The paper in Nature Biomedical Engineering is here: https://go.nature.com/2rZGBEn

When I joined Ralph Weissleder at the Center for Systems Biology near the end of 2016, there was a growing appreciation that tumor-associated macrophages (TAMs) promote tumor growth, metastasis, and resistance to treatment.^1,2 TAMs often dominate the immune cell population within a tumor, comprising as much as 30% of the entire tumor mass. To us, this made them a very attractive target for drug delivery and immunotherapy.

TAMs often display a tumor-supportive (i.e., M2-like) phenotype that prevents the body’s immune system from attacking the tumor. We therefore reasoned that re-educating TAMs to a tumor destructive (i.e., immune-supportive, M1-like) phenotype could promote anti-tumor immune response. However, two open questions remained to be addressed:

1. Which therapeutic drugs can efficiently re-polarize TAMs from a tumor-supportive (M2-like) to a tumor-destructive (M1-like) phenotype? To examine TAM’s phenotypes after drug treatment, I initially started working with conventional qPCR to look at gene expression. But it didn’t take long to realize this would be a daunting and costly bottleneck to examine the effects of even a small library of drugs. At the same time, Mike Cuccarese (another postdoc in the center) was working on methods to determine cell phenotype by image-based methods (high-content screening by extracting features of cell shape, Figure 1). We were able to easily adapt his methods to a 384-well plate format with automated image analysis to achieve a relatively high throughput and low cost drug screen. Ultimately, we found that inhibitors of M2-like education (e.g., CSF-1 receptor inhibitors) were far less effective than direct drivers of M1 activation (e.g., toll-like receptor agonists). From these screens, we identified a highly potent drug (R848) of interest to move forward.

Figure 1. Macrophages were polarized to an M2-like phenotype. Change in cell shape was then used as a rapid screen for drug induced re-education to an M1-like phenotype. Images are adapted from our paper.

2. How can small molecule drugs be delivered to TAMs in vivo? Even while working on drug selection, we knew that achieving drug delivery specifically to macrophages would be difficult. This is because small molecule drugs distribute throughout the body and are rapidly excreted in the urine. From my prior work with Jason Burdick at the University of Pennsylvania, I had become very familiar with cyclodextrin - a common molecule well known for its ability to bind small molecule drugs.³ The Weissleder lab also has shown extensively that nanoparticles made from dextran (very similar to cyclodextrin) are rapidly engulfed by macrophages.⁴ I therefore developed a nanoparticle composed almost entirely of cyclodextrin (Figure 2), allowing specific and rapid nanoparticle accumulation in macrophages while carrying the drug of interest.

Figure 2. Cyclodextrin nanoparticles were developed which can be rapidly internalized by macrophages while carrying the drug of choice. Images are reproduced from our paper.

As these initial investigations wrapped-up, the remaining work rapidly and organically came together through a collaborative effort between two world-class labs in imaging (Ralph Weissleder) and immunology (Mikael Pittet) and their respective research teams at the Center. Through a series of studies, we found that rapid uptake of the nanoparticle by TAMs in vivo enhanced local drug delivery, promoted an anti-tumor TAM phenotype, and controlled tumor growth (Figure 3). Importantly, combination with anti-PD-1 checkpoint inhibitor further improved response and sensitized otherwise resistant models to checkpoint blockade therapy—a significant advance, especially considering that nearly 1000 clinical trials are currently underway which hope to achieve similar results in humans.⁵ In sum, these studies demonstrate the potential of TAM re-education as a viable therapeutic strategy in cancer treatment, while highlighting the need for nanotherapeutic platforms with high TAM affinity.

Looking forward, I expect that this nanoparticle formulation will be a versatile tool for immunomodulatory drug delivery. Cyclodextrins and other host molecules are known to interact with a wide variety of small molecule drugs, enabling their use as a generalizable platform for drug delivery to macrophages, neutrophils, and other phagocytic immune cells.

Figure 3. Delivery of nanoformulated R848 to the tumor via the systemic circulation (left) can profoundly alter the phenotype of tumor-associated macrophages (TAMs, top right). These cells are converted from a tumor-supportive (M2-like) to a tumor destructive (M1-like) phenotype, resulting in elimination of tumors (bottom right). Figures are adapted from our paper.

Our paper: Rodell C.B., et al. TLR7/8-agonist-loaded nanoparticles promote the polarization of tumour-associated macrophages to enhance cancer immunotherapy. Nat. Biomed. Eng. (2018) doi:10.1038/s41551-018-0236-8.

References:
1. Ruffell, B. & Coussens, L. M. Macrophages and therapeutic resistance in cancer. Cancer Cell 27, 462-472 (2015).
2. Engblom, C., Pfirschke, C. & Pittet, M. J. The role of myeloid cells in cancer therapies. Nat. Rev. Cancer 16, 447-462 (2016).
3. Rodell, C. B., Mealy, J. E. & Burdick, J. A. Supramolecular Guest-Host Interactions for the Preparation of Biomedical Materials. Bioconjug. Chem. 26, 2279-2289 (2015).
4. Keliher, E. J. et al. Polyglucose nanoparticles with renal elimination and macrophage avidity facilitate PET imaging in ischaemic heart disease. Nat. Commun. 8, 14064 (2017).
5. Tang, J., Shalabi, A. & Hubbard-Lucey, V. M. Comprehensive analysis of the clinical immuno-oncology landscape. Ann. Oncol. 29, 84-91 (2017).