Molecular biology of malignant lymphoma: implications for diagnosis

S 12

Monday

22 October

Symposia

200 1

subunit of the human telomerase complex (hTERT). This enzyme is involved in immortalisation of cancer cells, and is expressed in high levels in 80-90% of all human tumors. Overexpression is partly due to gene amplification in the tumor ceils. The gene is turned off in most normal human tissues, and hTERT is therefore an attractive target for immunotherapy, where hTERT may represent a “universal’ tumor associated antigen. Based on these new, tumor specific and tumor associated antigens, candidate cancer vaccines h&e been tested in clinical trials involving patients with pancreas, colorectal and lung carcinomas and melanomas. The majority of these ttiafs have been small phase l/II protocols primarily focused on safety aspects, protocol optimisation and immunological endpoints. By using and intradermal route of injection and the use of GM-CSF as an adjuvant 50-90% of patients demonstrate immune responses to the vaccines, depending on the antigen and patient group. In some trials a significant association between increased survival and an immune response was observed, indicating that such vaccines may have a clinical potential.

of the same tumor. Of 12 invasive breast carcinomas with a relatively large in situ component, we have microdissected both components, isolated DNA and performed comparative genomic hybridization (CGH). In some tumors, we have found a few distinct tiflerences between otherwise identical genome profiles of both components, suggesting that the number of genetic alterations involved in breast cancer progression is limited. In one tumor, we found high-level amplification of a region on 8q in the invasive component only. This region was further analyzed by fluorescence in situ hybridization (FISH). We believe that amplification of a region on chromosome 8q is involved in the progression of carcinoma in situ to invasive breast carcinoma.

43 Genetic alterations cancer S.R. Lakhani. London,

41

Treatment of melanoma Btiqitte Dreno’ , Francine of Cellular

and Gene

by adoptive

transfer of CTL

Jotereau 2. ’ Department

Therapy;

2 INSERM

of Skin Oncobgy U 463, France

Unit

The two different approaches in the treatment of melanoma by immunotherapy are active immunotherapy which is represented by vaccination and passive immunotherapy with cellular therapy and mainly Tumor Infiltrating Lymphocytes (TIL). These ones used in melanoma stage IV (AJCC) induce a response rate of about 35% with a short duration of clinical response. Two main points are related to the efficiency of TIL: firstly the tumor burden and secondly the injection of specific T cytotoxic lymphocytes. Concerning the first point, it appears that the TIL would be more efficiency with a low tumor burden. Concerning specific cytotoxic T lymphocytes of melanoma antigens, they are considered as crucial in the anti-tumor response. However, until now a direct correlation between specific T cytotoxic activity and clinical response has never been demonstrated. In this context, the aim of our study was to demonstrate the efficiency of TIL used as adjuvant therapy in melanoma stage Ill after lymph node resection and to determine the percentage of specific TIL injected and the conelation with the clinical response. 88 eight patients have been included in an adjuvant randomized open study after positive lymph node resection. One arm (n = 44) received IL2 alone during 2 months, 6 weeks after the surgery and the other arm (n = 44) TIL and IL2 in the same condition. Moreover, among the TIL arm, the percentage of specific cytotoxic T lymphocytes has been determined in 26 patients by studying the intra-cytoplasmic production of gamma interferon by TIL incubated with autologous melanoma cell line. With a median follow up of 3 years, this study shows a significant decrease of relapse (73% without relapse in TIL arm versus 44% in IL2 arm p = 0.019) and an increase of survival in patients treated with TIL with only one invaded lymph node (p = 0.026). The percentage of specific TIL injected varied between 0, 3% and 10%. The injection of specific TIL was significantly associated with a decrease of relapse and an increase of survival (p = 0.02). In condusion, this study demonstrates for the first time, a clinical benefft of TIL used as adjuvant therapy in melanoma stage Ill with a correlation between relapse and overall survival and the injection of specific cytotoxic T cells.

42 Genetic alterations

in intraductal

and invasive breast cancer

M. Van de Vijver. Antoni van Leeuwenhoekhuis,

Amsterdam,

The

Netherlands

Breast cancer develops by transformation of normal epithelial cells into carcinoma in situ, followed by progression to invasive breast cancer, which can than acquire the potential to metastasize. There are several types of breast cancer, and consequently several distinct genetic pathways can be identified. Lobular carcinoma in situ and invasive lobular carcinoma are characterized by inactivation of the E-cadherin gene; well differentiated ductal carcinoma in in situ (DCIS) is characterized by frequent inactivation of an unidentified tumor suppressor gene on chromosome 16q; p53 mutation and HER2 gene amplification are involved in the development of poorly differentiated DCIS and invasive carcinoma. Several studies have revealed that the spectrum of genetic alterations in the in situ tumors is comparable to that of the invasive carcinomas. An important question remains which genetic alterations are involved in the progression from carcinoma in situ to invasive breast cancer. We have compared the genetic alterations in both the in situ and invasive component

in putative precursor

Breakthrough

Breast

Cancer

lesions to breast

Centfe,

Molecular

Pathdogx

UK

The multistep model of carcinogen&s in the breast suggests a transition from normal epithelium to invasive carcinoma via non-atypical and atypical hyperplasia and in situ carcinoma. The introduction of mammographic screening has led to the increased detection of ‘borderline lesions’ and this has highlighted deficiencies in our understanding of these lesions. Atypical Ductal Hyperplasia (ADH): ADH is a controversial lesion, which shares some but not all features of DCIS. Despite clear diagnostic criteria agreement even amongst experienced breast path&gists has been low. Lakhani (1995) demonstrated that LOH identified at loci on 16q and 17p in invasive carcinoma is also present in ADH with a similar frequency. This indicates that ADH is a neoplastic proliferation. There is support forthis view in the literature. The studies demonstrate that within the limits of current investigations, there is no significant difference between ADH and DCIS. ADH as currently defined may represent a small focus of DCIS rather than a separate entity. Hyperplasia of Usual Type (HUT): Retrospective studies indicate that this lesion has a relative risk of 2 for the subsequent development of invasive carcinoma. O’Connell et. al. have demonstrated LOH at many different loci in HUT with frequencies ranging from O-15%. These figures are similar to those of Lakhani et. al. (1996) - range 0% at 13q to 13% at 17q. These frequencies are lower than in DCIS and ADH (range 25-55%). At least a proportion of non-atypical hyperplasias are also donal, neoplastic proliferations. Apocrlne Hyperplasla: Apocrtne papillary hyperplasia is considered to be a benign l-ion desptie a similar architecture to low grade DCIS. Jones et al have investigated genetic alterations in 10 benign apocrine hyperplasia and compared these to apocrine ductal carcinoma in-situ (DCIS) and invasive apocrine carcinomas of the breast using comparative genomic hybridisation (CGH). All lesions exhibited DNA copy number changes. The average number of alterations in apocrine hyperplasia was 4.1 compared to 10.2 in apocrine DCIS and 14.8 in invasive carcinoma. The changes show considerable overlap with those identified in in-situ and invasive apocrine carcinoma. The data suggest that apocrine hvperplasia may be a precursor of apotine carcinoma. Normal tissues: LOH identified in invasive carcinoma is also present in morphological normal lobules. The frequency and significance remains unknown

44 Molecular diagnosls

biology

P.M. Kluin. Academisch Gmningen,

of malignant

Ziekenhuis The Netherlands

lymphoma: Groningen,

implications

for

Dept of Pathology;

Malignant lymphomas consist of Hodgkin lymphomas (HL) and Non-Hodgkin lymphomas (NHL; 90% of B- and 10% of T-cell origin). Normal and neoplastic B- and T-cells express unique B-cell receptori (BCR) and T-cell receptors (TCR), respectively. These receptors are shaped after gene rearrangements in precursor B- and T-cells. Thereafter BCR genes undergo additional modifications, hypermutation and class switching. Recent studies suggested that all gene alterations are mediated by DNA breaks and repair; therefore they also can lead to chromosomal translocation. B-cells undergo these alterations on the functional and non-functional allele. Many investigators focused on the functional BCR allele. Apart from providing useful clonality assays, these studies also led to the concept that B-NHL can be distinguished in pregerminal center, germinal center,and postgerminal center lymphomas. A clinically relevant implication is that chronic lymphocyiic leukemia (CLL), formerly thought to be one disease, consists

