Dietary 4-HPR suppresses the development of bone metastasis in vivo in a mouse model of prostate cancer progression

The effects of the synthetic retinoid N-(4-hydroxyphenyl) retinamide (4-HPR) on prostate cancer metastasis in vivo were evaluated in the mouse prostate reconstitution (MPR) model. MPRs were produced by infection of either heterozygous (+/-) or nullizygous (-/-) p53-mutant fetal prostatic epithelial cells with the recombinant retrovirus Zipras/myc 9. Previous studies have documented that loss of p53 function potentiates metastasis in this model system. MPRs were grafted into homozygous (+/+) p53 male mice, fed a 4-HPR containing diet or a control diet and maintained until the status of tumor progression dictated sacrifice. Under these experimental conditions, treatment with 4-HPR did not have a significant effect on primary tumor wet weight for either p53 +/- or p53 -/- MPRs. For, p53 +/- MPRs the animals fed the 4-HPR diet had a slight improvement in survival and a significant reduction in the number of mesenteric metastases (P=0.0477, t-test). Notably, in p53 +/- MPRs the incidence of metastasis to lumbar spine and sternum was 92% in the control animals compared to 54% in the 4-HPR treated animals (P = 0.035, chi (2)-test). In p53 -/- MPRs there was a trend toward a reduction in the number of soft tissue metastases to lung and liver in the 4-HPR group relative to the control diet group and a statistically significant reduction in the incidence of metastasis to bone was demonstrated in that 50% of control animals versus 30% of 4-HPR treated p53 -/- animals harbored bone metastases (P = 0 < 0.05, chi (2)-test). Cell lines were established from portions of the primary tumor and from selected metastatic deposits in each experimental group. Clonal analysis, by retroviral integration pattern, indicated increased clonal diversity in both the primary tumors and metastasis-derived cell lines from 4-HPR treated animals relative to the control animals. In vitro treatment with 4-HPR did not reveal discriminating differences between cell lines derived from primary tumors and bone metastases or control and treatment groups in regard to growth arrest or apoptotic responses. Overall these studies indicate limited anti-tumor and anti-metastatic activity in this highly aggressive in vivo mouse model of prostate cancer, yet 4-HPR treatment significantly suppressed the development of bone metastases in p53 +/- and p53 -/- MPRs revealing a novel and potentially clinically useful activity of this retinoid.