The journal "impact factor": a misnamed, misleading, misused measure.

The Institute for Scientific Information (ISI), a database publishing company that publishes Current Contents and Science Citation Index, has devised and promulgated what it terms the journal "impact factor." ISI describes this factor as a "measure of the frequency with which the 'average article' in a journal has been cited in a particular year." The factor is a ratio between citations and recent citable published items calculated by dividing the number of all current citations of items published in a journal during the preceding 2 years by the number of articles published in those 2 years by that journal. What, if anything, is wrong with the "impact factor"? There is absolutely nothing incorrect with the calculation of the ratio itself. However, the "impact factor" is misnamed and misleading. Being misnamed and misleading, the "impact factor" has been misused. It is being held out as a measure of the importance of a specific journal article and the journal in which the article appeared. By extension, the "impact factor" is also being misused to gauge the relative importance of individual researchers, research programs, and even the institution hosting the research. We recommend that the term "impact factor" be abolished and that this measure be renamed in keeping with its actual role, that merely of a time-specific "citation rate index" and nothing more. What is currently called the "impact factor" should not be misused to evaluate journals or to validate the scientific relevance of a particular researcher or research program, especially in decisions regarding employment, funding, and tenure.

BCL3 rearrangements and t(14;19) in chronic lymphocytic leukemia and other B-cell malignancies: a molecular and cytogenetic study.

The t(14;19)(q32.3;q13.1) is a recurring translocation found in the neoplastic cells of some patients with chronic lymphocytic leukemia (CLL) or other B-lymphocytic neoplasms. We previously cloned the translocation breakpoint junctions present in the leukemic cells from three such patients and identified a gene, BCL3, whose transcription is increased as a result of the translocation. In the present paper, we describe three additional patients with the t(14;19), one with lymphoma and two with CLL, and report the cloning and sequencing of the breakpoint junction in one of these patients as well as in a previously reported patient. We and others have found that the breakpoints on chromosome 14, with one exception, fall within the switch region upstream of the immunoglobulin heavy chain C alpha 1 or C alpha 2 sequences. Several of the breaks within chromosome 19 fall immediately upstream of the BCL3 gene, but several others are more than 16 kb 5' of the gene. Most patients with CLL and the t(14;19) also show trisomy 12.

Mapping FRA11A, a folate-sensitive fragile site in human chromosome band 11q13.3.

FRA11A, a rare folate-sensitive fragile site assigned to 11q13.3, lies in an area of genomic instability associated with several diseases and amplification events. To map FRA11A, we used fluorescence in situ hybridization with yeast artificial chromosome and cosmid probes on metaphase chromosomes of patients expressing the fragile site. FRA11A was found situated centromeric to ACTN3 and telomeric to D11S913, these markers being within an interval of approximately 1 Mb in the 11q13.3 region.

Cytogenetics of malignant gliomas: I. The autosomes with reference to rearrangements.

Autosomal chromosome abnormalities are far from always detectable and, when detected, far from fully consistent in malignant gliomas. In 15 of 41 malignant gliomas, we found autosomal chromosome aberrations ranging from solitary trisomy to a wildly abnormal polyploid complement. The sequence of chromosome events appears to proceed from the normal to the near-diploid state (via structural and numerical changes) to near-tetraploidy (via polyploidization), and finally toward near-triploidy (via chromosome loss and additional rearrangements). Characteristic chromosome changes of trisomy 7 and monosomy 10 were repeatedly found, usually together in the same cell clones. In only one case was trisomy 7 an isolated change. We observed structural rearrangements of chromosomes 7 and 10 which may be of some use in mapping specific genes duplicated or deleted by the whole-chromosome changes of chromosomes 7 and 10. Nonrandom structural changes of other autosomes, including chromosomes 1, 5, and 11, fit with the model of malignant glioma as a process involving multiple genes. An unusual concentration of breakpoints in 12q13, juxtaposing it to at least five other regions, reflects the presence of genetic information in 12q13 important to the development of malignant gliomas.

Cytogenetics of malignant gliomas. II. The sex chromosomes with reference to X isodisomy and the role of numerical X/Y changes.

