To accurately estimate the incidence of HOX11L2 expression, and determine the associated cytogenetic features, in T-cell acute lymphoblastic leukemia (T-ALL), the Groupe Français de Cytogénétique Hématologique (GFCH) carried out a retrospective study of both childhood and adult patients. In total, 364 patients were included (211 children ⩽15 years and 153 adults), and 67 (18.5%) [47 children (22.4%) and 20 adults (13.1%)] were shown to either harbor the t(5;14)q35;q32) translocation or express the HOX11L2 gene or both. Most of the common hematological parameters did not show significant differences within positive and negative populations, whereas the incidence of CD1a+/CD10+ and cytoplasmic CD3+ patients was significantly higher in positive than in negative children. Out of the 63 positive patients investigated by conventional cytogenetics, 32 exhibited normal karyotype, whereas the others 31 showed clonal chromosome abnormalities, which did not include classical T-ALL specific translocations. Involvement of the RANBP17/HOX11L2 locus was ascertained by fluorescence in situ hybridization in six variant or alternative (three-way translocation or cytogenetic partner other than 14q32) translocations out of the 223 patients. Our results also show that HOX11L2 expression essentially occurs as a result of a 5q35 rearrangement, but is not associated with another identified T-ALL specific recurrent genetic abnormality, such as SIL-TAL fusion or HOX11 expression.
The extensive use of fluorescence in situ hybridization (FISH) techniques in cytogenetic studies of hematopoietic malignancies has uncovered several examples of cryptic chromosome rearrangements. For instance, translocation t(12;21)(p13;q22)1 was unexpectedly found to be the most frequent translocation of childhood B-cell acute lymphoblastic leukemia (ALL) and t(5;11)(q35;p15.5) was recently identified in childhood acute myeloblastic leukemia.2 More recently, the translocation t(5;14)(q35;q32), only observed upon FISH analyses, was reported in approximately 20% of childhood T-cell ALLs.3,4 On chromosome 14, the breakpoints are scattered in the vicinity of the CTIP2/BCL11B gene, which is highly expressed during T-cell lymphoid differentiation. On chromosome 5, the t(5;14)(q35;q14) usually disrupts the RANBP17 gene, but is described to result in the ectopic transcriptional activation of HOX11L2, a gene encoding a homeobox transcription factor of the HOX11 family. The HOX11 gene, the founding member of the family, is located on human chromosome band 10q24 and was identified because of its involvement in T-ALL specific chromosomal translocations.5,6 The two translocations, t(10;14) and t(7;10), rearrange the TCR α/δ (on chromosome 14) or the TCR beta (on chromosome 7) with the HOX11 locus also resulting in its transcriptional activation. However, in a few percent of T ALL cases, HOX11 expression appears to be expressed in the absence of obvious chromosomal abnormality of 10q24 on banded karyotypes.
To further establish the association between 5q35 abnormality and HOX11L2 expression within T-ALL, the Groupe Français de Cytogénétique Hématologique (GFCH) initiated a collaborative effort to analyze a large number of adult and childhood T-ALL patients by cytogenetics, FISH, and RT-PCR.
Materials and methods
A retrospective study of childhood and adult T-ALL patients was carried out by the GFCH members, with the participation of specialists in morphology, immunology, and molecular biology (List in Appendix). All patients included in this study had T-ALL or T-cell lymphoblastic lymphoma, and were tested for t(5;14)/HOX11L2 expression either by FISH, molecular analysis (PCR), or both techniques. Out of 211 children, 97 included in the present series were also included in the EORTC-CLG series (H Cavé et al, submitted). On account of the heterogeneity of patient population, it was elected not to analyze outcome data such as percentage of complete remission (CR), duration of CR, or survival.
