disease-free survival (P=0.029), oestrogen receptor negativity (P=0.0004) and c-erbB-2 overexpression (P=0.006). Western blotting detected a 110 kDa molecule consistent with BRCA1 delta11b splice variant. BRCA1 protein is postulated to function as a tumour suppressor. We demonstrate cytoplasmic localisation in sporadic breast cancer suggesting excess delta11b splice variant production, reduced production of full-length BRCA1 and thus postulate reduced tumour suppressor activity. BRCA1 protein appears to have a significant role in both sporadic and hereditary breast cancers.

Human squamous cell carcinomas lose a mortality gene from chromosome 6q14.3 to q15

Normal human keratinocytes possess a finite replicative lifespan. Most advanced squamous cell carcinomas (SCCs), however, are immortal, a phenotype that is associated with p53 and INK4A dysfunction, high levels of telomerase and loss of heterozygosity (LOH) at several genetic loci, suggestive of the dysfunction of other mortality genes. We show here that human chromosome 6 specifically reduces the proliferation or viability of a human SCC line, BICR31, possessing LOH across the chromosome. This was determined by an 88% reduction in colony yield (P<0.001), following the reintroduction of an intact normal chromosome 6 by monochromosome transfer. Deletion analysis of immortal segregants using polymorphic markers revealed the loss of a 2.9 Mbp interval, centred on marker D6S1045 at 6q14.3-q15, in 6/19 segregants. Crucially, allelic losses of this region were not identified in control hybrids constructed between chromosome 6 and the BICR6 SCC cell line that is heterozygous for chromosome 6 and which showed no reduction in colony formation relative to the control chromosome transfers. This indicates that the minimally deleted region at D6S1045 is not the result of fragile sites, a recombination hot spot, or a feature of the monochromosome transfer technique. LOH of D6S1045 was found in 2/9 immortal SCC lines and was part of a minimally deleted region of line BICR19. Furthermore, allelic imbalance, consistent with LOH, was detected in 3/17 advanced SCCs of the tongue. These results suggest the existence of a suppressor of SCC immortality and tumour development at chromosome 6q14.3-q15, which is important to a subset of human SCCs.

Chromosome 17 aneusomy is associated with poor prognostic factors in invasive breast carcinoma

Aberrations of chromosome 17 are common in breast cancer. Fluorescence in situ hybridization (FISH) enables gene or chromosome copy number to be assessed in situ in archival tissues and related to morphology and clinical outcome. In this study direct labeled DNA probes for the chromosome 17 alpha satellite and the HER2/neu gene were applied simultaneously to 5 micron sections of 214 formalin-fixed paraffin-embedded invasive primary breast carcinomas. A high proportion (54%) of invasive breast carcinomas displayed aneusomy of chromosome 17. Polysomy 17 correlated with multiple copies of HER2/neu (p = < 0.001), but not with HER2/neu amplification. Eighty-six patients without HER2/neu amplification had aneusomy 17. Fifty-eight of the 86 patients that had aneusomy 17 had high HER2/neu copy number. Twelve patients with normal copy number for chromosome 17 had amplification of HER2/neu and 30 patients had amplification of HER2/neu with aneusomy 17. Aneusomy 17 was associated with grade 3 carcinoma (p = 0.008), ER negativity (p = 0.0032) and a Nottingham prognostic index of greater than 5.4 (p = 0.039) but was not associated with survival by univariate analysis. In conclusion, the determination of chromosome 17 copy number should be incorporated in assessment of HER2/neu status, as this will give an accurate measure of amplification of HER2/neu and may also be helpful in determining suitability for breast carcinoma trials.

Stages on the way to breast cancer

Pathways to breast cancer have been difficult to define using morphology alone. Although epithelial hyperplasia of usual type (HUT) is associated with moderately elevated breast cancer risk, molecular evidence placing it in a cancer precursor pathway is not clear-cut. Recent evidence suggests that small numbers of cytokeratin 5/6 positive cells are precursors of separate lineages which acquire either cytokeratin 8/18/19 or smooth muscle actin (SMA) and cytokeratin 14 on separate pathways to fully differentiated epithelial and myoepithelial phenotypes (and may ultimately lose cytokeratin 5/6 expression). Immunohistochemistry shows that most HUT have a mixed precursor phenotype resembling normal breast with co-expression of cytokeratin 5/6, 8/18/19 and SMA, in contrast to atypical ductal hyperplasia (ADH) and ductal carcinoma in situ which typically have a 'mature' luminal phenotype positive for cytokeratin 8/18/19 but lacking cytokeratin 5/6 expression. While this supports the idea of a biological discontinuity between HUT and ADH/DCIS, caution is called for in the diagnostic use of these reagents until greater experience has been accumulated, and other published data do show features of HUT intermediate between normal breast lobules and ADH.

Mucinous breast carcinoma

A case of mucinous breast carcinoma is discussed in which several unusual features in presentation and difficulties in management are addressed.

Genetic aberrations of NAT2 and chromosome 8: their association with progression in transitional cell carcinoma of the urinary bladder

INTRODUCTION/OBJECTIVE

N-acetyltransferase 2 (NAT2), mapped to 8p22, is a polymorphic enzyme which metabolizes aromatic amines. Loss of heterozygosity of 8p22 is associated with an increased risk of bladder cancer. This study evaluated NAT2 and chromosome 8 in sequential tumours from bladder cancer patients to determine if NAT2 alterations increase the risk of progression.

MATERIALS AND METHODS

Thirty-seven sequential carcinomas from 19 patients were assessed using fluorescence in situ hybridization.

RESULTS

Five carcinomas showed loss of NAT2; 4 of these were from pTa/pT1 tumours. Polysomy 8 was observed in 4 of 14 (29%) primary carcinomas (pTa/pT1), in 4 of 12 (33%) pTa/pT1 recurrences, and in 90% (9/10) of the detrusor muscle invasive tumours (pT2+). 6 of 8 (75%) locally invasive tumours with polysomy 8 were from patients who subsequently developed disease progression (pT2+). In total, 13.5% (5/37) of the carcinomas were abnormal for NAT2, and 46% (17/37) were abnormal for chromosome 8 copy number. Polysomy 8 was associated with high grade (p = 0.01) and stage (p = 0.03) and disease progression (p = 0.03).