Symposia

Monday 22 October 2001

of 2 disorders, pregerminal center cell derived CLL with poor prognosis and postgermlnal center cell CLL with an excellent prognosis. Combined with single cell analysis, these techniques also proved that HL represents a monoclonal disease of germinal center B-cells. Other investigators including our group, showed that many lymphomas contain chromosomal translocations activating oncogenes on the nonfunctional lg allele. These include t(14;18) activating BCL2 in 90% of follicular lymphoma and 1520% of diffuse large B cell lymphoma (DLBCL), t(l1 ;14) activating cyclin Dl in 90% of mantle cell lymphoma and IO-15% of myeloma and t(8;14) or variants activating MYC in 100% of Burkitt lymphomas and 510% of DLBCL. These translocations bear diagnostic significance. At present the best way to detect them is interphase fluorescence in situ hybridization (FISH). Our group designed appropriate probe sets to detect these breakpoints including variant types and adapted the assays to routinely processed paraffin embedded tissue sections. Optimal probes were selected using DNA fiber FISH, a powerful adjunct to map probes and breakooints. Other translocatiins involve BCL6 at 3~27 in 35% of DLBCL. the a&apoptotic API2 on llq21 involved in t(11;18) in 30% of gastric MALT lyrnphoma, and ALK on 2~23 in many CD3Ot T cell lymphomas called ALCL. Detection of t(ll;l8j in gastric limphoma predl&re&ance to eradication of Helicobacter Qlori. Detection of ALK translocations in ALCL predicts a relatively excellent prognosis.

45 Novel molecular

approaches

in various cancers

D. Sidransky. John Hopkins Universit): Department of Otolaryngobgy-Head and Neck Researc, Maryland, USA Cancers progress through a series of genetic and epigenetic changes. These alterations are rapidly being developed into novel molecular assays for cancer detection. Among the most promising, are microsatellite alterations, promoter hypermethylation and somatic mitochondrial mutations. Integrated viral DNA has also been identified as a specific marker for various viral associated malignancies. These alterations can be integrated into various molecular assays and targeted for the detection of a specific cancer type. Tested samples include bodily fluids that drain the organ of interest. Increasingly, plasma or serum DNA has become a fruitful target of new molecular assays to determine the absence or presence of cancer. In addition, the development of novel quantitative assays may allow us to accurately assess the burden of disease in addition to its presence or absence. Final clinical limitation will depend on further technological developments and the availability of large cohorts in which to test the candidate markers.

46 Tbmour-targeted Abstract

As a first approach, we generated a mouse fibroblast cell line~condiionaily immortalized by a terftperature sensitive allele of SV4O Large T antigen. Upon shift to the non-permissive ternpetiture. these c&s un&@o rapid senescence. tntmduction of a retroviral cDNA expressiti library prior to temperature shift led to the idenMcation of several geHes that allows escape from senescence. Interestingly, one these genes (BCLB) isfrequently activated by translocation in human lymphoma. As a second approach we introduced an activated RASvfz oncogene in the conditionally immortalized mouse fibmblasts, together with cDNA expression libraries. Using this approach, we identified hRRfL7 as a gene that allow escape of premature f3AS-induced senescence. Interestingly, myc genes were not found in this screen even though full-length cDNAs of myc genes were abundantly present in these libraries. Functional studies aimed at understanding how these novel genes interact with the p1tiRF-p53 pathway will be presented.

