3.2. Omomyc Efficacy and Side Effects In Vivo
Given the indispensable role of Myc and Mycn during development, when we first genetically-engineered a mouse model to express Omomyc in 2004, we limited its expression to one tissue only. The chosen tissue was skin epidermis, where Myc had been shown to drive proliferation, but where concomitant Myc-induced apoptosis was innately suppressed [69
]. Both Omomyc and a Myc-ERTAM
constructs were placed under the involucrin (inv) promoter, specific to keratinocytes. inv-Myc-ERTAM
mice treated with 4-hydroxytamoxifen to activate Myc, quickly presented papillomatosis associated with hyperkeratosis and hypergranulosis, but double transgenic inv-Myc-ERTAM/Omomyc
treated in the same way did not (Figure 2
). Strikingly, skin in Omomyc-expressing mice preserved the normal epidermal keratinocyte differentiation program, even in the sustained presence of activated Myc, when it was accompanied by a high degree of apoptosis [56
This study was the first demonstration that Myc inhibition by Omomyc has an anti-tumoral effect in vivo and did not seem detrimental to normal tissue homeostasis. However, two crucial questions still remained unanswered at that time: would Myc inhibition be effective in tumors where Myc was not the driver oncogene? And even if it did, would systemic Myc inhibition cause catastrophic side effects?
To answer these questions, we generated the first conditional ubiquitous Omomyc-expressing mouse: in this mouse, the Omomyc coding sequence was placed downstream of a tetracycline-responsive promoter element (TRE); the animals also harbored a reverse tetracycline transactivator (rtTA) under the promiscuous cytomegalovirus (CMV) or β-actin promoter. The resulting TRE-Omomyc;CMV-rtTA
double transgenic mice expressed Omomyc in all tested tissues besides bone marrow upon administration of dox to their drinking water. These animals were then crossed with the well characterized LSL-KrasG12D
mouse model of lung adenocarcinoma to test the therapeutic impact of a systemic Myc inhibitor. To our surprise, activation of the Omomyc transgene in tumor-bearing mice not only caused a decrease in proliferation, but also an increase in apoptosis and senescence, detected by Ki67, TUNEL, and β-galactosidase markers respectively, finally resulting in massive tumor regression [73
Maybe even more strikingly and against preconceived notions, Omomyc expression did not cause any significant change in body weight, general animal activity, blood biochemistry or histopathological examination in low proliferative tissues (pancreas, kidney, liver, heart, and lung). Only skin, testis and intestinal crypts showed decreased proliferation. In skin, this resulted in a moderate thinning of the epidermis and inhibition of hair regrowth after shaving. In testis, Omomyc caused atrophy of spermatogonia and reduction in spermatocyte counts. The small intestine showed shortening of villi, although without increased apoptosis, and with complete maintenance of intestinal absorption and the barrier against bacterial infections. TRE-Omomyc;β-actin-rtTA
mice showed the same phenotype [73
] and were used to study the effects of Omomyc expression in bone marrow, where it caused a decrease in proliferation and onset of transient anemia and leucopenia. These symptoms were quickly compensated by extramedullary hematopoiesis in the spleen, except for very mild polycythemia. Importantly, withdrawal of dox and consequent abrogation of Omomyc expression restored normal cell proliferation of all tissues within one week and tissues became indistinguishable from untreated mice [73
]. This manuscript, published in 2008, represented a real paradigm shift in the Myc field, being the first formal proof that Myc inhibition was feasible and extremely effective as a therapeutic approach against cancer, while being safe and well tolerated in normal tissues.
In a follow-up study in 2013, we focused on whether emergence of resistance to Myc inhibition by Omomyc could occur in mice bearing KrasG12D-driven lung tumors. We demonstrated that one cycle of 4 weeks of Omomyc expression was already sufficient to significantly extend survival in these animals. Importantly, tumors that re-appeared after doxycycline withdrawal were still sensitive to Myc inhibition, since addition of dox and reactivation of Omomyc eradicated newly-formed lesions. Based on these results, we proceeded to perform metronomic treatments with Omomyc (4 weeks on, 4 weeks off), and observed that, in these conditions, mice were kept alive indefinitely, showed no signs of severe side effects and presented a lower number of relapsing tumors after each round of treatment, until their complete eradication. When mice were euthanized after more than one year, the only tumors that were present in the lungs of dox-treated mice had silenced the Omomyc transgene, which suggests that the only way to evade Myc inhibition was to turn off its expression.
To address whether additional mutations might favor emergence of resistance, the same experiment was conducted by metronomic treatment of KrasG12D
tumors in a p53 defective background [71
] with similarly dramatic results (Figure 2
). Therefore, unlike for most therapies, no compensatory mechanisms can arise in these tumors to develop resistance to Myc inhibition, even in the absence of p53.
It is important to mention that RAS is able to extend Myc protein half-life by stabilizing it through phosphorylation [75
]. Therefore, despite not being the driver oncogene, Myc plays an essential role in KrasG12D
-driven tumors. In order to validate these findings in tumors arising in a different tissue and from a different oncogenic driver, and mostly to focus on whether Myc could have a role in tumor microenvironment maintenance, we made use of one of the best characterized mouse models of pancreatic islet tumors, developed by Doug Hanahan’s group. This model, called RIP1-Tag2,
is driven by the simian virus 40 (SV40)-T antigen, and mice develop insulinomas as a consequence of T antigen expression under the control of the insulin
promoter. These animals were crossed with two different Omomyc mouse models: TRE-Omomyc;CMV-rtTA,
for systemic expression of Omomyc, and TRE-Omomyc;RIP-rtTA
, for expression in the β-cell compartment only. Administration of dox to both models, with consequent activation of the Omomyc transgene, induced complete regression of insulinomas (Figure 2
). This regression started first as collapse of the tumor microenvironment and involution of tumor vasculature. It occurred through exclusion of macrophages and neutrophils from the periphery and the interior of the tumor, respectively, abrogation of Vegf:Vegfr2 interaction, endothelial cell death, and hypoxia, and it was then shortly followed by apoptosis of β cells. Again, as previously observed in the lung model, at end point, the only tumors still present in dox-treated mice had silenced Omomyc transgene expression [70
]. Hence, Myc in tumor cells was shown to be crucial for instruction of the tumor microenvironment and was confirmed to be a non-redundant node in cancer.
More recently and in the same line of reasoning, Omomyc expression in PDAC-bearing mice (pdx1-Cre;LSL-KRasG12D;p53ERTAM;TRE-Omomyc;CMV-rtTA
) was shown to trigger quantitative regression of these highly aggressive tumors and of their extensive fibroinflammatory stroma (Figure 2
), reiterating the point that Myc holds a key role in the coordination of the cross-talk between tumors and their microenvironment [72
The idea that Myc could be a universal target in multiple tumor types was further elaborated in two more studies, in which Omomyc efficacy was assessed in different models of glioma: an Ha-Ras-driven, genetically engineered mouse model of invasive astrocytoma (GFAP-V12Ha-Ras;TRE-Omomyc;CMV-rtTA
) and in patient-derived models of glioblastoma [66
]. In the first case, we showed that, in GFAP-V12Ha-Ras
-bearing mice, expression of Omomyc upon dox administration was able to both prevent the development of astrocytomas and reverse severe neurological symptoms associated with established disease, by decreasing proliferation, increasing apoptosis and causing mitotic defects, therefore reducing astrocytic cell density and increasing mouse survival [67
] (Figure 2
). In mice transplanted with patient-derived glioblastoma neurospheres harboring a dox-inducible Omomyc expression cassette, Omomyc expression recapitulated the same mitotic defects and greatly reduced intracranial cell density, conferring a survival advantage to the mice [66
]. In addition, Galiardi et al. showed that Omomyc expression reduced the GSC marker OLIG2 in glioblastoma cells, as well as the number of glioblastoma migrating cells and vascularization of the tumor stroma. As previously reported for other models, a fraction of the cells in this model also silenced Omomyc expression and retained their tumorigenic features [66
The study by Von Eyss et al. in 2015 added yet another model to the list of Omomyc sensitive tumors: the MMTV-Wnt1
mouse model of breast cancer, where the authors demonstrated that Omomyc expression could induce a strong decrease in proliferation in breast cancer cells [76
]. Additionally, in 2019, Alimova et al. showed that expression of Omomyc significantly reduces tumor growth and extends survival in an atypical teratoid rhabdoid mouse model [68
], demonstrating for the first time its efficacy against pediatric cancer.
These experiments as a whole confirmed the non-autonomous effects of Omomyc on the tumor microenvironment, while also reinforcing the notion of its safety in long-term systemic treatments, and expanded the list of tumor types as candidate for Myc inhibition treatment, regardless of their tissue of origin or driving mutation.