Perspectives On Animal Research

Volume 1, Supplement

An Evaluation of Ten Randomly-Chosen Animal Models of Human Disease

Large Granular Lymphocytic Leukemia (Rats)

Description of the Model:

The large granular lymphocyte (LGL) is important to the natural killer and antibody-dependent cellular cytotoxicity activity of the human immune system. Proliferations of LGL in people have heterogeneous presentations, and there are several neoplastic disorders that appear to be related. This has resulted in a confusing nomenclature in the literature, in which classifications often differ markedly. Loughran and Starkebaum summarized:

In 1977, McKenna et al. described a distinct clinicopathologic entity consisting of chronic neutropenia associated with increased numbers of circulating LGL. Since then additional cases of this syndrome have been described under a variety of names ... This confusing terminology arose in part because of the difficulty in determining whether this is a reactive or a neoplastic disorder. Two lines of evidence, however, indicated that most, if not all, of these patients have a neoplastic disorder ... Because of the neoplastic nature of the disorder, combined with the finding of infiltration of spleen, bone marrow, and liver by abnormai lymphocytes, we have proposed the term 'LGL leukemia.'(1)

They explained:

Initially, the term T-gamma lymphoproliferative disorder was used ... (but) occasional T-gamma lymphoproliferative disorders are reported that do not involve LGL ... Finally, the term 'T-cell CLL' is too broadly encompassing, since it includes malignancies arising from other T-cell populations...(1)

In 1970, Moloney et al. first noted spontaneous leukemias in older Fischer rats. They observed:

Leukemias of this cell type do not occur in man. On morphological and histochemical grounds, the cell cannot be classified as granulocytic, lymphocytic, or monocytic. However, the leukemic cells do have some characteristics of reticulum cells.(2)

Subsequent investigations indicated that there were similarities between these rat leukemias and human LGL leukemia.(3) Stromberg concluded:

Besides serving as a model for T-cell leukemias, LGL leukemia should provide investigators with abundant cells for the studies of certain aspects of LGL and NK-cell function. In addition, it is an excellent, reproducible model of tumor-induced autoimmunity. The ease of handling, high frequency of successful transplantation, and short latency period mark it as a convenient system for use.(3)

Criterion I: Concordance between the Animal Model and the Human Disease

Clinical Presentation:

Stromberg et al. published a series of three papers in 1983 that evaluated LGL leukemia in F344 rats morphologically, hematologically, and biochemically.(4-6) They found that this condition was a heterogeneous disorder with a wide range of clinical presentations. Among the salient features, they found splenic origin, liver involvement, and anemia, which was often Coombs-positive and was often associated with erythrophagocytosis.(4,5)

In 1985, Reynolds reviewed LGL leukemia in people and in animals.(7) He noted many similarities, including natural killer and antibody-dependent cellular cytotoxicity of tumor cells, probable splenic origin, liver involvement, and anemia. However, unlike the rat model, erythrophagocytosis was rare in human patients and it did not appear to contribute significantly to the anemia.(7) Furthermore, "Coombs' tests for the detection of autoantibodies to red cells have generally proven negative"(7) in people, but it is often positive in the rats.(5) Although the cause of the anemia is unclear in both human and rat LGL leukemia, it appears to be different.

Another difference between these diseases is the presence of skin lesions in human LGL leukemia, which are not found in the rat condition.(7) Perhaps most significantly, the course of the leukemia differs markedly between humans and rats.(7) In rats, it is an acute disease that has a rapidly progressive, fatal course. In contrast, people with LGL leukemia have a chronic disease so indolent that Loughran and Starkebaum wrote, "...the question remains whether the disease is truly malignant. In most patients, the disease appears to progress slowly if at all, and they generally do not require therapy."(1) Thus, despite some similarities between human and rat LGL function and LGL leukemia, there appears to be a fundamental difference in the clinical presentation of the disease.

Pathogenesis:

Little is known about the process of malignant transformation of LGL cells in rats or in people. Differences in clinical presentation, particularly in natural history, suggest different disease processes. However, this is not conclusive.

Criterion II: Citations

We located 41 articles from the Science Citation Index, 1983-1988, which cited at least one of the five reports describing this animal model.(2,4-6,8) (See Appendix A.) Of these, five were studies that included human patients or human tissue.

Buessow et al. referred to Ward and Reynolds' 1983 article as well as human clinical reports from 1982 and 1983 to support the following statement: "In light of recent reports concerning the heterogenicity of NK cell populations, it is tempting to speculate that HL-60 cytotoxicity may be another example of natural killer cell diversity."(9)

Similarly, Caulfiel et al. cited the 1983 paper by Ward and Reynolds as well as several clinical observations, some prior to 1983, when they noted:

The morphology of the granules in unstimulated granular lymphocytes is extremely variable within individual cells, within populations of cells, and among cells derived from different species and sources, e.g., blood, leukemias, cell lines, or clones.(10)

The reports by Loughran and Starkebaum (1) and by Semenzato et al.(12) will be discussed in greater detail in the next section. Loughran et al. reported three patients with leukocytosis of large granular lymphocytes. They observed:

The location of the neoplastic infiltrate in red pulp cords is consistent with the distribution of a subpopulation of natural killer cells in normal splenic tissue. Our findings are similar to those seen in the rat model of large granular lymphocyte leukemia in which tumor cells infiltrate splenic red pulp cords.(11)

In all five papers, including those reviewed below, the authors cited the animal model only to demonstrate parallels between the animal model and the human disease. None of the authors of these papers credited the animal model with any important insights into the pathophysiology, diagnosis, or treatment of LGL leukemia in man.