49 Cyclin-dependent kinase inhibitors: New compounds, selectivity, mechanisms of action and anti-proliferattve activities Laurent Meijer, Marie Knockaert. 29682 Roscoff cedex, France

detection

vectors for gene therapy

not received

s13

C.N.R.S.,

Station Biologique,

BP 74,

Cyclindependent kinases (CDKs) directly regulate the cell division cycle phases, transcription, apoptosis and neuronai cells & thymocytes functions. Intensive screening has led in the last few years to the identification of several families of chemical inhibitors of CDKs. Some of these compounds display a high selectivity and efficiency (I& 4 5 nM). Many have been co-crystallised with CDK2 and their atomic interactions with the kinase have been analysed in detail: all are located in the ATP-binding pocket of the enzyme. Some novel structures will be presented. Despite high selectivity, most CDK inhibitors (except purines) are potent inhibitors of glycogen synthase kinase-3 (GSK-3). Whether this GSK-3 inhibiory property is favourabie to the anti-mitotic properties of CDK inhibitors will be discussed. An overall method for the detemGnatlon of selectivity of kinase inhibitors, based on the affinity chromatography purification of targets on immobilised inhibitor, will be presented. This method has led to the identification of unexpected targets. CDK inhibitors display anti-proliferative properties, they arrest cells in Gl and in G2IM. Furthermore they facilitate or even Mgger apoptosis in proliferating ceils. In contrast, they protect neuronal cells and thymooytes from apoptosis. This suggest that CDKs may be involv+ both in triggering and inhibiting apoptosis. The consequences of this dual and conflicting effect need to be evaluated. The combination of CDK inhibitors with other anti-mitotic agents greatly enhances their anti-tumour activity. This will be illustrated by a few examples. The potential of CDK inhibitors is being extensively evaluated for cancer chemotherapy (clinical trials, phase I and II) and will be briefly reviewed. 50

47 Retinoic ackf and arsenic in acute premyelocytic Abstract

Drugs targeting leukemia

not received

48 Genes that allow escape from senescence Daniel Peeper, Avi Shvarts, Thijn Brummelkamp, Benjamin Rowland, Roderick Kortlever, Siriih Douma, Rend Bernards. Division of Molecular Carckwgenesis, The Netherlands Cancer Institute. Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands Senescence limits the proliferative capacity of primary cells in culture and contributes to resistance to transformation. In rodent cells, this process requires induction of the tumor suppressor genes p7p and ~5.3 as mutation of either of these two genes allows escape from replicative senescence and causes immottaH2ation. Expression of an active RAS oncogene in primary rodent cells accelerates this process of senescence and induces a state of premature senescence. Also in the case of F&S-induced premature senescence, the tumor suppressor genes pjpRF and ~5.3 appear to be critical downstream targets in this response. To identify novel regulators of the pltiRf-p53 pathway we petformed functional genetic screens with retroviral cDNA expression libraries.

p53 regulatory

mechanisms

S. Lain, D. Lane. Ninewells Hospifal & Medical School, Depatimenf Surgery and Molecular Biology, Dundee, United Kingdom

of

The p53 turnout suppressor is a short-lived protein that is maintained at low levels in nonal cells. In response to stresses, p53 is transiently stabilised and its function as a transcriptional activator is increased, Activation of the ~53 function causes cell cycle arrest and/or apoptosis, thus preventing the fixation of mutations in the following generations. The p53-mediated stress response is one of the most common mechanisms inactivated in tumour cells. More than 50% of all human cancers have mutations in the ~53 gene. In many other cases, there are lesions that affect proteins that modulate the levels and activity of p53. This is the case of tumour cells In whii the product of the MdmP gene is overexpressed, of turnour.~ where there is a deficiency in the expression pl4ARF tumour suppressor (the natural inhibitor of Mdm2) or of tumours associated with the expression of viral oncoproteins such as the HPV E6 protein. The search for non-genotoxic ways to activate the p53 response is widely thought to be a promising approach in the development of new low risk strategies to treat cancers. There are many reports showing that Mdm2 is the major regulator of the ~53 levels in the ceil. This is due to its abilitv to inhibit the interaction of ~53 with the transcriptii machinery and io its ubiquitin E3 ligase a&& that promotes the~ubiquitination and pmteasome degradation of ~53. Therefore, a direct and rational way to enhance the p53 response in a non-genotoxic way in turnouts where the