Non-Hodgkin's Lymphoma- family ties that bind!

 

The Swedish Family-Cancer Database is an extraordinary resource, containing a large amount of high quality data that essentially reflects the experience of the entire country. The database was firstly created in the mid-1990s by linking census information, death notifications, and the administrative families register at Statistics Sweden to the Swedish Cancer Registry. The Database was updated at the end of 2004 to include persons born in Sweden after 1931 with their biologic parents, totaling 10.5 million individuals. The Database is organized in 3.6 million nuclear families, with parents and offspring. A new study [ Altieri A et al, Blood epub ahead of print, April 2005] has examined familial risk of non-Hodgkin lymphoma (NHL) in a large number of families, with histology verified and classified in an up-to-date manner. Neoplasms were retrieved from the Swedish Cancer Registry from 1958 to 2002. The Swedish Cancer Registry is based on compulsory reports of individual cases provided by clinicians/pathologists or cytologists and is considered to have almost 100% completeness. Pathologists and cytologists separately reported every cancer diagnosis on surgically removed tissues, biopsies, cytological specimens, bone marrow aspirates and autopsies. Data on parity were complete, and data on socioeconomic index and area of residence were based on population censuses from the years 1960, 1970, 1980 and 1990. The data showed that having a parent or sibling with NHL almost doubled the risk of NHL (Standardized Incidence Ratio (SIR)=1.8, 95% Confidence Interval (CI) 1.4-2.2 and 1.9, 95% CI, 1.1-3.2 respectively). Of note, having a parent or sibling with T-cell NHL was not associated with an increase in risk.

 

COMMENT: This study increases our knowledge of familial influence on NHL. Importantly, ascertainment was essentially complete, and recall bias was erased by the use of a registry. The data show some familial influence on susceptibility to NHL that might reflect the known association of this disease with immune deficiency. The most striking feature of the report, however, is the power of this well-run and complete registry. This resource is the envy of those of us living in societies were compulsory data reporting and complete ascertainment are unlikely to be achieved in any academic setting.

Stella M. Davies

 

Nucleophosmin and AML, Etiology and Prognosis

 

Nucleophosmin (NPM) is a molecular chaperone that shuttles between the nucleus and cytoplasm and appears to regulate the assembly and transport of pre-ribosomal particles across the nuclear membrane. NPM has been thought to function as a tumor suppressor gene as it is a partner in a number of chromosomal translocations associated with leukemia and lymphoma, e.g. NPM-ALK and NPM-RAR a . Since the cellular localization of NPM is critical to its function Falini B et al , for the Italian GIMEMA acute leukemia working party [ New Engl. J Med. 2005, 352:254-66], hypothesized that changes in localization might contribute to leukemogenesis. These authors studied 591 patients with primary AML other than M3, and 135 with secondary AML, aged between 15 and 60 years. Cytoplasmic (i.e. altered) localization of NPM was found in 35.2% of cases of primary AML, but not in secondary AML or in 980 hematopoietic or non-hematopoietic neoplasms other than AML. Cytoplasmic NPM was associated with a wide spectrum of morphologic subtypes of AML, a normal karyotype, and lack of expression of CD34 and responsiveness to induction chemotherapy. AML specimens with cytoplasmic NPM carried frame-shift mutations in exon 12 of the NPM gene that were predicted to alter the protein at its C-terminal: this mutant gene caused cytoplasmic localization of NPM in transfected cells.

 

In a second paper, Alcalay M et al [Blood, epub ahead of print April 2005] report the gene expression profiles of 78 de novo AMLs (72 with normal karyotype; 6 with non-major chromosomal abnormalities) that were characterized for the subcellular localization and mutation status of NPM. Unsupervised clustering clearly separated AML cases with cytoplasmic NPM localization from those without, regardless of the presence of FLT3 mutations or non-major chromosomal rearrangements, supporting the concept that AML with cytoplasmic NPM represents a distinct entity. The molecular signature of AML with cytoplasmic NPM included up-regulation of several genes putatively involved in the maintenance of a stem cell phenotype, suggesting that AML with cytoplasmic NPM may derive from a multipotent hematopoietic progenitor.

 

In a third paper, Cazaniga G et al, [Blood, epub ahead of print April 2005] describe findings in 107 children with AML. The authors of this study sequenced NPM and identified mutations in exon 12 in 7 (6.5%) cases overall. All 7 were in children with normal karyotypes, accounting for 27% of cases with normal karyotypes. The effect of mutated NPM on prognosis was not addressed in this study.