Morphologic and immunologic validation
Morphologic and immunophenotypic studies of the ALL patients were carried out in each center and validated by groups of morphologists and immunologists responsible for the collection of data in various Working Groups, particularly involved in therapeutic protocols for children (EORTC-CLG,7 FRALLE8) and for adults (LALA87 and LALA949). Immunophenotyping was performed by flow cytometry in most of the cases. Diagnosis was made on bone marrow and/or blood, lymph node, or effusion cells. Membrane expression of CD1a, CD2, CD3, CD4, CD5, CD7, CD8, CD10, HLA-DR, CD19, CD20, CD13, CD14, CD33, CD34, CD117, and CD65 as well as TCRα/β and TCR γ/δ and intracytoplasmic CD3 (cCD3) was tested.10 Since the study was retrospective, all antigens could not be tested in each case, but the T-cell nature of ALLs could be ascertained in 364 patients who form the basis of this report, including 25 with lymphoblastic lymphoma.
Cytogenetic and FISH studies
Cytogenetic analysis was performed on bone marrow (274 patients), and/or blood (30/27 patients), or lymph node (four patients), or effusion cells (six patients), depending on the invasion by blast cells, after 24, 48, or even 72 h in vitro culture. R-bands (by heating or fluorescence) techniques were applied in the majority of the cases, but GTG and QFQ techniques were also occasionally used. Chromosomes were classified according to International Nomenclature.11 All chromosomal data were reviewed by the members of GFCH during successive workshops (subgroups and entire group) in order to assess the quality control and validate the karyotype interpretation. To be classified as normal, at least 20 metaphases were analyzed. The definition of clonal abnormality followed the recommendations of ISCN 1995.11
FISH analysis was performed either in the Hematology Center where the patient was initially examined or in one of the other Centers involved in the study.
Probes used to detect t(5;14) are as follows (Figure 1): YAC 885A6 covering all the known chromosomal breakpoints within the RanBP17/HOX11L2 region; BACs 45L16 and 546B8 covering most of these breakpoints; whole chromosome 14 specific painting probe (Q-Biogene, Illkirch), and BAC clones covering subregions of 14q32 (R Heilig, Genoscope, Evry) (Figure 2).
Dual-color FISH to metaphase chromosomes was the most frequently used technique, and FISH analysis on interphase nuclei was often carried out to increase the number of cells examined. Different approaches were used to evidence t(5;14) by FISH. As the chromosomal breakpoints are dispersed on 5q35 and even more dispersed on 14q32, dual-color FISH with YAC 885A6 covering the breakpoint region on 5q35 and whole chromosome painting for chromosome 14 were thought to be effective in detecting t(5;14)(q35;q32). BAC clones were used in a second step to localize the chromosome breakpoints more precisely. However, direct use of BACs 45L16 and 546B8, covering many breakpoints on 5q35 was preferred in some laboratories. IPM-FISH (IRS-PCR Multiplex FISH) was used in one laboratory.4 Occasionally, IGH BAC probe 158A2 (H. Avet-Loiseau, Nantes) was also used.
The χ2 and Student t-tests were used to compare data from t(5;14)/Hox11L2-positive and -negative patients.
The present retrospective study includes 364 patients with T-ALL (including 25 with lymphoblastic lymphoma) recruited between 1986 and 2002. They were 211 children (⩽15 years, 156 males, and 55 females), and 153 adults (117 males and 36 females). Eight patients were examined in relapse and the others at diagnosis prior to treatment. Both FISH and PCR analyses were carried out in 223 patients, FISH only in 73, and PCR only in 68 patients.
Frequency of t(5;14) or HOX11L2-positive patients
Patients demonstrating either a 5q35 breakpoint involving the RanBP17/HOX11L2 region by FISH or HOX11L2 expression by RT-PCR were considered as positive. The number of positive cases varied with the age of the patients: 47 in children (36 M and 11 F) that is, 22.3%, 20 in adults (16 M and four F), that is, 13.1%. The overall frequency was thus 18.4%. The distribution of ages did not differ significantly between positive and negative samples within a population: 7.4±0.5 (positive cases) vs 8.2±3.3 (negative) in children, and 29±3.26 vs 30.9±1.2 in adults. The prevailing proportion of male vs female patients was comparable in positive and negative patients.
