Tumor Associated Macrophages Can Contribute to Antitumor Activity Through Fcγr-Mediated Processing of Antibody-Drug Conjugates

Home , Monomethyl auristatin E

Author Manuscript Published OnlineFirst on March 24, 2017; DOI: 10.1158/1535-7163.MCT-17-0019 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Tumor associated macrophages can contribute to antitumor activity through FcγR-

mediated processing of antibody-drug conjugates

Fu Li1, Michelle Ulrich1, Mechthild Jonas2, Ivan J. Stone 1, Germein Linares2, Xinqun Zhang3,

Lori Westendorf2, Dennis R. Benjamin2, Che-Leung Law1

1. Preclinical Research, Seattle Genetics, Inc. 21823 30th DR SE, Bothell WA 98021

2. Translational Research, Seattle Genetics, Inc. 21823 30th DR SE, Bothell WA 98021

3. Chemistry, Seattle Genetics, Inc. 21823 30th DR SE, Bothell WA 98021

Correspondence

Fu Li, PhD

Seattle Genetics, Inc.

21823 30th Drive SE. Bothell WA 98021

Tel: 425-527-4626

Email: fli@seagen.com

Running title: Role of TAMs in ADC antitumor activity

Downloaded from mct.aacrjournals.org on September 24, 2021. © 2017 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 24, 2017; DOI: 10.1158/1535-7163.MCT-17-0019 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Abstract

The primary mechanism of antibody-drug conjugates (ADCs) is targeted delivery of a cytotoxic payload to tumor cells via cancer-associated membrane receptors. However, the tumor microenvironment likely plays a role in ADC penetration, distribution, and processing and thus impacts the overall antitumor activity. Here, we report on the potential contribution of Fc-FcγR interactions between ADCs and tumor-associated macrophages (TAMs) to the preclinical antitumor activities of ADCs.

In the CD30+ L-428 Hodgkin lymphoma model, anti-CD30-vcMMAE and a non-binding control

(hIgG-vcMMAE) demonstrated similar antitumor activity as well as similar payload release in the tumors. Immunohistochemistry analysis revealed L-428 tumors contained highly abundant

TAMs, which were confirmed to bind ADCs by IHC and flow cytometry. The infiltration of

TAMs was further found to correlate with the antitumor activity of the non-binding hIgG- vcMMAE in five additional xenograft models. hIgG1V1-vcMMAE, bearing a mutation in the

Fc region which ablates Fc gamma receptor (FcγR) binding, lost antitumor activity in three

TAM-high xenograft models, suggesting Fc-FcγR interactions modulate the TAM-ADC interaction.

Our results suggest that TAMs can contribute to ADC processing through FcγR interaction in preclinical tumor models and may represent an important additional mechanism for drug release from ADCs. Correlative studies in clinical trials will further shed light on whether TAMs play a role in patients’ response to ADC therapies.

Introduction

Antibody-drug conjugates (ADCs) represent an important class of therapeutic modalities for cancer (1). By conjugating a potent cytotoxic payload to targeted antibody, ADCs such as

ADCETRIS® (brentuximab vedotin) and KADCYLA® (ado-trastuzumab emtansine) have offered novel treatment options of Hodgkin lymphoma and HER2+ breast cancer, respectively

(2, 3). The key driver for the clinical activity of ADCs is attributed to the targeted delivery of a payload, which enters the cell through endocytic and lysosomal trafficking pathways. Once processed by lysosomal proteases (e.g. cathepsins), the released payload either induces apoptosis through disruption of microtubules (in the case of auristatins and maytansines), or nuclear DNA

(in the case of pyrrolobenzodiazepine dimers and calicheamicin). In addition, after antigen- mediated drug release has occurred, some membrane permeable payloads (such as DM1,

MMAE, and PBDs) have been shown to mediate bystander killing of adjacent tumor cells regardless of whether the targeted antigen is expressed (4, 5) . Although these mechanisms of action have been demonstrated in vitro and in vivo, antigen-mediated drug release does not account for all the potent antitumor activities observed in the clinic (6). In fact, some diffuse large B-cell lymphoma (DLBC) patients with minimal or undetectable CD30 expression on their lymphoma cells responded to brentuximab vedotin in a phase 2 clinical trial (7, 8). Similarly, it has been reported that CD22 and CD79b expression is not required for patients to respond to

ADCs targeting these receptors (6). These clinical observations raise the possibility that factors beyond target antigen expression on the cancer cells may be contributing to ADC antitumor activity (9, 10).

