One year ago on 1 October 2012, Dr Roland Berger, one of the most respected pioneers in the field of cytogenetics, died at the age of 78 years (Figure 1). A number of colleagues and friends, many of whom had known Roland since the 1950s, endorse that he was a unique scientist in his ability to combine remarkable intelligence and insight with French skepticism. Their contributions are collected into this article reflecting their memories and admiration of his scientific career.

Figure 1
figure 1

A rare photograph of Dr Roland Berger taken in 1995.

Full size image

Early years in cytogenetics

In the late 1950s, Marthe Gautier, a pediatrician, established the first laboratory in France for the study of genetics at the pediatric Trousseau Hospital on the eastern edges of Paris, as part of the service of Raymond Turpin. She was central to the successful development of tissue culture techniques used at that time. However, Marthe chose not to pursue her career in genetics and was succeeded as head of the laboratory by Jerome Lejeune. In 1962, when Roland had just completed his 4-year internship as a pediatrician in Turpin’s department, he wished to transfer to a career in research. He was tempted by the new discipline and technology of cytogenetics, which offered a vast and diverse field of research. Lejeune’s laboratory was ideally situated for this venture both in time, with the recent discovery of the trisomy 21 in Down syndrome, and in place, in that he was able to transfer seamlessly from one career to the other in the same location. Roland successfully convinced Lejeune to employ him and gratefully abandoned the formalities of the medical profession to develop his new career in research, under the watchful eye of the pediatric medical team. With Marthe among them, they used to visit the laboratory on a regular basis. They could be heard from afar, approaching down the corridor with the precise tap of their well-polished black shoes, each of them in a dark navy full-length cloak. Their arrival gave the air that Napoleon and his generals had arrived; however, when they sank to rest on the laboratory chairs and lit up their Gauloises it was a different story.

Roland’s initial studies were aimed at understanding the mechanisms behind constitutional chromosomal abnormalities in children. At that time, cytogenetic analysis was carried out on fibroblast cultures from skin biopsies, which, although lengthy to preform, was highly successful. Roland learnt how to take and culture skin cells among other laboratory techniques, including preparation of karyograms from photographs with scissors and glue. Everything seemed possible in cytogenetics in those early days: Lejeune hoped that by incubating human chromosomes with extracts from fruit fly salivary glands they would produce polytene chromosomes; one of his less successful ideas. Diagnosis of new congenital malformations was the laboratory’s daily fare, such as the excitement when Lejeune’s cry went up ‘Come and see this child who cries like a cat!’.1 One of Roland’s first publications was written in a French journal in 1963, co-authored by Lejeune and titled ‘On a case of 17 trisomy’.2

Roland presented his doctoral thesis in 1968 entitled ‘Méthodologie de l'Analyse des Chromosomes des Tumeurs’. As it was a thesis in the Faculty of Sciences, Paris, he needed an additional certificate in ‘Physiologie Cellulaire’ in order to qualify. After work he would dash through Paris in his ‘Deux Chevaux’ in order to arrive on time for the lectures. Roland expanded his knowledge by presenting his cases at scientific conferences, notably the International Congress of Genetics, The Hague, The Netherlands, 1963, where he first met Herman Van den Berghe, and the International Genetic Congress, Paris, 1970, with its explosion of different chromosomal-banding techniques, which followed the fluorescence method using Quinacrine.

Subsequently, Roland left Lejeune’s laboratory and founded one of his own. He arrived in Hôpital Saint Louis, the large French hematological center, in the 1970s. At that time, bone marrow transplantation of Fanconi’s anemia was hopeless: the children suffered enormous toxicity. The cytogenetics team demonstrated a high level of chromosome breakage in cells from Fanconi’s anemia patients treated with cyclophosphamide metabolites; drugs used in the transplant-conditioning procedure.3 As a result of these findings, Elaine Gluckman subsequently drastically reduced the use of this drug in Fanconi’s anemia patients, paving the way to the first successful cellular therapy for a genetic disease. Reflecting his continuing interest in Fanconi’s anemia, Roland published ‘Chromosomal studies of leukemic and preleukemic Fanconi’s anemia patients’: examples of acquired ‘chromosomal amplification’ with his loyal, long standing and devoted technician, Maryvonne Busson-Le Coniat.4

In Roland’s early years as a cytogeneticist, he also published a wide range of cytogenetic topics including the following: constitutional abnormalities, gene localization and chromosome mapping. Not deterred by their high failure rate and difficulties of chromosomal analysis, he developed an interest in malignancy and transferred to the study of hematological samples. He was interested in the effect of constitutional chromosomal abnormalities in leukemia.5 His first original experimental paper on leukemia concerned a Down’s syndrome baby with congenital leukemia who had acquired a leukemic clone with 54 chromosomes and subclones with duplications of each of the chromosomes involved in the main clone.6 He had entered the realms of understanding the way in which cancer progressed and introduced the term ‘clonal evolution’. In fact, in the book ‘Human Chromosome’, published 2 years later, it was concluded that ‘chromosomal abnormalities were the neoplastic process itself’. In the early 1980s, he produced several papers with leukemia or cancer as their principal theme, including acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), lymphoma and solid tumors.7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 We surely now, more than 50 years later, have the answers to the fundamental question he asked: are chromosomal abnormalities the cause or the consequence of cancer? In those days, with no bands on chromosomes, making the recognition of abnormal chromosomes almost impossible, he nevertheless developed a ‘global graphic technique’ or ‘profile method’ for comparing the karyotypes within and between over 40 malignant solid and liquid tumor effusions after direct or short-term culture.