Some hematological parameters were not statistically different in positive and negative patients. The mean percentage of blast cells in bone marrow was not different in positive and negative patients (83.2±2.2 vs 84.7±1.0), neither peripheral blood leukocyte counts (93.6±15.8 G/L vs 115.5±9.7), nor platelet counts (126.9±13.9 G/L vs 113.7± 8.1). The percentage of blast cells in peripheral blood was slightly lower in positive than in negative patients (55.4±4.3 vs 65.7±2.0, P=0.025), and the hemoglobin level was slightly higher in positive than in negative patients (13.7±2.5 g/dl vs 10.9±0.19, P=0.03).
Immunophenotypic data were heterogeneous due to both the large time period covered and the number of centers involved in the study. No significant differences were observed between positive and negative cases for most of the markers studied (Table 1). However, the number of CD1a+, and double-positive CD1a+/CD10+ was significantly higher in children with t(5;14)/HOX11L2 ALL compared with the others (P<0.001), whereas the differences were not significant in adults. However, when both children and adult cases were pooled, the difference remained significant (P<0.001). The number of HOX11L2-positive children with intracytoplasmic CD3 (cCD3) expression was significantly higher than the number of cCD3+ children without HOX11L2 expression (P = 0.006), whereas no such difference was noted in adult patients (P=0.2). It must be noted, however, that the number of adult patients is low compared with the children (Table 1).
From these studies, it can be concluded that the representation of cortical type of T-ALL (T-III in the EGIL nomenclature) with CD10 and cCD3 expression is higher within the t(5;14)/HOX11L2-positive population than within the negative one.
The common t(5;14)(q35;q32) is not detectable with banding techniques and FISH must be used to evidence its presence. As T-cell ALL is classically associated with a high proportion of karyotypes devoid of visible abnormalities, t(5;14)/HOX11L2-positive and -negative patients were compared for the number of NN (normal metaphases only), AA (only abnormal metaphases), and AN (mixture of normal and abnormal metaphases). No difference was observed between the two categories of patients (Table 2). Karyotype was successful in 63 out of 67 t(5;14)/HOX11L2-positive patients. Normal banded karyotype was found in 54 and 42% of successfully karyotyped children and adults, respectively. In the 31 t(5;14)/HOX11L2-positive patients with abnormal karyotype, no classical recurrent translocation was observed. Furthermore, no abnormality common to the patients could be observed, although trisomy 8 was present in six patients (only in one as the sole abnormality), del(9)(p21-22) in four, del(7)(p) in four, del(6)(q21) in three, del(1)(p21p23) in three, del(11) (q14q24) in two, and loss of Y in two children (Table 3).
Abnormalities of the long arm of chromosome 5 with different breakpoints were observed in five positive patients (Table 4). They were classified by banding techniques as add(5)(q34), del(5)(q35), t(2;5)(p21;q35), t(5;14)(q31;q32), and add(5)(q31), respectively (Tables 3 and 4). Of note, one also had visible rearrangement of 14q32, and another patient with common t(5;14) had add(14)(q32) chromosome without visible 5q rearrangement on banded karyotypes. All these patients were shown to harbor alternative or variant t(5;14) ascertained by FISH or HOX11L2 expression analyses.
FISH was performed in 296 patients, and 47 of them revealed a common t(5;14). Using the detection with YAC 885A6 and painting probes, a YAC hybridization extrasignal was visible within the painted part of chromosome 14, demonstrating the splitting of YAC signal and ascertaining the presence of translocation. Splitting signals (one on rearranged 5 and one on rearranged 14) were visible using BAC probes covering the chromosome 5 breakpoint. In one patient, FISH analyses demonstrated loss of the normal chromosome 5 and presence of two copies of the der(5) chromosome. FISH ascertained the involvement of the RANBP17/HOX11L2 region in the variant translocations. Two alternative translocations t(5;7)(35;q21) without involvement of chromosome 14 were found, one of which had been missed on banded karyotype. The alternative t(2;5)′ (p21;q35) translocation, also without involvement of chromosome 14, observed by classical cytogenetic analyses was also shown to involve RANBP17/HOX11L2 and an uncharacterized locus on chromosome 2. In another patient, t(5;22) translocation was detected, but the precise breakpoint could not be localized because of the shortage of material. A patient presenting with 5q31 breakpoint was found by FISH to have a complex rearrangement of 5q ultimately resembling in common t(5;14). One example of three-way translocation, involving chromosome 12p13, was also characterized (Table 4).