Criterion III: Historical Impact

All recent discussions of LGL leukemia in people have included clinical cases. Two recent reports of human LGL leukemia help to determine whether the animal model has made significant contributions. In 1987, Loughran and Starkebaum reported 38 cases and reviewed the literature.(1) They cited the paper by Ward and Reynolds in support of the asplenic origin of the tumor. They stated, "...in a rat model of LGL leukemia the spleen appears to be the site of origin of the leukemic cell."(1) They also cited two observations from human clinical studies indicating that the tumor originates in the spleen. However, it is not certain that the spleen is the source of LGL leukemia in rats.(4) Also, given the widespread distribution of LGL cells in human beings, it is not known how often the spleen is the primary source of the neoplasm. Loughran and Starkebaum also cited the animal model when they noted that several investigators have found lymphocytic infiltration of the lamina propria of the intestines in human patients and in experimental LGL leukemia. This does not represent an insight into human LGL but rather an observation of one of the many similarities between human and rat LGL leukemia.

The only citation of the rat model among the 46 references in a paper entitled "The lymphoproliferative disease of granular lymphocytes" by Semenzato et al.(12) was the 1983 paper by Ward and Reynolds. They commented, "It is of interest that the hepatosplenomegaly consistently observed in these patients is reminiscent of the rat model of large granular lymphocytic leukemia."(12) Once again, clinicians noted a parallel between the human and rat disease, but they did not credit the animal model with an important clinical insight. The other animal studies Semenzato et al. cited were two papers using mice to identify a marker for certain T-cells.

Because human LGL leukemia is uncommon, proponents of the animal model had observed that F344 rats could be an abundant source of cells for studies of LGL and natural killer cell function. The development of in vitro techniques has permitted these studies with human cells. Reynolds concluded:

...the results from a number of human and animal studies suggest that LGL leukemias/lymphomas may be quite useful in further studying the function of LGL ... Further studies which require a large number of highly purified functionally active cells can be easily done using LGL leukemia cells from either the human or animal models(7)

Conclusions:

Despite many similarities between human and rat LGL leukemia, their clinical courses differ markedly. This raises doubts about the usefulness of the animal model for providing insights into the pathophysiology, diagnosis, or treatment of human LGL leukemia. Indeed, the two recent clinical reviews did not appear to attribute any important findings to the animal research. The usefulness of the rat model in providing abundant cells for the study of LGL function has been diminished greatly by in vitro study of normal and leukemic human LGL cells.

References:

1. Loughran TP, Starkebaum G: Large granular lymphocyte leukemia: Report of 38 cases and review of the literature. Medicine 1987;66:397-405.

2. Moloney WC Boschetti AE, King VP: Spontaneous leukemia in Fischer rats. Cancer Res 1970;30:41-43.

3. Stromberg PC: Large granular lymphocyte leukemia, Model No. 319, in Capen CC, Jones TC, Migake G (eds): Handbook Animal Models of Human Disease, Fasc 14. Washington DC, Registry of Comparative Pathology, Armed Forces Institute of Pathology,1985.

4. Stromberg PC, Vogtsberger LM: Pathology of the mononuclear cell leukemia of Fischer rats. I. Morphology studies. Vet Pathol 1983;20:698-708.

5. Stromberg PC, Vogtsberger LM, Marsh LR, Wilson FD: Pathology of the mononuclear cell leukemia of Fischer rats. II Hematology. Vet Pathol 1983;20:709-717.

6. Stromberg PC, Vogtsberger LM, March LR: Pathology of the mononuclear cell leukemia of Fischer rats. III Clinical chemistçy. Vet Pathol 1983;2th718-726.

7. Reynolds CW: Large granular lymphocyte (LUL) lymphoproliferative diseases: Naturally cytotoxic tumors in man and experimental animals. CRC Crit Rev Oncol/Hematol 1985;2:185-208.

8. Ward JM, Reynolds GW: Large granular lymphocyte leukemia: A heterogeneous lymphocytic leukemia in F344 rats. Am J Pathol 1983;111:1-10.

9. Buessow SC, Gillespie Y, Mahaley MS: Tumoricidal activity of an acute promyelocytic leukemia cell line (HL-60) is augmented by human interferon alpha. Leukemia Res 1984;5:801-811.

10. Caulfield JP, Hein A, Schmidt RE, Ritz J: Ultrastructural evidence that the granules of human natural killer cell clones store membrane in a nonbilayer phase. Am J Pathol 1987;127:305-316.

11. Loughran TP, Kadin ME, Starkebaum G, et al.: Leukemia of large granular lymphocytes: Association with clonal chromosomal abnormalities and autoimmune neutropenia, thrombocytopenia, and hemolytic anemia. Ann Int Med 1985;102:169-175.

12. Semenzato G, Pandolfi F, Chisesi T, et al.: The lymphoproliferative disease of granular lymphocytes: a heterogeneous disorder ranging from indolent to aggressive conditions. Cancer 1987:60:2971-2978