International workshops on chromosomes in leukemia

Roland was an important participant in the International Workshops on Chromosomes in Leukemia of which there were six held over a period of more than 10 years. The reports resulting from these Workshops are listed in Table 1. They were an outgrowth of workshops on histocompatibility antigens organized by Rose Paine and others, including Walter and Julia Bodmer, Oxford. After the revolutionary cytogenetic findings achieved using chromosome banding, it was crucial to investigate how reproducible they were and what valid clinical and pathological correlations could be made. In general, no cytogenetics laboratory had more than 10–20 cases with any particular chromosomal abnormality; therefore, collaborations between laboratories and pooling of their data were critical if progress was to be made.

Table 1 Reports from the Six International Workshops on Chromosomes in Leukemia
Full size table

The First Workshop was held in Helsinki in 1977, in connection with the International Chromosome Conference organized by Albert de la Chapelle. There were 16 participants, three from the United States of America and the remainder from six European countries. They provided data on 223 patients with chronic myeloid leukemia and 241 patients with AML. One aim of the Workshop was to characterize chromosomal abnormalities in AML; only two translocations, t(8;21)(q22;q22) and t(15;17)(q21;q11), were identified. Among 17 patients with acute promyelocytic leukemia (APL), only nine had t(15;17) and six were thought to be normal. The Second Workshop was held in Leuven in 1979, hosted by Herman Van den Berghe, to which Roland was invited to participate. From the submitted data, the same division was noted; among 80 APL patients, 33 had the translocation and 40 had a normal karyotype. It was observed that the incidence of the translocation increased after several days of bone marrow cell culture. Thus, the close association of morphology and translocations was confirmed—t(8;21) and t(15;17) related to particular subtypes of AML—namely AML M2 and AML M3 (and AML M3 variants), respectively.

The Third Workshop in 1980 saw a change in several ways. It was organized and financed by Felix Mitelman and his colleagues in Lund. The subject was ALL and clinicians were invited for the first time. With 330 cases, about 50% children, one could not only compare the incidence of various abnormalities in children versus adults but also compare survival among the different abnormalities. It also became apparent that the high hyperdiploid group, greater than 50 chromosomes, was different from the hyperdiploid (47–49 chromosomes), both cytogenetically and clinically. It became evident that patients with karyotypes with a modal number >50 chromosomes had the best response to treatment, as reported earlier by Lorna Secker-Walker,19 whereas those with t(4;11) <46 chromosomes or Philadelphia chromosome had the worst outcome. For the first time, it was shown that karyotype was an essential and, in fact at that time, the most important independent predictor of survival. These associations are used for risk stratification of ALL patients for treatment to this day.

The Fourth Workshop in 1982, organized by Janet Rowley in Chicago, was the first one where data were collected prospectively. The planning for this Workshop began at the Second Workshop in 1979. It focussed on AML and included every patient with a successful cytogenetic result. It had the largest number of participants31 and included the original member cytogeneticists as well as invitées from South Africa, Australia and Japan, who contributed 660 de novo and 56 treatment-related AML. In addition to t(8;21) and APL, 30 patients had abnormalities of the long arm of chromosome 11 (11q) and a new group was observed, tentatively called inversion.16 Patients with treatment-related AML had complex karyotypes with abnormalities of chromosomes 5 and 7. The preponderance of translocations in younger patients less than 15 and up to 30 years, and of chromosome 5 and 7 abnormalities, as well as complex karyotypes in older individuals, was striking. Moreover, karyotype was the most important independent prognostic factor for survival.

The Fifth Workshop was held in Japan in 1984. One of the most striking findings was the difference in frequency of various clinical/pathological subtypes according to geography. Among 57 cases of follicular lymphoma with t(14;18), only one came from Japan! The last Workshop in 1987 was held in London. Its focus was the integration of cytogenetics with molecular genetics, especially cloned translocation breakpoints. This Workshop also examined long-term (>5 year) survival of the 66 AML patients who remained alive from the Fourth Workshop and whose clinical data were available for analysis at the Sixth Workshop. The predictors of long survival were younger age at diagnosis and translocations, t(8;21) and t(15;17), as well as inversion,16 with 10–30% alive at 10 years compared with those with chromosome 5 and 7 abnormalities, with no survivors after 6.5 years. Roland Berger was a major contributor of patient data but also brought scientific rigor to these meetings.

Banding (and DNA) years

With Georges Flandrin’s team, Roland focused on the study of AML. Among their multiple findings, they showed that in APL, erythroblasts with good-quality metaphases and a normal karyotype masked the poor quality leukemic metaphases with t(15;17). This observation explained the frequent karyotype discrepancy that consequently was efficiently overcome by a simple 2-day culture, as noted in the Second International Workshop. This observation was instrumental in defining the specificity of t(15;17) and its biological meaning, leading to the first model of targeted therapy in APL with retinoids.

