Digital image analysis agrees with visual estimates of adult bone marrow trephine biopsy cellularity.

INTRODUCTION: Evaluation of cellularity is an essential component of bone marrow trephine biopsy examination. The standard practice is to report the results as visual estimates (VE). Digital image analysis (DIA) offers the promise of more objective measurements of cellularity. METHODS: Adult bone marrow trephine biopsy sections were assessed for cellularity by VE. Sections were scanned using an Aperio AT2 Scanscope and analyzed using a Cytonuclear (version 1.4) algorithm on halo software. Intraclass correlation (ICC) was used to assess relatedness between VE and DIA, and between MRI and DIA for a separate subset of patients. Trephine biopsy sections from a subset of patients with bone marrow biopsies uninvolved by malignancy were assessed for age-related changes. RESULTS: Interobserver VE agreement was good to excellent. The ICC value was 0.81 for VE and DIA, and 0.50 for MRI and DIA. Linearity studies showed no statistically significant trend for age-related changes in cellularity in our cohort (r = -.29, P = .06). CONCLUSIONS: Agreement was good between VE and DIA. It may be possible to use DIA or VE to measure cellularity in the appropriate clinical scenario. The limited sample size precludes similar determinations for MRI calculations. Further studies examining healthy donors are necessary before making definitive conclusions regarding age and cellularity.

Detection of chromosomal translocations in formalin-fixed paraffin-embedded (FFPE) leukemic specimens by digital expression profiling.

INTRODUCTION: Detection of chromosomal translocations in formalin-fixed paraffin-embedded (FFPE) leukemic samples is important for confirmation of histopathological findings, classification, prognostication, and therapeutic decisions. Herein, we aim to determine whether digital expression profiling could detect chromosomal translocations in FFPE leukemic samples identified by RT-PCR, FISH, and/or karyotyping. METHODS: RNA was extracted from 28 FFPE bone marrow specimens from 19 patients diagnosed with leukemia. Eight patients were translocation t(9;22) positive, three inv(16)/t(16/16) positive, five translocation t(15;17) positive, and one translocation t(8;21) positive. Two patients (four specimens) were normal. The extracted RNA was hybridized to DNA reporter probes overnight at 65 °C, followed by purification of the labeled translocations. Six hundred fields of view were counted to enumerate the number of translocations. RESULTS: Digital expression profiling had 100% concordance with RT-PCR, FISH, or karyotyping analysis in the two normal individuals, eight translocation t(9;22) samples, five translocation t(15;17) samples, and one translocation t(8;21) sample. None of the inv(16) positive samples were detected. Digital expression profiling detected translocations with 0.014 p190 allelic burden. CONCLUSION: Digital expression profiling can be used to measure translocation t(9;22), t(15;17), and t(8;21) in FFPE samples and is useful when a confirmatory test from a FFPE sample is necessary.

Caspase-2 impacts lung tumorigenesis and chemotherapy response in vivo.

Caspase-2 is an atypical caspase that regulates apoptosis, cell cycle arrest and genome maintenance, although the mechanisms are not well understood. Caspase-2 has also been implicated in chemotherapy response in lung cancer, but this function has not been addressed in vivo. Here we show that Caspase-2 functions as a tumor suppressor in Kras-driven lung cancer in vivo. Loss of Caspase-2 leads to enhanced tumor proliferation and progression. Despite being more histologically advanced, Caspase-2-deficient tumors are sensitive to chemotherapy and exhibit a significant reduction in tumor volume following repeated treatment. However, Caspase-2-deficient tumors rapidly rebound from chemotherapy with enhanced proliferation, ultimately hindering long-term therapeutic benefit. In response to DNA damage, Caspase-2 cleaves and inhibits Mdm2 and thereby promotes the stability of the tumor-suppressor p53. Caspase-2 expression levels are significantly reduced in human lung tumors with wild-type p53, in agreement with the model whereby Caspase-2 functions through Mdm2/p53 regulation. Consistently, p53 target genes including p21, cyclin G1 and Msh2 are reduced in Caspase-2-deficient tumors. Finally, we show that phosphorylation of p53-induced protein with a death domain 1 leads to Caspase-2-mediated cleavage of Mdm2, directly impacting p53 levels, activity and chemotherapy response. Together, these studies elucidate a Caspase-2-p53 signaling network that impacts lung tumorigenesis and chemotherapy response in vivo.