Engagement between cancer cells and the tumor microenvironment has been proposed as one of the hallmarks for cancer (11, 12). In particular, the infiltrating immune cells have been shown to

be co-opted to promote tumor development and progression (13, 14). Numerous studies have shown that infiltration of tumor-associated macrophages (TAMs) often correlates with poor prognosis in multiple types of malignancy. Not surprisingly, infiltrating immune cells can also influence the response to anticancer therapies (15, 16). Indeed, TAMs-a prominent subpopulation of the infiltrating immune cells-have been linked to the response to conventional chemotherapies and targeted anticancer agents (16). On one hand, infiltrating TAMs may reduce sensitivity to paclitaxel and gemcitabine through suppression of CD8+ T cell activity (17, 18). On the other hand, TAMs can be crucial for executing the effector functions of certain antibodies. For example, trastuzumab has been shown to trigger macrophage engulfment of human epidermal growth factor receptor 2 (HER2)-positive cells (19, 20). These observations suggest the potential contribution of TAMs to the antitumor activity of antibody-based therapies.

Assessing ADC antitumor activity often relies on comparing tumor growth in xenograft models.

While most such studies support the antigen-specific activities of targeted ADCs (21, 22), we occasionally have observed a significant growth delay in tumors treated with non-binding controls, including a humanized IgG conjugated to the protease-cleavable monomethyl auristatin

E drug linker (hIgG-vcMMAE) (23). For example, while treatment with αLIV1-vcMMAE ADC led MCF-7 tumors to complete tumor regression for 50 days, these tumors also responded to hIgG-vcMMAE treatment, undergoing an evident growth delay (Figure S1B) (24).

Understanding the unexpected activity of hIgG-vcMMAE in vivo may help to better characterize the mechanisms of ADC drugs in clinic.

Here we evaluate the roles of tumor microenvironment on ADC antitumor activity in xenograft models, by comparing the antitumor activity and intratumoral drug concentrations resulting from targeted and non-targeted (hIgG-vcMMAE) monomethyl auristatin E (MMAE) conjugates. Our

results suggest non-targeted ADCs can bind to F4/80+ TAMs, the abundance of which correlated with the antitumor activity of non-targeted ADCs in lymphoma and breast cancer models in vivo.

Moreover, Fc gamma receptor (FcγR) binding-impaired hIgG1V1-vcMMAE conjugate was inactive in three xenograft models with high TAM infiltration, demonstrating that TAM mediated ADC activity are likely due to Fc-FcγR interactions.

Materials and Methods

Xenograft studies and treatment

All animal studies were performed according to Institutional Animal Care and Use Committee guidelines. L-428(ACC197), KM-H2(ACC8), DOHH-2(ACC47), Karpas 299 (ACC31) and L-

82(ACC597) cells were purchased from the German Collection of Microorganisms and Cell

Cultures (DSMZ, Germany) between 2012 and 2013. Of note, Karpas 299 cells were originally provided by Dr. Karpas from the University of Cambridge (UK) in 1998. SU-DHL-8(CRL2961) was purchased from American Type Culture Collection (ATCC) in 2013. These cell lines were all authenticated by IDEXX Laboratories (MO). Five million L-428 and KM-H2 cells were subcutaneously injected to NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ mice (NSG™, Jackson Laboratory,

CA) to establish xenografts. DOHH-2, SU-DHL-8, and Karpas 299 cells were implanted in severe combined immune deficient mice (SCID) (Harlan, IA). BR620 was a triple negative breast cancer patient-derived xenograft model passaged in NSG mice (Jackson Laboratory, CA).

Tumor bearing animals were treated with ADCs intraperitoneally (ip) when the tumor volume is between 100~200 mm3. Tumors were measured with a digital caliper twice a week and the tumor volume is calculated using the formula ½ X length X width2. Intratumoral MMAE from tumors was prepared and measured using mass spectrometry as previously described (25).

Immunohistochemistry and flow cytometry

Immunohistochemistry(IHC) on formalin fixed paraffin embedded xenograft tissues were performed using a Bond autostainer (Leica, Germany). Slides were deparaffinzied and antigen was retrieved in heated EDTA buffer. Macrophages were stained with rabbit anti-mouse F4/80

(Abcam, MA), while ADCs were stained with anti-MMAE antibody (SG15.22, Seattle

Genetics). Horseradish peroxidase-3,3'-diaminobenzidine (HRP-DAB) and alkaline phosphatase- fast red was used to detect bound primary antibodies. Hematoxylin was used to counter stain nuclei. Images were acquired on an Olympus BX41 microscope equipped with a Nikon DS-Fi1 camera.

For flow cytometric analysis, tumor samples were dissociated non-enzymatically and washed prior to staining with the following antibodies: PE conjugated rat anti mouse F4/80 (Biolegend,

CA), SG3.236 conjugated to Alexa Fluor 647 (Seattle Genetics). Flow cytometry was performed on FACS Calibur (BD Bioscience, CA) and data were analyzed using FlowJo (Tree Star, Inc.

OR)

Antibody-drug conjugates

Antibodies used for conjugation were all described previously (26). Anti-CD30 antibody was cAC10; Ab1-vcMMAE recognizes a surface receptor that is ubiquitously expressed and rapidly internalized (Seattle Genetics). Monomethyl auristatin E (MMAE) or monometheyl auristatin F

(MMAF) was conjugated to antibodies with an average drug: antibody ratio of 4.

Statistical analysis

Tumor volume was plotted in Prism software (GraphPad, CA). In general, two-way ANOVA analysis was used to compare the difference between each treatment agents, unless specified otherwise in the figures.

Results

Antigen-independent antitumor activity of ADCs in L-82 and MCF-7 xenografts

Consistent with most reports on xenograft response to targeted ADCs, tumors established from

anaplastic large cell lymphoma L-82 expressed CD30 and went to complete regression after a single dose of αCD30-vcMMAE for 50 days. hIgG-vcMMAE treatment resulted in an initial complete remission for 10 days, followed by regrowth of tumor (Figure S1A). The moderate anti-tumor activity of hIgG-vcMMAE was also previously reported in MCF-7 xenograft model

(Figure S1B) (24) . These observations illustrate while the targeting ADCs are certainly more

active, the non-binding hIgG-vcMMAE can also have pronounced antitumor activities in vivo.

Intratumoral drug release by hIgG-vcMMAE

We have previously shown that intratumoral MMAE release correlated with ADC antitumor

activity in vivo (5). To understand the antigen-independent activities of non-binding ADCs, we first determined whether tumors can process hIgG-vcMMAE and release MMAE. A panel of nine xenograft tumor models, including three Hodgkin lymphoma cell lines (L-428, KM-H2, and

L540cy), two anaplastic large cell lymphoma cell lines (Karpas-299, and L-82), two acute myeloid leukemia cell lines (TF1α and THP-1), one Burkitt’s lymphoma (Ramos), and a renal cell carcinoma cell line (786-O), were treated with hIgG-vcMMAE at 3 mg/kg (once ip.) and collected to measure intratumoral MMAE concentration 72 hours post dose. Liquid chromatography–mass spectrometry analysis revealed that MMAE was generated in these

tumors, although at varying concentrations (range: 23 to 193 nM) (Figure S1C). This result suggests that tumors can process non-targeted hIgG-vcMMAE in vivo, although some may do so better than others. Depending on the intrinsic sensitivity of tumor cells, MMAE released from

non-targeted hIgG-vcMMAE may accumulate in the tumor microenvironment to concentrations

sufficient to mediate tumor cell death.

To better characterize the kinetics of hIgG-vcMMAE induced activity, we measured the tumor

growth after the treatment of a targeting ADC (αCD30-vcMMAE) or hIgG-vcMMAE in the L-

428 tumors, in parallel to intratumoral MMAE concentration over a 10-day period. A single

injection of 2 mg/kg αCD30 -vcMMAE or hIgG-vcMMAE induced stable disease in these

tumors (Figure 1A). Released MMAE concentration increased during the first three days and slowly decreased in the tumor. Importantly, hIgG-vcMMAE and αCD30-vcMMAE yielded comparable MMAE exposure in the tumor (669 vs. 907 day.nM; Figure 1B). These data suggest that, despite its non-binding nature, hIgG-vcMMAE was processed within the tumor microenvironment to release its payload without engaging any antigen on the tumor cells.

We next evaluated how the hIgG ADCs were processed in the L-428 tumors. First, we compared the activity of hIgG-vcMMAE and hIgG-vcMMAF, two ADCs which only differ in the released payload. Released MMAE is more membrane permeable than MMAF and can mediate bystander killing in vivo (5, 27). Anti-CD30 ADCs utilizing either MMAE or MMAF led to regression of

CD30+ L-428 tumors, suggesting both payloads are potent in vivo (Figure 2A). Interestingly,

although both hIgG-vcMMAE and hIgG-vcMMAF released payloads in the tumors, only hIgG-

vcMMAE caused delay of tumor growth (Figure 2A and B). These data are consistent with our recent observation that MMAE is significantly more membrane permeable relative to MMAF enabling it to mediate bystander tumor cell killing with the tumor microenvironment.

TAMs can process ADCs in L-428 tumors

We next performed immunohistochemistry (IHC) analysis for several components in the tumor microenvironment including collagen, endothelial cells, and macrophages to identify the

mechanism underlying processing of non-targeted hIgG-vcMMAE. A distinguishing feature of

L-428 tumors was the significant amount of macrophage infiltrates (Figure 3A), as illustrated by the macrophage marker F4/80 staining. Co-staining of F4/80 and anti-auristatin antibody revealed that F4/80 positive macrophages co-localized with hIgG-vcMMAE ADCs in the L-428 tumors, whereas most of the F4/80 negative cells, including L-428 tumor cells, were not stained by the anti-auristatin antibody (Figure 3A). The physical interaction between hIgG-vcMMAE with TAMs was further confirmed by flow cytometric analysis showing that F4/80 expressing murine host macrophages retrieved from hIgG-vcMMAE-treated L-428 xenografts could be recognized by an anti-auristatin antibody (Figure 3B). Therefore, hIgG-vcMMAE can bind to

TAMs in vivo.

TAMs correlate with hIgG-vcMMAE antitumor activity in xenograft models

We then stained a panel of xenograft models to evaluate whether abundance of TAMs correlated with hIgG-vcMMAE mediated antitumor activity. Using F4/80 IHC, we found L-428, KM-H2, and BR620 (a patient-derived triple negative breast cancer model) contain abundant TAMs

(Figure 4A). In line with our hypothesis, hIgG-vcMMAE treatment resulted tumor remission or significant growth delay in these tumor models (Figure 4B). For example, 3 mg/kg hIgG- vcMMAE caused remission and growth delay in L-428 tumors. Same treatment with hIgG- vcMMAE led to growth delay in the KM-H2 xenografts, although the growth delay is shorter than a CD30-targeting ADC. In the BR620 model, hIgG-vcMMAE also caused a significant growth delay.

In contrast, hIgG-vcMMAE was not active in xenograft models lacking TAMs. For example,

TAMs were not found in tumors from DOHH-2 (B cell non-Hodgkin lymphoma), SU-DHL-8

(large cell B lymphoma), and Karpas 299 (anaplastic large cell lymphoma) (Figure 4C).

Concordantly, treatment of hIgG-vcMMAE had no impact the growth of DOHH-2 and SU-DHL-

8 xenografts, and a minor delay of growth in Karpas-299 xenografts (Figure 4D). Importantly, a positive control ADC (Ab1-vcMMAE) induced tumor growth delay in DOHH-2 and remission in SU-DHL-8 tumors, suggesting that (i) these models were sensitive to the payload MMAE and

(ii) target-mediated ADC internalization and payload release mechanisms were intact in these models. Similarly, αCD30-vcMMAE induced complete remission of Karpas 299 tumors.

Collectively, these data suggests the abundance of TAMs in preclinical tumor models may contribute to the antitumor activity of non-binding ADCs.

FcγR mediates TAM-ADC interaction

We then assessed whether reducing TAMs can attenuate non-binding ADC activity. Clodronate- encapsulated liposome (CEL) can deplete TAMs in vivo (28). Injecting L-428 tumor bearing mice with CEL prior to hIgG-vcMMAE dosing indeed attenuated the effect of antigen- independent activity (Figure S2). CEL injection alone did not impact tumor growth in this model, supporting that TAMs are contributing to the antigen-independent antitumor activity in this model.

To further understand the mechanisms by which TAMs interact with ADCs, we generated a

G1V1 mutant of the hIgG1 (E233P:L234V:L235A), which has profoundly reduced binding to

CD64 (FcγR1) for both antibody and ADC (Figure S3) (29). We then compared the activity of hIgG-vcMMAE and hIgG1V1-vcMMAE in L-428 xenograft model. As shown previously, a single dose of 3 mg/kg hIgG-vcMMAE resulted in a 30-day regression of L-428 tumors. In contrast, hIgG1V1-vcMMAE did not mediate any apparent tumor regression and gave limited growth delay of L-428 tumors (Figure 5A). In the Hodgkin lymphoma model KM-H2, while 3 mg/kg single dose of hIgG-vcMMAE yielded remission for a week and followed by growth

delay for 10 more days, hIgG1V1-vcMMAE had no obvious impact on the growth of the tumors

(Figure 5B). In the patient-derived xenograft model BR620, 4 doses of 3 mg/kg hIgG-vcMMAE led to tumor regression and growth delay for 4 weeks. On the other hand, these tumors did not respond to hIgG1V1-vcMMAE at the same dose level (Figure 5C). These three studies suggest

that abrogating FcγR interaction reduces TAM-mediated ADC activity.

An alternative method to reduce the Fc-FcγR interaction is to saturate the plasma compartment

with excess IgG. We treated KM-H2 tumor bearing mice with 10mg/kg human intravenous

immunoglobulin (hIVIg) one day prior to the ADC administration. Mice receiving hIVIg had a

shorter growth delay response to the hIgG-vcMMAE treatment (Figure 5D). This result suggests

the interaction between host FcγR and ADCs can be attenuated by introducing circulating IgG.

Collectively, these data indicate that Fc binding function of the ADC was required for TAM-

mediated antitumor activity.

Discussion

With the development of ADCETRIS® (brentuximab vedotin) and KADCYLA® (ado-

trastuzumab emantasine), ADCs have become a major class of novel drugs developed for the

treatment of cancer. More ADC drugs have demonstrated responses in indications that were

challenging to treat, such as acute myeloid leukemia and glioblastoma (30, 31). However, the

mechanisms by which ADCs exert antitumor effects independent of ADC-target interaction still

remain unclear. For example, while CD30 is detected in ~25% of diffuse large B-cell lymphoma

(DLBCL) samples (32-34), brentuximab vedotin has surprisingly achieved a 44% objective

response rate in patients with relapsed /refractory DLBCL with variable CD30 expression in a

recent phase 2 trial (7). On the other hand, ADCs targeting pan-B cell markers (e.g. CD19,

CD22, and CD79b) resulted in response rate between 31% and 54% (35, 36). Collectively, these

clinical studies suggest other factors are involved in ADC activities in addition to antigen

expression.

Antigen-independent antitumor activity in xenograft models

While mounting evidence has shown that ADCs mediate antigen-specific cytotoxicity in vitro,

the antitumor activity of ADCs in vivo is only partially explained by antigen-specific uptake and

processing. For example, in tumors displaying heterogeneous antigen expression, bystander

killing of antigen-negative tumor cells can be mediated by membrane permeable payloads (4, 5).

However, we and others reported that non-binding ADCs may lead to growth delay or remission

in xenograft models (9, 24). In this study, we found hIgG-vcMMAE, a non-binding humanized

IgG control, lead to tumor regression of two lymphoma and one breast cancer tumor models.

Importantly, at the same dose level, hIgG-vcMMAE did not impact the growth of other

lymphoma tumor models or many other carcinoma models (25, 37). The highly variable activity

of hIgG-vcMMAE activity is consistent with the findings reported by others (10), indicating that

ADCs may inhibit tumor growth in the absence of specifically targeted antigens, which may be

associated with a clinical benefit for cancer patients. Interestingly, several studies have reported

that a few patients with limited detectable target expression responded to ADC treatment (6, 7).

While antigen expression can be heterogeneous with a given tumor and thus non-detectable on single biopsy of that tumor, more studies are required to better document such phenomena in future clinical trials and to elucidate the mechanisms underlying such observations.

TAMs and Fc interaction in therapeutic responses to ADCs

It is now widely accepted that the TAMs can influence disease progression and metastasis (11,

12). TAMs have been found to correlate with poor prognosis for cancers in breast, cervix, and

bladder (13). In breast cancer, increased CD68+ macrophage counts correlated with shorter

relapse-free survival and overall survival (38). In classical Hodgkin lymphoma, malignant Reed-

Sternberg cells are outnumbered by non-malignant stromal cells. Gene expression profiling and

IHC analysis have suggested higher TAMs correlate with reduced rate of disease-specific survival in adult Hodgkin lymphoma (39, 40). The roles of TAMs in another type of lymphoma,

DLBCL, remains controversial (41). While TAMs have been linked to poor prognosis in patients received chemotherapies such as CHOP, DLBCL patients with higher TAMs responded favorably when rituximab was given in combination with chemotherapy (42-45). All of these clinical studies demonstrate a critical role of TAMs in cancer.

In contrast to the growing consensus on the biology of TAMs, their involvement in the response to cancer therapeutics is more complex (46). Recently, TAMs were found to dampen the immune response and may drive resistance to chemotherapies such as gemcitabine, paclitaxel, and doxorubicin in animal models (17, 18). Monoclonal antibody based therapies, such as rituximab and trastuzumab, may require the engagement of macrophages to assist phagocytosis. SGN-30, the parental antibody of brentuximab vedotin, was found to mediate ADCP in animal models

(47). While these studies in general support the hypothesis that targeting antibodies can interact with macrophages, our study here is the first to demonstrate TAMs can interact with therapeutic

ADCs even in the absence of antigen binding. The interaction between TAMs and ADCs was mediated through FcγRs, and such interaction led to ADC internalization and processing by the

TAMs, with subsequent payload release within the tumor microenvironment. When the payload is membrane permeable, e.g. MMAE, the TAM-processed ADC can effectively mediate

“bystander killing” of tumor cells. In addition, it is likely that other mechanisms contribute to processing of ADCs in the tumor microenvironment (10). Tumor-specific activation of

Probody™ antibodies have been recently developed on the basis of extracellular

metalloproteases (48). While it is beyond the scope of this study, further correlative studies in

clinical trials on ADCs will be needed to more critically evaluate the role of TAMs and

extracellular proteases in ADC processing, payload release, and antitumor effects in human

patients.

TAMs may either promote or suppress tumor progression depending on how they are polarized,

thus representing a potentially targetable cell population for cancer therapy (14). Indeed, an anti-

CSF-1R antibody RG7155 led to depletion of TAMs and increases of CD8/CD4 T-cell ratio in various cancer models, and resulted in objective responses in patients with diffuse-type giant cell tumors (49). Our study clearly demonstrates the ability of TAMs to internalize ADCs using their

FcγRs and subsequently process ADCs to release their payloads. Continued efforts to develop

ADC drugs for targeted depletion of TAMs may provide another therapeutic avenue for treating

cancer.

References

1. de Goeij BE, Lambert JM. New developments for antibody-drug conjugate-based therapeutic approaches. Current opinion in immunology. 2016;40:14-23. 2. Kumar A, Younes A. Role of CD30 targeting in malignant lymphoma. Current treatment options in oncology. 2014;15:210-25. 3. Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, et al. Trastuzumab emtansine for HER2-positive advanced breast cancer. The New England journal of medicine. 2012;367:1783-91. 4. Kovtun YV, Audette CA, Ye Y, Xie H, Ruberti MF, Phinney SJ, et al. Antibody-drug conjugates designed to eradicate tumors with homogeneous and heterogeneous expression of the target antigen. Cancer research. 2006;66:3214-21. 5. Li F, Emmerton KK, Jonas M, Zhang X, Miyamoto JB, Setter JR, et al. Intratumoral payload release influences the potency and bystander-killing effects of antibody-drug conjugates in preclinical models. Cancer research. 2016. 6. Pfeifer M, Zheng B, Erdmann T, Koeppen H, McCord R, Grau M, et al. Anti-CD22 and anti-CD79B antibody drug conjugates are active in different molecular diffuse large B-cell lymphoma subtypes. Leukemia. 2015;29:1578-86. 7. Jacobsen ED, Sharman JP, Oki Y, Advani RH, Winter JN, Bello CM, et al. Brentuximab vedotin demonstrates objective responses in a phase 2 study of relapsed/refractory DLBCL with variable CD30 expression. Blood. 2015;125:1394-402. 8. Bartlett NL, Smith MR, Siddiqi T, Advani RH, O'Connor OA, Sharman JP, et al. Brentuximab vedotin activity in diffuse large B-cell lymphoma with CD30 undetectable by visual assessment of conventional immunohistochemistry. Leukemia & lymphoma. 2016:1-10. 9. Boghaert ER, Khandke K, Sridharan L, Armellino D, Dougher M, Dijoseph JF, et al. Tumoricidal effect of calicheamicin immuno-conjugates using a passive targeting strategy. International journal of oncology. 2006;28:675-84. 10. Lam M-H, Lucas J, Maderna A, Wald H, Wojciechowicz M, Dushin R, et al. Abstract 4837: Extracellular proteolytic cleavage of peptide-linked antibody-drug conjugates promotes bystander killing of cancer cells. Cancer research. 2014;74:4837-. 11. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646- 74. 12. Hanahan D, Coussens LM. Accessories to the crime: functions of cells recruited to the tumor microenvironment. Cancer cell. 2012;21:309-22. 13. Bingle L, Brown NJ, Lewis CE. The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. The Journal of pathology. 2002;196:254- 65. 14. Cassetta L, Pollard JW. Cancer immunosurveillance: role of patrolling monocytes. Cell research. 2016;26:3-4. 15. Allavena P, Mantovani A. Immunology in the clinic review series; focus on cancer: tumour-associated macrophages: undisputed stars of the inflammatory tumour microenvironment. Clinical and experimental immunology. 2012;167:195-205. 16. Mantovani A, Allavena P. The interaction of anticancer therapies with tumor-associated macrophages. The Journal of experimental medicine. 2015;212:435-45.

17. Mitchem JB, Brennan DJ, Knolhoff BL, Belt BA, Zhu Y, Sanford DE, et al. Targeting tumor-infiltrating macrophages decreases tumor-initiating cells, relieves immunosuppression, and improves chemotherapeutic responses. Cancer research. 2013;73:1128-41. 18. DeNardo DG, Brennan DJ, Rexhepaj E, Ruffell B, Shiao SL, Madden SF, et al. Leukocyte complexity predicts breast cancer survival and functionally regulates response to chemotherapy. Cancer discovery. 2011;1:54-67. 19. Petricevic B, Laengle J, Singer J, Sachet M, Fazekas J, Steger G, et al. Trastuzumab mediates antibody-dependent cell-mediated cytotoxicity and phagocytosis to the same extent in both adjuvant and metastatic HER2/neu breast cancer patients. Journal of translational medicine. 2013;11:307. 20. Shi Y, Fan X, Deng H, Brezski RJ, Rycyzyn M, Jordan RE, et al. Trastuzumab triggers phagocytic killing of high HER2 cancer cells in vitro and in vivo by interaction with Fcgamma receptors on macrophages. Journal of immunology. 2015;194:4379-86. 21. Lewis Phillips GD, Li G, Dugger DL, Crocker LM, Parsons KL, Mai E, et al. Targeting HER2-positive breast cancer with trastuzumab-DM1, an antibody-cytotoxic drug conjugate. Cancer research. 2008;68:9280-90. 22. Li D, Poon KA, Yu SF, Dere R, Go M, