Cancer cells cannot proliferate and survive unless they obtain sufficient levels of the 20 proteinogenic amino acids (AAs). Unlike normal cells, cancer cells have genetic and metabolic alterations that may limit their capacity to obtain adequate levels of the 20 AAs in challenging metabolic environments. However, since normal diets provide all AAs at relatively constant levels and ratios, these potentially lethal genetic and metabolic defects are eventually harmless to cancer cells. If we temporarily replace the normal diet of cancer patients with artificial diets in which the levels of specific AAs are manipulated, cancer cells may be unable to proliferate and survive. This article reviews in vivo studies that have evaluated the antitumor activity of diets restricted in or supplemented with the 20 proteinogenic AAs, individually and in combination. It also reviews our recent studies that show that manipulating the levels of several AAs simultaneously can lead to marked survival improvements in mice with metastatic cancers.

Targeted therapies with antiangiogenic drugs (e.g., sunitinib) and immune checkpoint inhibitors (e.g., anti-PD-1 antibodies) are the standard of care for patients with metastatic renal cell carcinoma. Although these treatments improve patient survival, they are rarely curative. We previously hypothesized that advanced cancers might be treated without drugs by using artificial diets in which the levels of specific amino acids (AAs) are manipulated. In this work, after showing that AA manipulation induces selective anticancer activity in renal cell carcinoma cells in vitro, we screened 18 artificial diets for anticancer activity in a challenging animal model of renal cell carcinoma. The model was established by injecting murine renal cell carcinoma (Renca) cells into the peritoneum of immunocompetent BALB/cAnNRj mice. Mice survival was markedly improved when their normal diet was replaced with our artificial diets. Mice fed a diet lacking six AAs (diet T2) lived longer than mice treated with sunitinib or anti-PD-1 immunotherapy; several animals lived very long or were cured. Controlling the levels of several AAs (e.g., cysteine, methionine, and leucine) and lipids was important for the anticancer activity of the diets. Additional studies are needed to further evaluate the therapeutic potential and mechanism of action of this simple and inexpensive anticancer strategy.

New therapies are needed to improve the low survival rates of patients with metastatic colon cancer. Evidence suggests that amino acid (AA) restriction can be used to target the altered metabolism of cancer cells. In this work, we evaluated the therapeutic potential of selective AA restriction in colon cancer. After observing anticancer activity in vitro, we prepared several artificial diets and evaluated their anticancer activity in two challenging animal models of metastatic colon cancer. These models were established by injecting CT26.WT murine colon cancer cells in the peritoneum (peritoneal dissemination) or in the tail vein (pulmonary metastases) of immunocompetent BALB/cAnNRj mice. Capecitabine, which is a first-line treatment for patients with metastatic colon cancer, was also evaluated in these models. Mice fed diet TC1 (a diet lacking 10 AAs) and diet TC5 (a diet with 6% casein, 5% glutamine, and 2.5% leucine) lived longer than untreated mice in both models; several mice survived the treatment. Diet TC5 was better than several cycles of capecitabine in both cancer models. Cysteine supplementation blocked the activity of diets TC1 and TC5, but cysteine restriction was not sufficient for activity. Our results indicated that artificial diets based on selective AA restriction have therapeutic potential for colon cancer.

Patients with metastatic triple negative breast cancer (TNBC) need new therapies to improve the low survival rates achieved with standard treatments. In this work, we show for the first time that the survival of mice with metastatic TNBC can be markedly increased by replacing their normal diet with artificial diets in which the levels of amino acids (AAs) and lipids are strongly manipulated. After observing selective anticancer activity in vitro, we prepared five artificial diets and evaluated their anticancer activity in a challenging model of metastatic TNBC. The model was established by injecting 4T1 murine TNBC cells into the tail vein of immunocompetent BALB/cAnNRj mice. First-line drugs doxorubicin and capecitabine were also tested in this model. AA manipulation led to modest improvements in mice survival when the levels of lipids were normal. Reducing lipid levels to 1% markedly improved the activity of several diets with different AA content. Some mice fed the artificial diets as monotherapy lived much longer than mice treated with doxorubicin and capecitabine. An artificial diet without 10 non-essential AAs, with reduced levels of essential AAs, and with 1% lipids improved the survival not only of mice with TNBC but also of mice with other types of metastatic cancers.

Since plants are an important source of anticancer drugs, we have carried out a random screening for selective anticancer activity of 57 extracts from 45 plants collected in Grazalema Natural Park, an area in the South of Spain of high plant diversity and endemism. Using lung cancer cells (A549) and lung non-malignant cells (MRC-5), we found that several extracts were more cytotoxic and selective against the cancer cells than the standard anticancer agent cisplatin. Five active extracts were further tested in cancer and normal cell lines from other tissues, including three skin cell lines with increasing degree of malignancy. An extract from the leaves of Daphne laureola L. (Thymelaeaceae) showed a striking potency and selectivity on lung cancer cells and leukemia cells; the IC₅₀ values against these cancer cells were approximately 10,000-fold lower than against the normal cells. Daphnane-type diterpene orthoesters may be responsible for this highly selective anticancer activity.