CONCLUSION

Whilst there does not appear to be an association between loss of NAT2 and risk of progression in transitional cell carcinoma, the high rate of polysomy of chromosome 8 implies that other genes on this chromosome significantly influence disease progression.

Aberrant expression of minichromosome maintenance proteins 2 and 5, and Ki-67 in dysplastic squamous oesophageal epithelium and Barrett's mucosa

BACKGROUND

Minichromosome maintenance (Mcm) proteins are essential for eukaryotic DNA replication, and their expression implies potential for cell proliferation. Expression is dysregulated in dysplastic states but data for oesophageal squamous mucosa and Barrett's mucosa have not been published.

AIM

To test the hypothesis that Mcm proteins are downregulated together with the proliferation marker Ki-67 in differentiating epithelial compartments of non-dysplastic squamous and Barrett's epithelium, and that this process does not occur in dysplastic mucosae. METHODS AND CASES: Forty five patients with Barrett's oesophagus included 20 with glandular dysplasia (10 low grade, eight high grade, two both, and four with invasive adenocarcinoma). Twenty five other patients included 12 with oesophageal squamous dysplasia (three low grade, six high grade, three both, and four with invasive squamous carcinoma). Formalin fixed paraffin embedded tissue sections from biopsy series and resections were immunostained using antibodies to Mcm2, Mcm5, and Ki-67. Percentage of nuclei positive for Mcm2, Mcm5, and Ki-67 was estimated and scored from 0 to 6 as: 0, none +; 1, <10%+; 2, 10-30%+; 3, 30-70%+; 4, 70-90%+; 5, >90%+; 6, all+. Four separate epithelial strata were scored: in squamous epithelium the basal layer and thirds to the surface, in Barrett's mucosa the luminal surface, upper and lower crypt, and deep glands.

RESULTS

In non-dysplastic squamous epithelium and Barrett's mucosa, high level expression of Mcm2, Mcm5, and Ki-67 proteins was largely confined to the proliferative compartments and downregulated in differentiated compartments. Expression persisted up to the mucosal surface in dysplastic squamous epithelium and Barrett's mucosa.

CONCLUSIONS

Persistent expression of Mcm2, Mcm5, and Ki-67 proteins in luminal compartments of dysplastic oesophageal squamous epithelium and dysplastic Barrett's mucosa may be diagnostic markers and imply disruption of cell cycle control and differentiation in these dysplastic epithelia.

Loss of heterozygosity on chromosomes 11 and 17 are markers of recurrence in TCC of the bladder

Approximately 2/3 of patients diagnosed with superficial transitional cell carcinoma of the urinary bladder (TCC) will recur within 2 years. Loss of chromosome 9 and loss of heterozygosity (LOH) at 9q34 in index TCCs identify a subset of patients at high risk of recurrence. This study explores genetic alterations on chromosomes 4, 8, 11 and 17 as predictors of recurrence. A total of 109 carcinomas were investigated at 26 loci. DNA was extracted from microdissected archival normal/tumour tissue and was analysed for loss of heterozygosity (LOH). Fluorescent PCR was performed and genotyping carried out on a Perkin Elmer ABI377 sequencer. LOH of D11S490 or D17S928 was significantly more frequent in index carcinomas of patients who experienced recurrence compared to those with no recurrence (P = 0.004 and 0.019 respectively). These results suggest that loss of these regions is associated with recurrence of TCC. Further investigation is required to identify genes in these regions, which might be responsible for driving recurrence in TCC of the urinary bladder.

Evaluating HER2 amplification and overexpression in breast cancer

The development of Herceptin (Trazumatab) makes testing for HER2 status important for choosing optimal therapy in breast cancer. This study addresses the precision, accuracy, and reproducibility of HER2 assays. HER2 was assessed retrospectively by immunohistochemistry (IHC) with Dako 'Herceptest', by IHC with the monoclonal antibody CB11, and by fluorescence in situ hybridization (FISH, PathVysion), in a series of 216 formalin-fixed breast carcinomas including 191 for which quantitative HER2 data from radioimmunohistochemistry (Q-IHC) were available. All tests were scored independently by two observers. Positivity rates varied between Herceptest (12.6%), FISH (19.4%), and CB11 IHC (28.5%). Kappa values showed that IHC-based tests were more susceptible to inter-observer variation (kappa=0.67 and 0.74 for Herceptest and CB11, respectively) than FISH (kappa=0.973). Overall test accuracy (see the Materials and methods section) for CB11 IHC (83.8%) was lower than Herceptest (87.4%) or FISH (93.2%). FISH predicted p185 HER2 overexpression (determined by Q-IHC) better (concordance index C.Ind. 0.90) than CB11 IHC (C.Ind.=0.85) or Herceptest (C.Ind.=0.81). Of 42 cases with gene amplification by FISH, 67% were positive in the Herceptest (2+ or 3+) vs. 83% with CB11. Of 174 cases negative by FISH, 96% were negative in the Herceptest and 68% with CB11. In conclusion, FISH is the most accurate, reproducible, and precise predictor of HER2 overexpression in routine diagnostic laboratories.

Expression of telomerase RNA in oesophageal and oral cancer

BACKGROUND

Telomerase is detectable in 85% of cancers, but is largely repressed in normal tissues. Human telomerase RNA (hTR) inhibition is a promising anti-cancer strategy, but requires differential expression between malignant and normal tissue.

METHOD

Archival paraffin sections from 48 oral squamous cell carcinomas (SCC) (23 floor of mouth, 25 tongue) and 56 oesophageal carcinomas (41 SCC, 15 adenocarcinomas) were evaluated for hTR expression using a radiolabelled riboprobe. Results were compared with expression in controls and adjacent histologically normal tissue. Statistical analysis was by the chi2 test.

