Background Tumor associated macrophages (TAMs) promote an immunosuppressive and protumor immune microenvironment reducing the efficacies of cancer immunotherapies. Altering TAM phenotypes from immunosuppressive to proinflammatory should stimulate TAMs directly and lymphocytes indirectly to attack cancer cells and potentially increase the efficacies of check point inhibitor therapies, like anti-CTLA4. Mannosylated Amine Dextrans (MADs) are synthetic ligands for CD206 that can carry small molecule payloads on degradable linkers to CD206 expressing cells, i.e. TAMs. Three MAD constructs were made carrying payloads of paclitaxel or one of two novel bisphosphonates and were evaluated in human GM-CSF macrophages for their abilities to durably alter macrophage phenotypes towards pro-inflammatory. The paclitaxel MAD was tested in the CT26 tumor model with or without anti-CTLA4.

Methods >Human monocytes from 3 donors were differentiated to CD206+ macrophages in GM-CSF supplemented RPMI1640 with 10% FBS (fresh medium) for 5 days.

>Medium was replaced with fresh media with varying concentrations of MAD + drug constructs or molar equivalents of free drugs or zoledronate and incubated for 24 hours.

>Saline or MAD without payload were controls.

>Drug containing media were replaced with fresh medium and incubated for 3 days (for durable responses)

>Cells were assessed by flow cytometry (MFI) for viability, CD206, CD163, CD80, CD86, MHC1, MHC2, SIRPalpha, and PD1.

>CT26 tumors implanted in Balb/c mice, treatment began when tumors were 100mm2.

>Mice treated with MAN-Paclitaxel (127 micrograms/dose), free paclitaxel, saline, or MAD no payload twice/week for 5 doses. Tumor volumes measured with calipers were the output.

Results Cell/FlowAssays

>None of the constructs reduced macrophage viability significantly.

>All three MAD-Drug constructs reduced expression of CD206 and CD163 (immunosuppression markers) and increased expression of CD80 and CD86 (pro-inflammatory markers).

>PD1 expression increased.

>MHC1 and MHC2 expression was not significantly changed.

>Both MAD-bisphosphonates – but not free drugs or MAD-paclitaxel – significantly reduced expression of SIRPalpha (p<.0001).

CT26 Studies

>MAD-Paclitaxel and free paclitaxel were equally effective at tumor growth reduction.

>MAD-Paclitaxel plus anti-CTLA4 reduced tumor growth by 76%. Anti-CTLA4 alone by 52%.

Conclusions > All 3 MAD-drug constructs shifted macrophage phenotypes towards pro-inflammatory and more effectively the free drugs.

> Both MAD-Bisphosphonated constructs significantly reduced expression of SIRPalpha (checkpoint)

>MAD-paclitaxel and free paclitaxel increased the efficacy of anti-CTLA4, but

>CD206 targeted delivery can avoid off target toxicities of paclitaxel.

Shifting TAM phenotypes towards a pro-inflammatory state in a targeted fashion has the potential to increase the efficacies of other immunotherapies to reduce tumor burdens.

Acknowledgements Macrophage cell culture assays were performed at Discovery Life Sciences (Huntsville, AL). CT26/Balb/c studies were performed by Charles River Laboratories (North Carolina).

Ethics Approval All studies performed at Charles River Laboratories were reviewed and approved by an institutional IACUC committee.