Dissemination of tumor cells to regional lymph nodes via the lymphatic system represents the first step in HNSCC metastasis and is the most important poor prognostic factor for disease recurrence. Tumor-associated lymphangiogenesis plays an active role in metastatic disease spread by providing escape routes for cancer cells and is supported by significant correlation between intratumoral lymphatic vessel density and lymph node metastasis [25, 26]. HNSCC are highly vascular tumors with remarkable expansion of both blood and lymphatic vascular networks in head and neck area. In our previous study we showed an equally high density of blood and lymphatic vessels in HNSCC patients, underscoring the fact that HNSCC is not only highly angiogenic, but also highly lymphangiogenic . Accumulating evidence now supports rapalogues potent activity against tumor blood vasculature and we have shown that mTOR inhibitors have potent anti-angiogenic effects in HNSCC. Temsirolimus (CCI-779) significantly suppressed angiogenesis in HNSCC xenografts, decreasing intra-tumoral microvessel density by 42% (P < 0.001) . Similarly in our current study we found a significant 36% inhibition of blood microvessel density by rapamycin in the HNSCC orthotopic tumor model as well. Several studies show rapamycin also exerts anti-lymphangiogenic effects in vitro, blocks in vivo lymphangiogenesis in pancreatic cancer , and reduces regenerative lymphangiogenesis in a skin flap model . Together these findings underscore the importance of mTOR-targeted therapy in inhibiting both tumor angio- and lymphangiogenesis. Unlike blood vessel angiogenesis, rapalogues effects on tumor-associated lymphangiogenesis are not well understood, but could provide critical additional target for mTOR inhibitors in the treatment of HNSCC. Recently, in the study by Gutkind et al we demonstrated anti-lymphatic properties of rapalogues in an orthotopic model of HNSCC generated by injection of UMSCC2 cells into the tongue of SCID/NOD mice . In this study we obtained further evidence for the anti-lymphatic properties of mTOR inhibitors employing OSC-19 orthotopic model of HNSCC and investigated the mechanisms of rapalogues anti-lymphatic effects using in vitro and in vivo models.
Treatment of SCID mice with 5 mg/kg of rapamycin for 16 days significantly lowered lymphatic microvessel density and significantly reduced lymphovascular invasion and decreased the incidence of cervical lymph node metastasis compared to vehicle-treated controls. Furthermore, rapamycin significantly suppressed the extent of metastatic tumor cell spread within the lymph nodes. Most tumor-positive lymph nodes in the control group (78%) demonstrated complete replacement of the normal lymph node architecture with tumor cells. Conversely, the majority (74%) of positive cervical lymph nodes extracted from rapamycin-treated mice demonstrated only minimal tumor cell spread, with only few metastatic tumor cells localized to subcapsular sinuses, an early stage of cervical lymphatic metastasis known as ‘micrometastasis’. This suggests that rapamycin can delay lymphatogenous metastatic spread in head and neck cancer, potentially impeding extracapsular extension of squamous cell carcinoma nodal metastases, a significant poor prognostic factor for decreased patient survival .
The results obtained in the animal experiment employing an orthotopic murine model of HNSCC were further supported by in vitro study findings. The LEC proliferation assay showed that mouse and human lymphatic endothelial cells are highly sensitive to mTOR inhibitors, which decreases LEC proliferation by >35% in 72h of treatment. Interestingly we observed a moderate, but significant increase in apoptotic cell death after rapamycin treatment for a faster proliferating SV-LEC cell line, but not for HMEC-1A cell line, which showed only a minimal increase in the number of apoptotic cells. Potent anti-lymphatic effects of the rapalogues have now been associated with inhibition of mTOR signaling.
Not only angiogenesis, but lymphangiogenesis too plays an important role in promoting tumor growth and metastasis. The lymphatic system is a main conduit for initial metastasis for many types of solid tumors, including head and neck cancer. VEGF-C and VEGFR-3 are not only expressed by lymphatic EC, but also by a variety of HNSCC cell lines, including the HNSCC cell lines used in this study (SCC40, FaDu, PCI-15a, OSC-19) (Figure 5A). The VEGF-C/VEGFR-3 axis plays an important role in cancer progression through several cellular pathways . Activation of the VEGF-C/VEGFR-3 axis in lymphatic ECs promotes lymph node metastasis, while binding of VEGF-C to VEGFR-3 creates a positive-feedback ‘autocrine loop’ which further enhances VEGF-C release, to dramatically stimulate cancer cell proliferation as well as lymphangiogenesis . In our study we found that rapamycin strongly suppressed VEGFR-3 expression in both human and mouse lymphatic EC (Figure 5B). Rapalogues also significantly inhibited VEGFR-3 expression in several HNSCC cell lines. Because rapalogues down-regulate VEGFR-3 expression in lymphatic endothelial cells and some HNSCC cells it suggests mTOR inhibitors can suppress this vicious cycle of autocrine growth stimulation to decrease the number of lymph node metastasis, one of the most important factors contributing to poor head and neck cancer prognosis and survival. Mechanistically, another study coauthored by one of the authors of this paper showed that rapamycin affects VEGFR-3 protein expression in LEC cells by inhibiting protein synthesis and promoting protein degradation of VEGFR-3. Importantly rapamycin did not alter the VEGFR-3 mRNA level .
Another important observation from this study was that rapamycin significantly increased the level of soluble VEGFR-2 in serum samples in SCID mice implanted with HNSCC. We also observed a rapamycin-induced upregulation in the level of soluble VEGFR-2 in serum samples of nude mice with FaDu HNSCC xenograft tumors (Ekshyyan O., Moore-Medlin T., Nathan CO; unpublished observation). Recently, a soluble form of VEGFR-2 (sVEGFR-2) that is produced by alternative splicing has been identified as an endogenous selective inhibitor of lymphatic vessel growth [32, 33].
In a recent study by Silver et al  sVEGFR-2 expression was found to be inversely correlated with lymphatic vessel density in head and neck malignant tumors. Interestingly sVEGFR-2 was not expressed in lymphatic vessels, but its expression was specific to the endothelial cells in blood vessels in both malignant tissue as well as adjacent normal tissues . Furthermore it was demonstrated that gene therapy with a splicing variant esVEGFR-2 that produces soluble VEGFR-2 significantly suppresses tumor growth and lymph node metastasis in a mouse mammary cancer model .
Because soluble VEGFR-2 binds VEGF-C it may competitively inhibit VEGF-C-induced activation of pro-lymphangiogenic and angiogenic signaling. sVEGFR-2 release could be used as a potential biomarker of anti-lymphangiogenic and angiogenic responsiveness in clinical trials of mTOR inhibitors and warrants further investigation.