 

COMMENT: These studies are an important step forward in AML biology. Major advances in the treatment of leukemia have depended on understanding genetic heterogeneity and using this knowledge to apply risk-adapted therapy to improve outcome. Approximately 40-50% of AMLs have no frequent recurring chromosomal abnormalities to guide risk assignment. The studies described above report that 60% of AML cases with no frequent recurring chromosomal abnormality will express cytoplasmic NPM, providing a new tool (along with flt3 internal tandem duplications) for risk assignment in this group, as those cases appear to have superior responses to induction therapy. The frequency of cytoplasmic NPM in pediatric AML remains to be reported and pediatric studies are currently underway. Cytoplasmic NPM can be detected simply and reliably with immunochemistry, so could be used widely and inexpensively for large-scale screening in a cooperative group setting. The technique requires biopsy slides only so can be used even in cases where aspirates are of poor quality or culture for standard cytogenetics has failed.

Stella M. Davies

Smoke'n gene(s) but no smoking gun

 

The majority of studies that have examined links between maternal cigarette smoking during pregnancy and risk of childhood leukemia have been null. In fact, some studies have suggested a possible protective effect. This has been true whether the study utilizes a case-control (e.g., interviewing mothers after the child has developed malignancy) or prospective cohort (e.g., interviewing mothers during pregnancy and then following them to see who develops malignancy) approach. Nevertheless, there have been some interesting studies that have evaluated the effect of maternal smoking on either carcinogens present in newborn urine or on potential biomarkers of risk (such as chromosomal aberrations). In this new study from Spain, Ana de la Chica R et al [JAMA 2005; 293:1212-1222] studied chromosome instability in amniocytes obtained from pregnant smokers and non-smokers. A total of 25 women in each group, all of whom did not consume alcohol, tea, or coffee, were recruited. A routine amniocentesis was performed and the investigators examined chromosomal instability by evaluating gaps and breaks, as well as structural abnormalities. When comparing the cytogenetic data between the two groups, amniocytes from smokers had a statistically significantly higher proportion of structural abnormalities among 793 and 752 metaphases examined, respectively (smokers 12.1%, controls 3.5%; p=0.002). Abnormalities appeared localized to certain chromosomal regions including 5q31 (8 smokers, 3 controls);17q21 (10 smokers, 5 controls) and 11q23 (6 smokers, 0 controls). The authors conclude that transplacental exposure to tobacco smoke could be associated with an increased risk of hematopoietic malignancies, given that 11q23 is involved in pediatric leukemias.

 

COMMENT: There are a number of comments to be made on this paper. It is notable that the authors had to interview 800 mothers to find 25 in each group that fit their criteria (smoker or non-smoker who did not drink alcohol, coffee or tea). This is no small task and demonstrates the large amount of work that often accompanies what seems like a relatively simple research question. This study is intriguing as it suggests that constituents of cigarette smoke directly affect certain loci on chromosomes. In particular, the finding of breakpoints on 11q23 solely in the amniocytes obtained from cigarette smokers suggests a possible direct link with infant leukemia. In our current study of infant leukemia, we took a quick look at maternal cigarette smoking. We found that maternal smoking during pregnancy was neither associated with MLL+ leukemia (odds ratio (OR) = 0.90; 95% CI: 0.47-1.73) nor with MLL- leukemia (OR = 0.89; 95% CI: 0.43-1.85) [personal communication, Spector LG and Ross JA]. Thus, even in light of the evidence that maternal cigarette smoking can cause genetic damage, the majority of epidemiological studies are not finding that maternal cigarette smoking influences risk of infant or childhood leukemia. It is unlikely that this finding is due to bias, since large cohorts from Europe have also reported negative associations. Nevertheless, evidence of fetal chromosomal aberrations that are directly linked to maternal cigarette smoking should serve as a strong impetus to not smoke while pregnant. Finally, it would also be of interest to determine whether these abnormalities persist after birth, as the in utero experience may have longer implications in health. Few studies have explored whether individuals exposed to cigarette smoke in utero are at an increased risk of malignancy later in life.

Julie A. Ross

 

Briefly Noted

 

Mutations in the perforin gene are familiar to pediatric hematologists as a cause of familial hemophagocytic lymphohistiocytosis. In a fascinating paper, Clementi R et al [ Blood 2005; 105:4424-4428] describe the presence of constitutional biallelic perforin mutations in 4 of 29 Hodgkin or non-Hodgkin's lymphoma cases screened. In 2/4 cases available for evaluation, perforin and NK cell activity were absent. In addition monoallelic perforin mutations were present in 4 cases. These data indicate that perforin deficiency may play an important role in susceptibility to lymphoma.

Stella M. Davies

 

C3 Mailing List

 

The editors of C3 are constantly updating the mailing list. If you know of colleagues including postdocs and trainees, who wish to receive our newsletter, please have them email us at pedsepi@umn.edu. There is no charge. The newsletter is also available at the website listed on the C3 header.

C3 Quarterly Newsletter
Children's Cancer Research Fund
Epidemiology Research Unit
Division of Pediatric Epidemiology
Clinical Research
University of Minnesota
420 Delaware St. SE, Box 422
Minneapolis, MN 55455
pedsepi@umn.edu

Editors: 
Stella M. Davies, MD, PhD, and Julie A. Ross, PhD