The biological signification of translocations was the central question in those years. During his work on Burkitt’s lymphoma, Roland’s group was the first to locate the translocation breakpoints of the immunoglobulin heavy and light chains, thereby implicating the location of an oncogene, later identified as MYC, on 8q24, to be dysregulated by a position effect.8, 9, 10 From this finding, the convergence of cytogenetics and molecular biology began, leading to the discovery of large numbers of significant genetic alterations associated with translocations in malignancies. Roland was mindful of the importance of human gene mapping and he became the chair of the first committee on ‘Chromosome rearrangements in acquired malignant diseases’ established in 1981 in Human Gene Mapping 6 conference in Oslo. In his own laboratory, he pioneered molecular cytogenetics by introducing chromosome sorting, chromosome microdissection, in situ hybridization with banding and early fluorescence in situ hybridizationof unique sequences.

Roland had a keen interest in abnormalities of 11q, which culminated in the publication of a review with Olivier Bernard highlighting the ‘Molecular Basis of 11q23 Rearrangements in Hematopoietic Malignant Proliferations’.20 He was an active participant in the ‘European Union Concerted Action Workshop on 11q23’, organized by Lorna Secker-Walker in London in 1997.21, 22 This Workshop defined the individual 11q23 partners both cytogenetically and clinically; these papers continue to be referred to today.

One highly significant finding was Roland’s discovery of the cryptic translocation, t(12;21)(p13;q22), with his then PhD student, Serge Romana.23, 24, 25 This late discovery in 1994 was made by the application of chromosome paints to metaphases from patients with abnormalities of chromosome 12. The molecular nature of the translocation arising from a fusion between the TEL and AML1 genes was made simultaneously by Roland’s group and Todd Golub in United States of America in 1995,26 as well as the strong association of this translocation with deletions of the short arm of the chromosome 12 not involved in the translocation. The TEL–AML1 fusion (now known as ETV6–RUNX1) occurs in ∼25% childhood B-lineage ALL. It is associated with a young age and an excellent outcome, thus remaining important for risk stratification for treatment.

In the discovery of novel chromosomal abnormalities, Roland’s group was one of the first to identify that often in ALL, marker chromosomes in the absence of one copy of chromosome 21 were duplicated chromosome 21 with multiple copies of the AML1 gene.27, 28 This intriguing abnormality now defines a novel cytogenetic subgroup, named intrachromosomal amplification of chromosome 21 (iAMP21), with a poor outcome on standard therapy.29, 30, 31, 32, 33 Intensified treatment has been shown to significantly improve outcome, which has become standard practice on clinical trials.34, 35 From these few examples, it is evident that his cytogenetic discoveries in human leukemia led to improved precision in classification of leukemic subtypes combined with a deeper understanding of their prognostic significance.

Professional duties

Roland was an ethical man with a sound perception of ‘public service’. He was known as an ‘activist’ because, through his enthusiasm for the importance of acquired cytogenetic abnormalities, he persuaded the ‘Securité Sociale’ (the French National Health Service) that cytogenetic testing should be included in the list of diagnostic procedures that were reimbursed, by pleading that they were essential to the optimum care of patients who suffered from hematological diseases. Thanks to his efforts at the French Ministry of Health, the external quality assessment for the field of acquired cytogenetics was initiated.

Roland strongly considered teaching and mentoring of his students as an essential part of his professional duty. He trained a number of pupils who continue in cytogenetics or research activity in French public institutions. He mentored Philippe Jonveaux, Serge Romana, Florence Nguyen-Khac towards University qualifications and Alain Bernheim and Olivier Bernard in research careers. He inspired many cytogeneticists and researchers at some point in their lives, for which the list is too long to be included.

Groupe francophone de cytogénétique hématologique

Roland succeeded in maintaining cytogenetics as an active domain in France. Together with Joseph Tanzer in 1978, Roland was one of the founders and leading scientists of the Groupe Français (later changed to Francophone) de Cytogénétique Hématologique (GFCH). This group performed excellent collaborative studies and produced a series of authoritative papers, which characterized novel and established cytogenetic subgroups. Roland was immersed in GFCH studies, both for the subject, the studies themselves, and article writing, which was eloquently undertaken. He led the group until his retirement in 2008. The GFCH meant a lot to him, and he followed with interest their ongoing work, long after his retirement. It was important that this group continued after him; hence, two of his pupils took his place: Serge Romana (2008–2012) and Florence Nguyen-Khac (from 2012). At this time, 56 original articles have been published on behalf of the GFCH, which are listed in Table 2. His enormous commitment to GFCH was equal to every cause that he deemed worthy of fighting for. He is fondly remembered by his friends and colleagues with quotations listed in Table 3.

Table 2 Articles from Groupe Français de Cytogénétique Hématologique
Full size table
Table 3 Quotations of fond memories from friends and colleagues of Dr Roland Berger
Full size table