RESULTS

hTR was detectable in 76% of oral SCC overall (floor of mouth 65%, tongue 88%, P=0.61), and in 54% of oesophageal cancers (SCC 51%, adenocarcinoma 60%, P=0.56). Detectable hTR expression was significantly more frequent in oral SCC than oesophageal SCC (P=0.01). hTR expression was only detected in normal tissue at low levels in basal squamous epithelium. There was agreement of hTR expression between 8/9 surgically excised carcinomas and their initial diagnostic biopsies.

CONCLUSION

Tumour-specific hTR expression confirms hTR inhibition as a possible therapeutic strategy in some if not all oesophageal and oral cancers. Generally concordant hTR status between biopsy and resection suggest that biopsy may have a role in selecting candidates for telomerase inhibition therapy.

What the clinician needs from the pathologist: evidence-based reporting in breast cancer

Histopathology has a vital role in determining breast cancer management and pathologists must be part of the clinical team. Carcinoma size, grade, and especially lymph node status remain the best available prognostic factors. Metastatic carcinoma in axillary nodes is more important than any other prognostic factor presently available. ER status is an important predictor of response to endocrine manipulation, but its independent prognostic significance, and that of micrometastatic disease, circulating carcinoma cells and other molecular factors, even well-studied ones such as HER2 status, are less clear. Pathology is the first clinical speciality to subject its practice to rigorous scientific analysis, and it has stood up well. However, workers without appropriate experience in Pathology or scientific design have created difficulties by undertaking poorly planned studies with ill-defined end-points, lacking appropriate quality control. New analytical techniques and therapeutic targets make it essential that we learn from past mistakes and integrate pathologists into the research teams pursing clinical trials and the assessment of new bio-markers. Without this, input resource will be wasted on false leads that could have been curtailed. Morphology alone will not be enough to select patients likely to benefit in trials of new therapies, but selection 'tests' must be appropriate. The confusion of tests for selection of patients to receive Herceptin shows what happens when this process fails. Much of the microarray data being put into data-bases has no quality control, and meta-analysis of this data will produce even more conflict than the clinical trials. This can be avoided, as the ability to standardise is available.

Scoring nuclear pleomorphism in breast cancer

AIMS

Nuclear grade has equal weight with mitotic index and acinus formation in grading breast cancer, but criteria for its assessment are less well defined. This study examines consistency of nuclear grading in breast cancer and whether improved nuclear grading criteria are required.

METHODS AND RESULTS

Photographic prints of haematoxylin-eosin sections of 100 unselected symptomatic breast cancers were circulated to histopathologists who assigned each carcinoma a nuclear pleomorphism score on a linear analogue scale 0-100 (0-33 equating to nuclear pleomorphism grade 1, 34-66 to grade 2, and 67-100 to grade 3). Seventeen histopathologists completed the exercise, including 11 breast specialists. While kappa scores for the implied nuclear grades indicated 'moderate' or 'good' agreement between individuals and the group as a whole, seven pathologists allocated analogue scores significantly lower than the median score allocated by the group to each case, while five allocated significantly higher scores. The range was from analogue scores 11.3 units lower on average than the median (assigning 27% of carcinomas nuclear grade 1, 60% grade 2, and 13% grade 3) to scores 7.5 units higher on average than the median (assigning only 2% carcinomas nuclear grade 1, 46% grade 2, and 52% grade 3). Five of six non-specialists allocated scores significantly lower than the group medians but only two of 11 specialists did so (P=0.018).

CONCLUSIONS

Systematic differences between pathologists in scoring nuclear pleomorphism in breast cancer potentially contribute to differences in allocating overall grade and confirm the need for improved nuclear grading criteria. Specialists tend to allocate higher pleomorphism scores than non-specialists.

Clonal origins of human breast cancer

Tumours are usually considered as the clonal progeny of single transformed cells. An X-chromosome inactivation assay has been applied to exploring clonal relationships in human breast cancer. Analysis of X-inactivation in DNA extracted from microdissected in situ and invasive breast carcinoma by Hpa II restriction and polymerase chain reaction (PCR) of the androgen receptor exon I CAG polymorphism confirmed monoclonality in 105/133 samples of carcinoma cells from 31/32 informative breast cancers. Clonality was identical in seven cases between in situ and invasive carcinoma. Unexpectedly, 4 of 12 cancers (33%) with two or more monoclonal samples available were mosaic (polyclonal) in respect of X-chromosome inactivation between separate morphologically homogeneous tumour cell samples. Concordant clonality supports a common clonal origin of in situ and invasive breast cancers, but frequent apparently mosaic X-inactivation in breast cancer cannot be explained by non-tumour cell contamination. It is concluded that these carcinomas may be genuinely multiclonal. Possible mechanisms of multiclonality include simultaneous transformation of cell groups straddling X-chromosome inactivation patch boundaries, tumour-initiating mutations prior to X-inactivation, or recruitment of bystander stem cells by DNA transfer from necrotic or apoptotic tumour cells. Collision of independent cancers appears implausible at this frequency. Further studies using independent analytical techniques are required to test the important possibility that a significant proportion of mammary carcinomas are not monoclonal.

Lack of telomerase RNA gene hTERC expression in alternative lengthening of telomeres cells is associated with methylation of the hTERC promoter

The immortal phenotype of most human cancers is attributable to telomerase expression. However, a number of immortal cell lines and tumors achieve telomere maintenance in the absence of telomerase via alternative mechanisms known as ALT (alternative lengthening of telomeres). Here we show that the promoter of the telomerase RNA gene (hTERC) is methylated in three of five ALT cell lines and is associated with a total absence of hTERC expression in the three lines. Treatment with 5-azacytidine in combination with trichostatin A resulted in partial demethylation of the hTERC promoter and expression of the gene. Partial methylation was detected in tumors (5%) and in immortal cell lines (27%). Cell lines with partial methylation express hTERC. Only in ALT cell lines does there appear to be a strong correlation between hTERC promoter hypermethylation and lack of hTERC expression.