Sex chromosomal monosomy with total loss of an X or Y is frequently observed in malignant gliomas. Beyond that, not much is known about the behavior of the sex chromosomes in these tumors. We noted loss of the X from 3 of 13 gliomas from women (23%) compared to loss of the Y from 16 of 28 gliomas from men (57%). There were two structural rearrangements of the Y (an inversion and a translocation with chromosome 4). Most unexpectedly, clones with sex chromosome reversal were encountered in 3 cases. These XX clones in gliomas from men are perforce the consequence of Y loss coupled with X isodisomy, a nonrandom sequence of sex chromosome changes. We examined the company kept by numerical X and Y changes in clones and found that clones with numerical sex chromosome changes had fewer autosomal abnormalities, reflecting a distinct tendency to clonal separation of sex chromosome from autosomal abnormalities. We conclude that the sex chromosome changes are not a necessary part of the neoplastic process in malignant gliomas but that they must be of biologic significance to the brain since they are highly nonrandom in frequency, type, and sequence in brain cells.

Chromosomes in gliomatosis cerebri.

Gliomatosis cerebri is a rare brain tumor which histologically resembles a diffuse cerebral astrocytoma. It can simultaneously infiltrate multiple sites in the cerebrum, cerebellum, brainstem, and spinal cord. This remarkable diffuseness has led to the idea that gliomatosis cerebri does not derive from a solitary focus but must arise from a broad field of glial cells. We studied the chromosomes from gliomatosis cerebri in a 12-year-old boy by conventional cytogenetics and fluorescence in situ hybridization (FISH). Aside from normal cells, we found a majority of cells with the karyotype 44,XY,del(6)(q25),del(14)(q21), der(15;21)(q10;q10),add(18)(q22),del(19)(p12),add(20)(p13),-21. A smaller proportion of cells had 88 chromosomes with a doubling of this abnormal karyotype. These findings are consistent with a clonal neoplasm stemming from a single cell. The chromosome changes we observed, with the possible exception of the chromosome 6 deletion, did not resemble those frequently found in astrocytomas. Gliomatosis cerebri may therefore belong to a separate category of brain tumors.

X;6 translocation in a child with congenital acute lymphocytic leukemia.

A case of congenital acute lymphoblastic leukemia (ALL) displayed an X;6 translocation. This is the third reported case of ALL with an X;6 translocation. In addition, two of the three ALL cases occurred during infancy, at ages 2 months and newborn, and both translocations involved the band q15-16 region of chromosome 6. Anomalies of the long arm of chromosome 6, mainly interstitial and terminal deletions, have been reported as a recurrent karyotypic event in a significant number of ALL cases. The molecular basis and propensity of an X;6 rearrangement in this case of congenital ALL is unclear and merits further investigation. The similarities in this case and the other infant ALL case cited suggest that an X;6 rearrangement with a breakpoint in bands q15-16 of chromosome 6 is characteristic of a form of congenital ALL.

Cancer in ataxia-telangiectasia patients.

A gene locus for ataxia-telangiectasia (A-T) is in chromosome region 11q22 to 11q23 and predisposes to cancer. Ataxia-telangiectasia patients appear to have two separate clinical patterns of malignancy. One pattern involves solid tumors, which have not been stressed and which include malignancies in the oral cavity, breast, stomach, pancreas, ovary, and bladder. Detection of a solid tumor in an A-T patient should serve as a warning. It heralds a markedly elevated risk of another malignancy in that patient. The second pattern of neoplasia in A-T is well recognized and consists of lymphocytic leukemia and non-Hodgkin's lymphoma. These malignancies may relate to immunodeficiency in A-T and to chromosome breakage and rearrangement, which are a feature of A-T. These two patterns of malignancy may be truly separate and reflect different mechanisms of malignancy in A-T, or they may not really be separate but instead reflect a single mechanism of malignancy. The situation in A-T is reminiscent of that in the acquired immunodeficiency syndrome (AIDS), in which Kaposi's sarcoma occurs with mild immunodeficiency and pneumocystis carinii pneumonia occurs with more profound immunodeficiency owing to the human immunodeficiency virus. Next to pulmonary disease, cancer is the leading cause of death in A-T.