Re-engineered p53 activates apoptosis in vivo and causes primary tumor regression in a dominant negative breast cancer xenograft model.

Inactivation of p53 pathway is reported in more than half of all human tumors and can be correlated to malignant development. Missense mutation in the DNA binding region of p53 is the most common mechanism of p53 inactivation in cancer cells. The resulting tumor-derived p53 variants, similar to wild-type (wt) p53, retain their ability to oligomerize via the tetramerization domain. Upon hetero-oligomerization, mutant p53 enforces a dominant negative effect over active wt-p53 in cancer cells. To overcome this barrier, we have previously designed a chimeric superactive p53 (p53-CC) with an alternative oligomerization domain capable of escaping transdominant inhibition by mutant p53 in vitro. In this report, we demonstrate the superior tumor suppressor activity of p53-CC and its ability to cause tumor regression of the MDA-MB-468 aggressive p53-dominant negative breast cancer tumor model in vivo. In addition, we illustrate the profound effects of the dominant negative effect of endogenous mutant p53 over wt-p53 in cancer cells. Finally, we investigate the underlying differential mechanisms of activity for p53-CC and wt-p53 delivered using viral-mediated gene therapy approach in the MDA-MB-468 tumor model.

pT1 substaging in renal cell carcinoma: validation of the 2002 TNM staging modification of malignant renal epithelial tumors.

PURPOSE: Tumor size has been used as one of the criteria to stratify renal cell carcinoma (RCC) into different pathological stages (pT). The recent 2002 UICC/TNM classification of malignant epithelial renal tumors is modified to substratify pT1 RCC into pT1a (less than 4.0 cm) and pT1b (greater than 4.0 but less than 7.0 cm). In this study we ascertained if this stage modification has prognostic relevance. MATERIALS AND METHODS: A total of 259 consecutive radical nephrectomy specimens of organ confined RCC from 1970 to 1997 at 1 institution, including 153 of conventional RCC (CRCC), 71 of papillary RCC, 28 of chromophobe RCC, 1 of collecting duct carcinoma and 6 of RCC not otherwise specified, with a mean clinical followup of 7.5 years (median 6.4) were included in the study. RESULTS: There were 115 pT1a (44.4%), 95 pT1b (36.7%) and 49 pT2 tumors (18.9%). Disease recurrences (DR) and disease specific death occurred in 2 (1.7%) and 0 cases (0%) of pT1a, 7 (7.3%) and 5 (5.3%) of pT1b, and 16 (32.6%) and 12 (24.5%) of pT2. DR for pT1b was higher compared with pT1a (all histological subtypes RR 3.68), although this difference was not statistically significant (p = 0.106). If only CRCCs were analyzed, DR in the pT1b group was statistically higher compared with pT1a (RR 8.54, p = 0.047). Disease specific survival in pT1a could not be evaluated because no deaths occurred in this subgroup. DR and disease specific survival were significantly different between pT1b and pT2 tumors for all histological subtypes (RR 5.51, p = 0.001 and 5.49, p = 0.001) and for the CRCC subtype (RR 5.50, p = 0.001 and 5.18, p = 0.005, respectively). Using size as a continuous variable the logarithmic change in tumor size was a significant predictor of DR (RR 8.82, p = 0.001). All statistical analyses were adjusted for age and sex. CONCLUSIONS: Substaging RCC into pT1a and pT1b yields prognostically important information, validating the 2002 TNM modification for malignant renal epithelial malignancies. The substratification of pT1 is particularly useful in tumors with CRCC histology.

p16INK4a expression in actinic keratosis and Bowen's disease.