Aneusomy of chromosomes 7 and 17 predicts the recurrence of transitional cell carcinoma of the urinary bladder

OBJECTIVE

To determine if changes in chromosome 7 and 17 copy number can be used to predict recurrence in patients with primary noninvasive (pTa) or superficially invasive (pT1) transitional cell carcinoma (TCC) of the urinary bladder.

PATIENTS AND METHODS

Tissue specimens for 129 tumours from 52 patients (38 men and 14 women) with pTa/pT1 TCC at first diagnosis were retrieved from pathology archives. All patient notes were accessed and disease outcome documented for superficial (pTa/pT1) recurrence or progression to detrusor muscle invasion (>/= pT2). The tumours were examined for chromosomal copy number of chromosomes 7 and 17 using fluorescence in situ hybridization (FISH) with chromosome-specific probes. The copy number of chromosomes 7 and 17 was determined in interphase nuclei on intact 6 microm tissue sections.

RESULTS

Aneusomy of chromosomes 7 and 17 was detected in the index primary tumours of 10 of 32 (31%) patients with subsequent recurrent disease. No aneusomy for these chromosomes was detected in primary tumours from 20 patients with no detect-able recurrence (P = 0.0082). The relative risk of recurrence was 3.62 times greater (95% confidence interval 1.6-8.1, Cox's multiple regression P = 0.0019) for patients with chromosomal aneusomy in primary TCC. Neither stage nor grade of the primary tumours was associated with recurrence in these patients, nor was there a significant association with increased grade (G2/3) or stage (>/= pT2) at recurrence.

CONCLUSION

These results suggest that the measurement of aneusomy by FISH, using markers for chromosomes 7 and 17, predict recurrence in a subgroup of patients with pTa/pT1 tumours at presentation. This finding may offer a new objective and quantitative test for patients destined to recur.

Molecular pathology and future developments

There has already been a 'molecular' revolution in pathology. Demonstrating transcription of specific single genes or small gene sets and their protein products by in situ hybridisation and immunocytochemistry is routine in diagnostic and experimental pathology. A perhaps-greater revolution is imminent with the application of more recently established and emergent technologies in pathology. These include new approaches to polymerase chain reaction (PCR); simultaneous studies of multiple genes and their expression using oligonucleotide and cDNA arrays; serial analysis of gene expression (SAGE); expressed sequence tag (EST) sequencing, subtractive cloning and differential display; high-throughput sequencing; comparative genomic hybridization, multiplex fluorescence in situ hybridisation (FISH) (spectral karyotyping); reverse chromosome painting; knockout and transgenic organisms; laser microdissection and micro-machining; and new methods in bio-informatics, 'data mining' and data visualisation. Molecular methods will profoundly change diagnosis, prognosis and treatment targeting in oncology and elucidate fundamental mechanisms of neoplastic transformation. Individual susceptibility to specific diseases will become assessable and screening will be refined. The new molecular biology will be most fruitful in partnership with classical approaches to pathology: the expectation that molecular methods alone will answer all pathological questions is unrealistic. A further challenge for the biomedical community in the 'genome era' will be to ensure that the benefits of these sophisticated technologies are enjoyed globally.

Predicting the future: a critical appraisal of cancer prognosis studies

Many studies have attempted to define useful prognostic and predictive factors in cancer but few have achieved acceptance in clinical practice because of methodological weaknesses. These include failure to test clearly formulated hypotheses, inadequate sample size, inappropriate multiple significance testing, arbitrary definition of patient groups, inadequately reproducible assays, and failure to verify prognostic factors with data independent of the data which suggested the original hypothesis. This unsatisfactory situation will persist until critical attention is routinely paid to study design and prospective validation of supposed prognostic and predictive factors, without which classical approaches will be suboptimally exploited and the flood of data from new molecular technologies will not be used effectively. We propose that prognostic factors should be evaluated in three phases: I, assay definition; II, retrospective testing; III, prospective testing, ideally as a designed part of clinical trials.

Chromosomal aberrations in transitional cell carcinoma that are predictive of disease outcome are independent of polyploidy

OBJECTIVE

To determine whether aneusomy for chromosomes 7, 9 and 17 (reported to predict recurrence in up to 65% of patients with superficial transitional cell bladder cancer and thus providing the opportunity for early and effective treatment) reflects specific genetic events on these chromosomes or merely wider unspecific genetic damage to the cell, e.g. that increased copy numbers for 7 and 17 reflect tumour polyploidy.

MATERIALS AND METHODS

The study comprised 25 primary tumours; 6 microm thick sections from formalin-fixed and paraffin-embedded tumours were analysed. Chromosome copy numbers were determined by fluorescence in situ hybridization (FISH) using pericentromeric probes for chromosomes 7, 8, 9, 10, 11 and 17. A minimum of 200 nuclei per tumour area were scored by two independent observers.

RESULTS

Eight of the 25 tumours examined (32%) showed no evidence of chromosomal abnormalities as detected by FISH for any chromosomes analysed. Twelve tumours (48%) showed abnormalities for one or two chromosomes, five tumours (20%) showed abnormalities for multiple chromosomes and one tumour showed abnormalities for all chromosomes analysed, suggestive of polyploidy.

CONCLUSIONS

Chromosomal abnormalities predictive of recurrence occur largely in the absence of other gross chromosomal lesions. In a small proportion of cases other chromosomes are affected, but this is almost always distinct from tumour polyploidy.

Therapeutic implications of the sentinel lymph node in breast cancer

Immunohistochemistry of histologically negative axillary lymph nodes in breast-cancer patients resulted in upstaging of the sentinel lymph node in eight (14%) of 52 patients. The resulting information altered clinical management in six of these patients. Thus, this technique may affect clinical decision-making in breast-cancer patients.

Is small cell lung cancer the perfect target for anti-telomerase treatment?