RQ-PCR (real-time RT PCR) analysis was performed in the majority of the 291 patients studied, whereas RT-PCR was used in some instances. HOX11L2 was found to be expressed in 55 patients, 42 children, and 13 adults. HOX11L2 expression was detected on each sample with 5q35 abnormality, including those with variant or alternative translocations. Owing to the prevalence of t(5;14), HOX11L2 expression is very likely associated with t(5;14) in the 68 samples that were not analyzed by FISH.
BACs and breakpoint localization
The localization of chromosomal breakpoints by FISH analysis was performed with BAC probes selected to encompass the 5q35 and 14q32 breakpoints. The results obtained in the same laboratory on 25 patients showed that most of the breakpoints on 5q35 were located within sequences covered by BACs 45L16 (eight patients) and 546B8 (14 patients), and between the two BACs in one patient (Figure 1). The translocation of BAC 546B8 onto 14q32 was accompanied by the loss of signal of 45L16 (deletion) in one patient, and the breakpoint localization could not be ascertained in two other patients in whom the translocation was only shown by splitting of YAC 885A6. The main conclusion of these data is that translocations may escape detection if only two of the BACs 45L16 and 546B8 are used.
The breakpoints were more dispersed on 14q32, spanning a minimal interval of 1700 kb (Figure 2). However, seven out of 22 breakpoints were covered by BAC 2576L4, centromeric and distant (more than 1200 kb) from the BCL11B gene locus.
The dispersion of breakpoints justifies the use of whole chromosome 14 specific painting probe in the first step of a search for t(5;14).
Discrepancies between FISH and PCR analysis
Six patients out of 223 (2.7%) showed HOX11L2 expression in the absence of rearrangement detected by FISH (Table 5). The analysis of individual situations shows that karyotype or FISH analysis (four metaphases only) failed in two patients and that the low number of blast cells in one sample studied precluded any conclusion. It is worth noting that the malignant clone can be present as a minor population, as examplified by a positive case with only one t(5;14)-positive metaphase out of 19 analyzed. In two patients, the two BAC clones used as chromosome 5 FISH probes cover only a limited area of 5q35 and, conceivably the breakpoints could have been missed. In another patient, the very careful analysis of chromosome painting (wcp14) showed a very thin signal on chromosome 5 with chromosome 14 probe, suggesting a hidden and complex rearrangement. Taken together, the discrepancy could not be explained by practical reasons in only one patient.
Recurrent chromosome abnormalities in t(5;14)/Hox11L2-negative ALL patients
Karyotype analysis of the 364 ALL cases confirms the low incidence of other known translocations of T-ALLs (Table 6). The t(10;11)(p12;q14) that results in CALM-AF10 gene fusion was observed in five children and three adults, that is, 2.3% of successfully karyotyped patients. Translocations affecting the 10q24 band [t(10;14) and t(7;10)] were more frequent in adult patients (15 cases) than in children (five cases). As a whole, these abnormalities were present in 5.8% of the estimable T-ALL samples.
In our series of negative patients, the frequency of partial deletions of the long arm of chromosome 6 (6q-) with variable breakpoint localization is also high: 16 in children and 17 in adults (11.9%) as well as del(9)(p12;p24) (eight in children and nine in adults, ie 6.1%).
Rearrangements of 5q with variable breakpoint localization were present in 16 negative patients, that is, 5.8% of the cases without expression of HOX11L2. As expected, the presence of 5q rearrangement is therefore not always associated with t(5;14)/HOX11L2-positive subtype of T-ALL.
Taken together, the chromosomal features of this series are very close to what one can expect from the literature.
The introduction of FISH techniques allowed the description of cryptic acquired chromosomal abnormalities previously overlooked in hematopoietic malignancies. The recently identified recurrent translocation t(5;14)(q35;q32) has been recognized as the most frequent translocation in T-cell ALL.3
To obtain a better estimate of the frequency of t(5;14)(q35;q32) and of HOX11L2 deregulation in T-ALL, a series of 364 T-ALLs (211 children from 0 to 15 years, and 153 adults more than 15-year- old) were collected in a retrospective collaborative study of the GFCH. Both FISH and RT-PCR techniques were systematically used in order to pick up all the potential deregulation examples of HOX11L2.