BACKGROUND: Progression of cutaneous squamous neoplasms from actinic keratosis (AK) to Bowen's disease (BD; squamous cell carcinoma in situ) has important implications for clinical management and treatment, thus requiring accurate diagnosis. p16INK4a is a cell cycle regulatory tumour suppressor protein that negatively regulates D-type cyclins in the G1 cell cycle phase via intimate interplay with the retinoblastoma gene. Expression of a paraffin-reactive p16INK4a marker has recently been shown to increase in cervical squamous neoplasms as lesions progress from low-grade dysplasia to squamous cell carcinoma in situ. p16INK4a expression in the progression of squamous cutaneous neoplasia, however, has not been evaluated. OBJECTIVES: To evaluate p16INK4a expression in the progression of squamous cutaneous neoplasia. METHODS: Biopsies of 203 squamous cutaneous neoplasms with unequivocal features of AK (n = 87) and BD (n = 116) as well as a benign squamous control group (verruca vulgaris: n = 10; seborrhoeic keratosis: n = 11; scar tissue: n = 8) obtained between January and December 2001 at Henry Ford Hospital (Detroit, MI, U.S.A.) were immunostained for p16INK4a (Dako; clone E6H4; dilution 1 : 50) using large core (1.5 mm) tissue microarray analysis. Nuclear/cytoplasmic immunoreactivity in > 10% of neoplastic cells was considered positive. RESULTS: Of 203 cases, 166 (81.8%) were interpretable (AK 59; BD 107). Mean patient age was 71.0 years (range 33-93); 57% were male. Sites of involvement were: head and extremities 75.9%, trunk/buttocks 21.7%, genital region 2.4%. p16INK4a immunostaining was positive in 90 of 107 (84.1%) BD cases, four of 59 (6.8%) AK cases and none of 29 benign squamous controls. The sensitivity and specificity of p16INK4a for a diagnosis of BD (vs. benign squamous controls/AK) was 84.1% and 95.5%, respectively (P < 0.0001, Fisher's exact test, two-sided). CONCLUSIONS: p16INK4a is a sensitive and specific marker for distinguishing BD from AK/benign squamous cutaneous lesions and may be helpful as an adjunct to histomorphology in the diagnosis and appropriate clinical management of these lesions.

Malignant papillary renal tumors with extensive clear cell change: a molecular analysis by microsatellite analysis and fluorescence in situ hybridization.

CONTEXT: Histologic subtyping of renal cell carcinomas (RCCs) is based not only on cytoarchitectural pattern but also on distinct cytogenetic abnormalities. Some renal tumors demonstrate overlapping morphologic features, rendering histologic subtyping difficult. One such group of tumors is papillary renal neoplasms with extensive clear cell change. Because histologic subtyping has been shown to be of prognostic value, it is important that malignant epithelial renal tumors be accurately subtyped. It is not known if these tumors should be classified as papillary RCC (PRCC) or as conventional/(clear cell) RCC (CRCC). OBJECTIVE: To ascertain if this subgroup of renal neoplasms demonstrates the cytogenetic abnormalities seen typically in PRCC, that is, trisomy 7 and 17 or CRCC, that is, loss of 3p, using microsatellite analysis for loss of heterozygosity (LOH), and fluorescence in situ hybridization (FISH) for trisomies. DESIGN: Seven RCCs from 6 patients that showed more than 75% papillary architecture and more than 75% clear cell change were included in the study. Tumor size ranged from 2.5 to 7.0 cm (mean 4.7 cm) and all were confined to the kidney (stage I). DNA was extracted from formalin-fixed paraffin-embedded tissue. FISH was done using In Situ Kits for centromere probes for chromosomes 7 and 17. For LOH, microsatellite analysis using labeled primers for 4 markers in the 3p13 through 3p24.2 region were used. The amplified polymerase chain reaction products were analyzed using an automated DNA sequencer. As compared with normal DNA, LOH in tumor was recognized as a loss of 1 allele, and microsatellite instability as the addition of an extra allele. RESULTS: LOH in at least 1 of the markers spanning for 3pl3 through 3p24.2 was detected in 6 of 7 specimens (86%), of which 1 also showed concomitant microsatellite instability. FISH did not demonstrate trisomy for either chromosome 7 or 17. Instead, monosomy 7 was observed in 4 of 6 tumors (67%) and monosomy 17 in all tumors (100%). CONCLUSION: Because malignant papillary renal tumors with extensive clear cell change show molecular changes identical to CRCC, this subgroup of tumors may have to be classified as CRCC. This study underscores the utility of molecular studies in refining light-microscopic criteria in accurate histologic subtyping of RCCs.