Small cell lung cancer (SCLC) is common in men and women, has a very poor prognosis, and is therefore a major cause of premature mortality. As such, any prospects for improved therapy are of great significance. The promise of telomerase as a therapeutic target is now close to realization with extremely encouraging preclinical studies aimed at the RNA component (hTR) of telomerase. The rational integration of telomerase therapeutics into clinical trials will therefore require tumours to be well characterized for hTR expression. Despite the large number of cancer types now characterized for telomerase or telomerase component gene expression, only a handful of SCLC samples have been analysed. Given the major clinical problem with treating SCLC, we specifically set out to address the issue of hTR expression in neuroendocrine tumours. Our study covers 91 pulmonary neuroendocrine tumours (62 SCLC and 29 carcinoid tumours). We present data to show that upregulation of the RNA component of telomerase occurs in 98% of human SCLCs. Interestingly, the less aggressive carcinoid tumours of the lung had a significantly lower frequency of hTR expression (P < 0.01). Importantly, we compare hTR expression in this series to the well characterized biological targets p53 and BCL2, and show hTR to be expressed more frequently. Therapies directed at the RNA component of human telomerase are in active development and these data show SCLC to be a prime target for such therapies.

Accuracy of intraoperative frozen-section analysis of axillary nodes

The Editors welcome topical correspondence from readers relating to articles published in the Journal. Letters should be no more than 250 words in length and should be typed on A4-sized paper in double spacing.

Comparative genomic hybridization analysis of primary colorectal carcinomas and their synchronous metastases

We have analyzed 26 tumors from 12 patients with metastatic colorectal adenocarcinoma by comparative genomic hybridization (CGH). Primary tumors and their lymph node metastases from five Dukes' C patients as well as primary tumors and their liver metastases from seven Dukes' D patients were used to assess the extent of genetic differences between primary and secondary colorectal carcinomas from the same patients, to calculate the degree of clonal divergence and genetic heterogeneity in metastatic colorectal cancer, and to determine the differences in genetic imbalances between Dukes' C and D stage tumors. We show that the same genetic aberrations were frequently found in the primary tumors and their metastases. However, metastases often contained genetic aberrations not found in the corresponding primary tumors. The comparison of Dukes' stages C and D revealed genetic aberrations common to both. However, reduced copy number of chromosome arm 17p (5/5 vs. 0/7; P = 0.001) was significantly associated with Dukes' stage C and lymph node metastases, while increased copy number of chromosome arms 6p (6/7 vs. 0/5; P = 0.007) and 17q (5/7 vs. 0/5; P = 0.027) was associated more with Dukes' stage D and liver metastases. Our results established a repertoire of chromosomal alterations associated with metastatic colorectal cancer and suggest that Dukes' C (lymph node metastasis) tumors are not always simply an earlier stage of Dukes' D (liver metastasis) tumors and, thus, in some instances at least, they are distinct forms of the disease.

Comparative genomic hybridization analysis of primary colorectal carcinomas and their synchronous metastases

We have analyzed 26 tumors from 12 patients with metastatic colorectal adenocarcinoma by comparative genomic hybridization (CGH). Primary tumors and their lymph node metastases from five Dukes' C patients as well as primary tumors and their liver metastases from seven Dukes' D patients were used to assess the extent of genetic differences between primary and secondary colorectal carcinomas from the same patients, to calculate the degree of clonal divergence and genetic heterogeneity in metastatic colorectal cancer, and to determine the differences in genetic imbalances between Dukes' C and D stage tumors. We show that the same genetic aberrations were frequently found in the primary tumors and their metastases. However, metastases often contained genetic aberrations not found in the corresponding primary tumors. The comparison of Dukes' stages C and D revealed genetic aberrations common to both. However, reduced copy number of chromosome arm 17p (5/5 vs. 0/7; P = 0.001) was significantly associated with Dukes' stage C and lymph node metastases, while increased copy number of chromosome arms 6p (6/7 vs. 0/5; P = 0.007) and 17q (5/7 vs. 0/5; P = 0.027) was associated more with Dukes' stage D and liver metastases. Our results established a repertoire of chromosomal alterations associated with metastatic colorectal cancer and suggest that Dukes' C (lymph node metastasis) tumors are not always simply an earlier stage of Dukes' D (liver metastasis) tumors and, thus, in some instances at least, they are distinct forms of the disease.

Structural differences between valine-12 and aspartate-12 Ras proteins may modify carcinoma aggression

Recent evidence associates the codon 12 valine-for-glycine (G12V) mutant Ki-Ras protein with higher stage and increased lethality of colorectal carcinomas, while the codon 12 aspartate-for-glycine (G12D) Ras mutation shows no such association. Several observations may be relevant to this phenomenon. First, GTPase activity of G12V Ras is one-quarter that of G12D Ras and one-tenth that of wild-type (WT) Ras. Second, binding of the GTP analogue GppNp to G12D Ras is 8-fold weaker than its binding to G12V or WT Ras and crystal structures indicate that electrostatic repulsion between the carboxylate group of the G12D Asp-12 side-chain and the gamma phosphate of the bound nucleotide may make GTP binding to G12D Ras weaker even than that of GppNp. It is proposed that this lowering of affinity for GTP allows G12D Ras an escape from the oncogenic GTP-bound state, whereas GTP tightly bound to G12V mutant Ras generates a more persistent, potentially oncogenic, signal. Structural comparisons also suggest that differences between the Switch I (effector) region of G12D and G12V Ras could modify interactions with downstream signalling molecules such as Raf-1, neurofibromin, and phosphatidylinositol 3-hydroxy-kinase. Other differences between the G12D and G12V mutant Ras proteins include a lower affinity of the GTPase activating protein GAP for G12V than for G12D or WT Ras; but, as both G12D and G12V Ras are refractory to GTPase activation by GAP binding, this may be less significant. These studies complement experimental data showing that such Ras mutations differ in their effects in vitro and in vivo and, with recent data indicating heterogeneity of ras mutation in colorectal carcinomas and other tumours, make it plausible that codon 12 Ras mutations differ in carcinogenic potential and prognostic significance.

