Two Mouse Models Illustrate Different Mechanisms of Anemia of Cancer

Anemia is a well-known complication of many cancers, and is associated with increased morbidity and shortened survival. Anemia of cancer (AC) has been assumed to have mechanistic similarities to the anemia of inflammation, but the roles of inflammation and hepcidin in the development of AC have yet to be elucidated. We used animal models of cancer to investigate the roles of inflammation and hepcidin in the development of AC.

We created two mouse models of cancer using the syngeneic lung and ovarian cancer cell lines, TC-1 and ID8, respectively. C57/Bl6 mice were injected intraperitoneally with TC−1, ID8, or PBS.

By day 14, the TC-1 mice had visible tumors and were more anemic than the control mice (TC-1 13.4 g/dL, PBS 15.5 g/dL; p=0.004). Zinc protoporphyrin (ZPP), a measure of iron-restricted erythropoiesis, was significantly elevated in the tumor mice compared to controls (TC−1 ZPP 142, control ZPP 86; p<0.001), which reflects that the anemia was due to iron restriction. The tumor mice had a higher MCV at day 14 (TC-1 46.2 fL, PBS 42.4 fL; p=0.035) which may have been due to a vigorous reticulocytosis (TC-1 RPI 7.62, PBS RPI 2.49; p=0.022). Liver hepcidin mRNA measurements were not statistically significantly different in the TC-1 and control mice at day 7. However, by day 14, the TC-1 mice had higher hepcidin levels than the control mice with a trend towards significance (TC−1 7.14 arbitrary units, PBS 3.28, p=0.052). This increase, coupled with a concurrent increase in SAA-1 (TC-1 0.02810 arbitrary units, PBS 0.00013; p=0.004), points to an inflammation-mediated mechanism in the development of anemia.

The ID8 mice demonstrated a different profile of anemia. By day 133, these tumor mice had also developed anemia (ID8 hgb drop 5.8 g/dL, PBS 2.7 g/dL; p=0.181) with increased ZPP levels (ID8 272, PBS 136; p=0.072). However, the ID8 and control mice had comparable MCV values (ID8 43.4 fL, PBS 44.6 fL). The lack of an increase in MCV in the tumor mice can be explained by a lack of the robust reticulocytosis that was seen in the TC-1 mice (ID8 RPI 14.4, PBS RPI 21.5; p=0.014) with a concurrent iron-restricted anemia. In addition, in contrast to the TC-1 tumor mice, the ID8 mice did not have elevations in hepcidin (ID8 0.202, PBS 2.026; p=0.152) or SAA-1 (ID8 0.0077, PBS 0.0116; p=0.463).

We were able to create mouse models of anemia of cancer with lung and ovarian tumor cells. By the day of sacrifice, the mice had developed evidence of tumor load (masses, ascites) and anemia. While both mouse models appeared to develop iron-restricted anemia with increased ZPP, they differed in their hepcidin and SAA-1 responses. The TC-1 mice had inappropriately elevated hepcidin despite the presence of significant anemia, indicating that hepcidin causes the iron restriction. This is likely mediated by inflammation since there is a hepatic acute phase response. ID8 mice, by contrast, do not have increased hepcidin levels or develop evidence of inflammation. These variable profiles indicate different mechanisms of the development of anemia in different cancer models.

Our mouse models demonstrate that there may be varying mechanisms for the development of anemia of cancer. Despite the development of anemia and iron restriction in both models, there are large differences in hepcidin measurements and inflammatory markers. Further studies are ongoing using hepcidin knockout mice for further elucidation of the mechanisms of anemia of cancer.

No relevant conflicts of interest to declare.