Taken together, the frequency of t(5;14)(q35;q32)/HOX11L2-positive ALL was found to be higher in children (22.3%) than in adults (13.1%) (P<0.02). Hematological data (percentage of blast cells in bone marrow and blood, number of leukocytes and platelets in peripheral blood, hemoglobin level) did not differ between t(5;14)/HOX11L2-positive and -negative patients. A high number of positive-ALLs was of the cortical type (TIII in EGIL classification10) with frequent coexpression of CD1a and CD10. Another feature of the t(5;14)/HOX11L2-positive ALL is the frequent expression of cCD3.
Conventional cytogenetics detected clonal chromosomal abnormality in 31 out of 63 (49%) estimable patients with t(5;14)/HOX11L2-positive ALL. Therefore, the percentage of ‘normal’ banded karyotypes (51%) may account, in part, for the classical high proportion of T-ALL without chromosome abnormality often underscored in the past. Interestingly, no recurrent classical translocation of T-ALL was observed in positive patients ruling out oncogenic cooperation between those events.
Additional molecular data were available for some of the samples. In the patients tested, SIL-TAL rearrangements were present in 15 out of 96 children (15.6%) and in three of 35 adults (8.6%) (Table 7). Comparison with HOX11L2 and HOX11 data (see below) showed that they were mutually exclusive, except in one sample, which demonstrated both SIL-TAL1 fusion and HOX11 expression. HOX 11 expression was detected in 23 out of the 135 patients tested (Table 7). The frequency was markedly higher in adults (37.5%) than in children (12%). HOX11 expression was associated with a chromosomal translocation involving 10q24 in 20 patients. In four children and three adults expressing HOX11, no abnormality of 10q24 had been observed. The HOX11 gene is not expressed during normal hematopoietic differentiation and the reasons underlying HOX11 expression in these instances remain elusive. This observation has been reported several times and deserves further investigations.
The example of HOX11L2 appears more trivial. As HOX11, HOX11L2 is not expressed during normal hematopoietic differentiation. Although not yet formally proven, it is accepted that HOX11L2 is transcriptionally activated by the transcription regulatory elements of CTIP2, which is highly expressed during T-cell differentiation, and is transposed to this region as a result of t(5;14). The t(5;14)(q35;q11), which affects the RanBP17/HOX11L2 region and the TCR α/δ gene, is also expected to activate HOX11L2 expression through the TCR gene regulatory elements.14,15 HOX11L2 expression was detected only in samples harboring a rearranged 5q35 region. The existence of alternative translocations such as t(5;7) and t(2;5) shows that other genes can donate regulatory elements. Similar examples have been described, in which deregulation of the TAL1/SCL gene is achieved by juxtaposition to SIL, TCR, and TCTA genes.16,17
The study of HOX11L2 expression is not always sufficient to characterize the genetic abnormality of the subtype of ALL because of the existence of alternative translocations. Similarly, the use of molecular probes covering only the most common region of breakpoints on chromosome 5, such as BACs 45L16 and 546B8, may be insufficient to detect the 5q rearrangement, because there are some ALLs in which the breakpoint is located outside the region covered by these BACs. Some discrepancies between FISH and RT-PCR results would result from this limited approach. The situation is worse for the determination of chromosomal breakpoints on 14q32 because they are scattered over more than 1700 kb. Finally, HOX11L2 expression studied by RT-PCR remains a fast means to determine the presence of the rearrangement, since it seems specific to the ALL subtype and absent in normal lymphoid cells. Within our samples, only six cases demonstrated HOX11L2 expression in the absence of a detectable 5q35 abnormality. In most of them, the abnormality could have escaped detection for practical reasons, such as number of blast cell in the sample or limited FISH analyses. As a whole, although one example remains truly puzzling, our data indicate that HOX11L2 expression in T-ALLs occurs only because of 5q35 rearrangement. In T-ALL samples, expression of TAL1/SCL and HOX11 has been claimed to occur in the absence of abnormality of their respective locus.18 In our series, HOX11 is expressed in seven samples in the absence of 10q24 rearrangement, as judged from conventional cytogenetic data. Further investigation is needed to understand the molecular reasons for these abnormal expressions.
Romana SP, Mauchauffé M, Le Coniat M, Chumakov I, Le Paslier D, Berger R et al. The t(12;21) of acute lymphoblastic leukemia results in a tel-AML1 gene fusion. Blood 1995; 85: 3662–3670.
Jaju RJ, Haas OA, Neat M, Harbott J, Saha V, Boultwood J et al. A new recurrent translocation, t(5;11)(q35;p15.5), associated with del(5q) in childhood acute myeloid leukemia. Blood 1999; 94: 773–780.
Bernard OA, Busson-Le Coniat M, Ballerini P, Mauchauffé M, Della Valle V, Monni R et al. A new recurrent and specific cryptic translocation, t(5;14)(q35;q32), is associated with expression of the Hox11L2 gene in T acute lymphoblastic leukemia. Leukemia 2001; 15: 1495–1504.
Hélias C, Leymarie V, Entz-Werle N, Falkenrodt A, Eyer D, Aurich-Costa J et al. Translocation t(5;14)(q35;q32) in 3 cases of childhood T-cell acute lymphoblastic leukemia: a new recurring and cryptic abnormality. Leukemia 2002; 16: 7–12.
Dubé ID, Kamel-Reid S, Yuan CC, Lu M, Wu X, Corpus G et al. A novel homeobox gene lies at the chromosome 10 breakpoint in lymphoid neoplasias with chromosomal translocation t(10;14). Blood 1991; 78: 2996–3003.
Hatano M, Roberts CW, Minden M, Crist WM, Korsmeyer SJ . Deregulation of a homeobox gene, HOX11, by the t(10;14) in T cell leukemia. Science 1991; 253: 79–82.
Vilmer E, Suciu S, Ferster A, Bertrand Y, Cavé H, Thyss A et al. Long-term results of three randomized trials (58831, 58832, 58881) in childhood acute lymphoblastic leukemia: a CLCG-EORTC report. Leukemia 2000; 14: 2257–2266.
Donadieu J, Auclerc MF, Baruchel A, Perel Y, Bordigoni P, Landman-Parker J et al. Prognostic study of continuous variables (white blood cell count, peripheral blast cell count, haemoglobin level, platelet count and age) in childhood acute lymphoblastic leukaemia. Analysis of a population of 1545 children treated by the French Acute Lymphoblastic Leukaemia Group (FRALLE). Br J Cancer 2000; 83: 1617–1622.
Dombret H, Gabert J, Boiron J-M, Rigal-Huret F, Blaise D, Thomas X et al. Outcome of treatment in adults with Philadelphia chromosome-positive acute lymphoblastic leukemia – results of the prospective multicenter LALA-94 trial. Blood 2002; 100: 2057–2066.
European Group for the Immunological Characterization of Leukemias. (EGIL): Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, van't Veer MB . Proposals for the immunological classification of acute leukemias. Leukemia 1995; 9: 1783–1786.
ISCN 1995. An International System for Human Cytogenetic Nomenclature. In: Mitelman F (ed). Basel, Switzerland: Karger, 1995.
Ballerini P, Blaise A, Busson-Le Coniat M, Su X-Y, Zucman-Rossi J, Adam M et al. Hox11L2 expression defines a clinical subtype of pediatrc T-ALL associated with poor prognostic. Blood 2002; 100: 991–997.
Mauvieux L, Leymarie V, Helias C, Perusson N, Falkenrodt A, Lioure B et al. High incidence of HOX11L2 expression in children with T-ALL. Leukemia 2002; 16: 2417–2422.
Whitlock JA, Raimondi SC, Harbott J, Morris SW, McCurley TL, Hansen-Hagge TE et al. t(5;14)(q33-34;q11), a new recurring cytogenetic abnormality in childhood acute leukemia. Leukemia 1994; 8: 1539–1543.
Hansen-Haage TE, Schäfer M, Kiyoi H, Morris SW, Whitlock JA, Koch P et al. Disruption of the RanbBP17/Hox11L2 region by recombination with the TCRd locus in acute lymphoblastic leukemias with t(5;14)(q34;q11). Leukemia 2002; 16: 22205–22212.
Begley CG, Green AR . The SCL gene: from case report to critical hematopoietic regulator. Blood 1999; 93: 1760–1770.
Aplan PD, Johnson BE, Russell E, Chervinsky DS, Kirsch IR . Cloning and characterization of TCTA, a gene located at the site of a t(1;3) translocation. Cancer Res 1995; 55: 1917–1921.
Ferrando AA, Neuberg DS, Staunton J, Loh ML, Huard C, Raimondi SC et al. Gene expression signatures define novel oncogenic pathways in T cell acute lymphoblastic leukemia. Cancer Cell 2002; 1: 75–87.
The Belgian contribution to this work presents research results of the Belgian Programme of Interuniversity Poles of Attraction initiated by the Belgian State, Prime Minister's Office, Science Policy Programming. We own scientific responsibility of this work.
The Appendix lists the name of the city, the name of the institute, followed by the names of other participants and the number of cases in parentheses.
Toulouse: Centre Hospitalier Universitaire Hôpital Purpan: EE Kuhlein, E Duchayne, A Robert, F Huguet (53).
Paris: Hôpital Trousseau et CHU Saint Antoine: M Adam, L Douay, J Landman-Parker, G Leverger (45).
Paris: Hôpital Saint Louis: X-Y Su, M-T Daniel (40).
Louvain: Katholieke Universiteit Leuven: A Uyttebroek, C Doyen, A Delannoy, B Deprijck; Université Catholique de Louvain: C Verellen-Dumoulin, JM Libouton, JP Scheiff, V Deneys, C Vermylen, A Bruwier, D Latinne, J-L Vaerman, A Ferrant, A Bruniez (32).
Lyon: Hôpital Edouard Herriot and Hôpital Debrousse: I Tigaud, E Callet-Bauchu, D Treille-Ritouet, X Thomas (30).
Dijon: Centre Hopitalier Universitaire: B Favre, M Maynadié, D Caillo (25).
Lille: Centre Hospitalier Universitaire: P Lepelley, C Roumieux, N Grardel (18).
Nantes: Centre Hospitalier Universitaire Hôtel-Dieu: H Hervé-Loiseau (18).
Strasbourg: Centre Hospitalier Régional Universitaire: M Lessard, V Leymarie, C Gervais (18).
Marseille: Institut Paoli-Calmettes: D Sainty, C Arnoulet, R Bouabdallah, D Coso, A Charbonnier, D Blaise, A-M Stoppa, N Vey (14).
Reims: Centre Hospitalier Universitaire: I Luquet, S Daliphard (14).
Paris: Centre Hospitalier Universitaire Necker-Enfants Malades: S Romana (12).
Bordeaux: Centre Hospitalier Universitaire Hôpital Haut Lévêque: J-P Vial, F Lacombe, J-M Boiron (11).
Paris: Hôpital Pitié-Salpétrière: H Merle-Béral, CERBA: H Mossafa (11).
Gand: Centre for Medical Genetics, Belgique: J Philippé, B Verhasselt, Y Benoit (8).
Rouen: Centre Anticancéreux Henri Becquerel: A Stamboulas, MP Callat (7).
Marseille: Hôpital Timone-Enfants: H Zattara (3).
Liège: Université de Liège, ULCB, Belgique (3).
Nancy: Centre Hospitalier Universitaire de Nancy Brabois: P Jonveaux, J Buisine, L Mansury (2).
About this article
Advances in Biological Regulation (2019)
The American Journal of Surgical Pathology (2019)
Insertional oncogenesis by HPV70 revealed by multiple genomic analyses in a clinically HPV‐negative cervical cancer
Genes, Chromosomes and Cancer (2019)
Cancer Letters (2018)
Blood Advances (2017)