Oestrogen receptor status predicts local recurrence following breast conservation surgery for early breast cancer

AIMS

To evaluate factors predicting locoregional recurrence in patients treated for early breast carcinoma by breast conservation surgery with or without radiotherapy.

METHODS

A retrospective study of 256 patients was carried out, with special emphasis on the role of oestrogen receptor status. Other parameters studied included age, menopausal status, size of primary tumour, tumour type, axillary nodal status and adjuvant therapy.

RESULTS

Multivariate analysis showed the following parameters to be independent predictors of locoregional recurrence. Radiotherapy (57% reduction in hazard, P = 0.004): expression of oestrogen receptors (52% reduction in hazard, P = 0.008); tamoxifen therapy (46% reduction in hazard, P = 0.023); tumour size (40% increase in hazard per cm, P<0.001).

CONCLUSIONS

Within this study, lack of oestrogen receptor expression was a strong independent factor associated with a higher rate of locoregional recurrence. This may be of value in selecting a group of patients less suitable for breast conservation surgery.

Sentinel node localization in patients with breast cancer

BACKGROUND

Intraoperative lymphatic mapping and identification of the first draining lymph node (the sentinel node) may allow some patients with breast cancer to avoid the morbidity of formal axillary clearance. The aim of this pilot study was to establish the reliability of the technique in predicting axillary node status.

METHODS

Sixty-eight consecutive patients with breast cancer, 38 undergoing mastectomy and 30 wide local excision, were included. Some 2-4 ml of 2.5 per cent Patent Blue dye was injected into adjacent breast tissue on the axillary side of the primary tumour. After 5-10 min, the axilla was explored. Blue-stained lymphatics were dissected to the sentinel node, which was removed for frozen-section examination, followed by routine histology. Formal axillary dissection was then completed.

RESULTS

A sentinel lymph node was identified successfully in 56 (82 per cent) of 68 patients. Histology of the sentinel node accurately predicted axillary node status in 53 (95 per cent). There were three false negatives (5 per cent). In each case, only a single non-sentinel node was tumour positive. Sensitivity and specificity were 83 and 100 per cent respectively.

CONCLUSION

This technique would allow a selective policy of formal axillary dissection in only node-positive patients; however, further experience and refinement are needed.

Heterogeneity of mutant versus wild-type Ki-ras in primary and metastatic colorectal carcinomas, and association of codon-12 valine with early mortality

The point mutations occurring in codons 12 and 13 of Ki-ras in 78 patients with colorectal carcinoma (31 Dukes' A and B, 21 Dukes' C, and 26 Dukes' D) have been determined by allele-specific oligonucleotide hybridization and sequencing. Duplicate samples of invasive primary carcinoma, adjacent normal tissue, and available lymph node and liver metastases from the same patients were microdissected from paraffin sections. There were no differences in the mutation rate between primary carcinomas and secondary deposits: 26 of 78 (33 per cent) primary carcinomas, 10 of 32 (31 per cent) lymph node metastases, and 10 of 26 (38 per cent) liver metastases. Multiple sampling revealed frequent heterogeneity within carcinomas: 9 of 26 primaries with Ki-ras mutations also contained areas of carcinoma with only the wild-type gene, implying that Ki-ras mutation, even when present in a colonic carcinoma, may not have been necessary for establishing the malignant phenotype. Also, 2 of 26 (8 per cent) Dukes' D patients had a mutation in their primary carcinoma but none in liver metastases and 6 of 47 (13 per cent) Dukes' C and D patients had mutations in liver or lymph node metastases but none in the primary carcinoma. Such heterogeneity may modify the effectiveness of novel therapies targeting mutant Ki-ras function, such as farnesyltransferase inhibition. Mutation of codon 12 from GGT (glycine) to GTT (valine) was more prevalent in primary and metastatic deposits of Dukes' C/D carcinomas (P = 0.01) than in primary carcinomas from Dukes' A/B patients. Mutations of codon 12 to GAT, AGT, GCT and codon 13 GGC to GAC were also found, but no correlation with carcinoma aggressiveness was apparent. Follow-up of 71/78 patients (up to 12 years) revealed decreased overall survival (P = 0.001) in patients with the GGT to GTT transversion in codon 12, even when the analysis was restricted to Dukes' D cases, supporting the suggestion that this mutation may confer a more aggressive phenotype in colorectal carcinoma.

Is chromosome 9 loss a marker of disease recurrence in transitional cell carcinoma of the urinary bladder?

Investigation of transitional cell carcinoma of the urinary bladder (TCC) patients classified by recurrence and/or progression has demonstrated that loss of chromosome 9, as detected by FISH analysis of the pericentromeric classical satellite marker at 9q12, occurs early. A total of 105 TCCs from 53 patients were analysed in situ by two independent observers for loss of chromosome 9 using quantitative fluorescence in situ hybridization (FISH). All 53 primary tumours were evaluated for chromosomes 9, 7 and 17. Normal ranges for chromosomal copy number were defined for normal skin epidermis and bladder epithelium. Values for chromosome 9 copy number outwith the range 1.51-2.10 (mean +/- 3 x s.d. of normal values) were significantly abnormal. Twenty-five TCCs were detected with consistent monosomic scores. Of 89 TCCs, in which multiple tumour areas were analysed, 85 tumours (96%) demonstrated the same chromosome 9 copy number in all areas (2-6) analysed; only three tumours demonstrated heterogeneity for this locus. A total of 36% (12 out of 33) of patients with subsequent disease recurrence demonstrated loss of chromosome 9 in their primary and all subsequent TCCs analysed. Only a single patient (n = 20) with non-recurrent TCC showed loss of chromosome 9 (P = 0.0085). Of 53 primary tumours, eight showed significant elevation of chromosome 17. Of these patients, six demonstrated elevation in chromosome 7 copy number. No abnormalities were observed in non-recurrent patients. This study describes rapid quantitation of chromosomal copy number by FISH using a pericentromeric probe for chromosome 9 in TCC of the urinary bladder. Routinely fixed and processed material was evaluated without disaggregation. Strict quality control of FISH demonstrated that this technique was reproducible in a clinical environment and could be used to detect genetic changes relevant to patient outcome. It is proposed that loss of chromosome 9 from primary TCC of the urinary bladder identified patients at high risk of recurrence and possible progression.

Expression of gap junction proteins connexin 26 and connexin 43 in normal human breast and in breast tumours

Gap junctional intercellular communication (GJIC) has been proposed as a cellular mechanism for tumour suppression and there is experimental evidence in support of this. If aberrant GJIC contributes to the formation of human breast tumours, one might expect that the connexins (gap junction proteins) expressed by epithelial cells in normal human breast would be down-regulated in tumour epithelial cells, or that tumour cells might show aberrant expression of other connexin family members. This study examines the immunocytochemical expression of connexins 26 (Cx26) and 43 (Cx43) in normal human breast, 11 benign breast lesions, two special-type carcinomas, and 27 invasive carcinomas of no special histological type (NST). Cx26 generally was not expressed at detectable level in normal human breast, but punctate Cx43 immunostaining of the myoepithelial cells was found. Cx43 staining of the myoepithelium was also a feature of the benign lesions and ductal carcinoma in situ (DCIS). In general, the epithelial cells of benign lesions failed to stain for either connexin. Similarly, a lobular carcinoma did not express Cx26 or Cx43, but there was punctate Cx43 in the epithelial cells of a mucoid carcinoma. Cx26 was up-regulated in the carcinoma cells of 15 of the 27 invasive NST carcinomas, although the staining was usually cytoplasmic and heterogeneous. Cx43 was expressed by stromal cells, possibly myofibroblasts, in all NST carcinomas. Furthermore, there was heterogeneous Cx43 expression in the carcinoma cells of 14 of the 27 NST carcinomas and the staining was often intercellular and punctate, characteristic of functional connexins. Up-regulated of Cx26 and/or Cx43 in the carcinoma cells of over two-thirds of invasive lesions of NST is not necessarily inconsistent with a tumour suppressor role for GJIC. However, the role of gap junctions in the formation and progression of solid human tumours is likely to be more complex than indicated from experimental systems.

Tumour specific regulation of telomerase RNA gene expression visualized by in situ hybridization

Maintenance of telomere structure by the ribonucleoprotein enzyme telomerase is considered central to the development of most human cancers. However, regulatory mechanisms governing telomerase expression during oncogenesis are largely unknown. We address potential tumour-specific regulation of telomerase RNA gene expression by RNA in situ hybridization to over 300 tumour samples of germ cell and epithelial origin. Twenty-six per cent of non-small cell lung cancers (NSCLC), expressed detectable levels of the telomerase RNA gene (hTR), and interestingly expression was almost confined to squamous carcinomas (41%), being rare in pulmonary adenocarcinomas and large-cell anaplastic carcinomas (P=0.006). Low frequency hTR expression was also associated with adenocarcinoma of the breast (13%), and ovary (17%). In comparison, hTR expression was detected in 43% of cervical cancers with no significant differences in frequency between squamous-cell carcinoma and adenocarcinoma or in transitions between intraepithelial neoplasia and invasive carcinoma. In contrast to the common epithelial cancers, the malignant cells in 73% of testicular germ-cell tumours (seminomas and teratomas), expressed hTR consistent with hTR expression in normal testicular germ cells. Differentiated tissues within ovarian germ cell tumours and in testicular teratomas lacked detectable hTR expression. These studies show that different tumour types have distinct patterns of hTR expression, which has implications for our understanding of mechanisms regulating telomerase activity and for targeting the telomerase RNA component as an anti-cancer therapy.

Expression of gap junction proteins connexin 26 and connexin 43 in normal human breast and in breast tumours

Gap junctional intercellular communication (GJIC) has been proposed as a cellular mechanism for tumour suppression and there is experimental evidence in support of this. If aberrant GJIC contributes to the formation of human breast tumours, one might expect that the connexins (gap junction proteins) expressed by epithelial cells in normal human breast would be down-regulated in tumour epithelial cells, or that tumour cells might show aberrant expression of other connexin family members. This study examines the immunocytochemical expression of connexins 26 (Cx26) and 43 (Cx43) in normal human breast, 11 benign breast lesions, two special-type carcinomas, and 27 invasive carcinomas of no special histological type (NST). Cx26 generally was not expressed at detectable level in normal human breast, but punctate Cx43 immunostaining of the myoepithelial cells was found. Cx43 staining of the myoepithelium was also a feature of the benign lesions and ductal carcinoma in situ (DCIS). In general, the epithelial cells of benign lesions failed to stain for either connexin. Similarly, a lobular carcinoma did not express Cx26 or Cx43, but there was punctate Cx43 in the epithelial cells of a mucoid carcinoma. Cx26 was up-regulated in the carcinoma cells of 15 of the 27 invasive NST carcinomas, although the staining was usually cytoplasmic and heterogeneous. Cx43 was expressed by stromal cells, possibly myofibroblasts, in all NST carcinomas. Furthermore, there was heterogeneous Cx43 expression in the carcinoma cells of 14 of the 27 NST carcinomas and the staining was often intercellular and punctate, characteristic of functional connexins. Up-regulated of Cx26 and/or Cx43 in the carcinoma cells of over two-thirds of invasive lesions of NST is not necessarily inconsistent with a tumour suppressor role for GJIC. However, the role of gap junctions in the formation and progression of solid human tumours is likely to be more complex than indicated from experimental systems.

Amplification, increased dosage and in situ expression of the telomerase RNA gene in human cancer

Telomere length is maintained by the enzyme, telomerase, which has been linked to cellular immortality and tumour progression. However, the reasons for the high levels of telomerase found in human tumours are unknown. We have mapped the human telomerase RNA gene, (hTR), to chromosome 3q26.3 and show the hTR gene to be amplified in four carcinomas, (2/33 cervix, 1/31 head and neck, 1/9 lung). In addition, increased copy numbers of the hTR locus was also observed in 97% of tumours. By in situ hybridisation, the histological distribution of high levels of hTR expression could be demonstrated in a lung tumour and its metastasis with hTR amplification. These results are the first report of genetic alterations involving a known component of telomerase in human cancer. Indeed, it is also the first report of the amplification of a specific locus within the chromosome 3q region frequently subject to copy number gains in human tumours. In addition, we also show for the first time the histological distribution of the RNA component of telomerase in human tumours.

Quantitative radioimmunohistochemical measurements of p185(erbB-2) in frozen tissue sections

The relationship between expression of the c-erbB-2 proto-oncogene and the biology of breast cancer has been investigated widely, most studies using immunohistochemistry in formalin-fixed, paraffin-embedded tissues. This technique is at best semiquantitative and there is a high degree of interstudy variability because of its subjective nature and poor methodological standardization. The relationship between the levels of expression and biology can be examined thoroughly only with an accurately quantitative technique. We have developed a radioimmunohistochemical assay to measure p185(erbB-2) in tissue biopsy specimens. The method involves incubating frozen sections with 125I-labeled monoclonal antibody, microautoradiograpy, and grain counting with image analysis. Sections of cell pellets with known c-erbB-2 levels are processed with each batch of samples as internal calibration standards. We have quantified c-erbB-2 expression in 60 breast carcinomas and compared the results with conventional immunohistochemistry. Radioimmunohistochemistry measured receptor levels throughout the range of expression in breast carcinomas, whereas conventional immunohistochemistry detected the protein only in the highest expressing tumors. The quantitative, objective data produced by radioimmunohistochemistry allow a more thorough evaluation of the relationship between c-erbB-2 expression and tumor biology. This technique may have applications in other fields where quantitative data is required and relevant monoclonal antibodies are available.

Practical histological microdissection for PCR analysis

Recovery of cells by histological microdissection is increasingly used for analysis by polymerase chain reaction (PCR) or microchemical techniques. This paper describes techniques of histological microdissection. Sections of archival formalin-fixed, paraffin-embedded tissue up to 15 years old were mounted on plain glass slides. Sections 6-7 microns in thickness stained with toluidine blue were dissected under proteinase K buffer solution, using an electrolytically sharpened tungsten needle in a bacteriological loop-holder and a Leitz mechanical micromanipulator (model M). Detached cell groups were recovered in a silicone-coated pipette tip for PCR analysis after digestion in 25-50 microliters of proteinase K (500/ml) in TRIS-HCl buffer (pH 8.3). Consistent amplification and analysis of microsatellite loci were obtained from 2 microliters of crude lysate using 28-30 cycles of PCR incorporating a 32P 5'-end-labelled primer, electrophoresis under denaturing conditions on 6 per cent polyacrylamide gels, and autoradiographic detection.

Practical histological microdissection for PCR analysis

Recovery of cells by histological microdissection is increasingly used for analysis by polymerase chain reaction (PCR) or microchemical techniques. This paper describes techniques of histological microdissection. Sections of archival formalin-fixed, paraffin-embedded tissue up to 15 years old were mounted on plain glass slides. Sections 6-7 microns in thickness stained with toluidine blue were dissected under proteinase K buffer solution, using an electrolytically sharpened tungsten needle in a bacteriological loop-holder and a Leitz mechanical micromanipulator (model M). Detached cell groups were recovered in a silicone-coated pipette tip for PCR analysis after digestion in 25-50 microliters of proteinase K (500/ml) in TRIS-HCl buffer (pH 8.3). Consistent amplification and analysis of microsatellite loci were obtained from 2 microliters of crude lysate using 28-30 cycles of PCR incorporating a 32P 5'-end-labelled primer, electrophoresis under denaturing conditions on 6 per cent polyacrylamide gels, and autoradiographic detection.

Allelic imbalance at NME1 in microdissected primary and metastatic human colorectal carcinomas is frequent but not associated with metastasis to lymph nodes or liver

Allelic imbalance at the NME locus on chromosome 17q21 was analyzed in colorectal cancer patients using a highly polymorphic microsatellite repeat sequence within NME1 itself. Duplicate samples of carcinoma and adjacent normal tissue was obtained by microdissection from 6 to 7-microns paraffin sections of 94 primary carcinomas (treatment years 1979-1993) and available lymph node and liver secondaries. In 55 patients informative (heterozygous) at this locus, allelic imbalance was examined in primary and secondary carcinomas. Microsatellite instability prevented assessment of allelic balance in two cases, and there was no evidence of homozygous loss at NME1 in any case analyzed. Allelic imbalance at the NME locus in carcinomas was frequent (27/53; 51%), and concordant results were obtained between primary carcinoma and secondary deposits in 30 of 33 (91%) cases. Three discordant cases showed allelic imbalance in secondary deposits but not the primary lesion. Although frequent, allelic imbalance at NME1 had no relationship to Dukes' stage at presentation or with subsequent hepatic metastasis, nor with the primary carcinoma site (proximal versus distal), tumor size, or mitotic or apoptotic index. Moreover, neither disease-free nor overall survival differed between patients with carcinomas showing NME1 allelic imbalance and patients with carcinomas that did not. Our results show that although allelic imbalance is frequent at the NME locus in primary and secondary colorectal carcinomas, there is no evidence to link this with clinical or pathological features or with metastatic potential. Microsatellite PCR and microdissection of enriched populations of carcinoma cells allowed uniformly successful analysis of samples from archival formalin-fixed paraffin-embedded tissue up to 15 years old and clear assessment of allelic imbalance in tumor specimens. Target sequences (e.g., microsatellites and minisatellites) up to approximately 200-250 bp may be reliably analyzed for allelic balance, suggesting that this method is of general utility in the genetic analysis of primary and metastatic neoplasia.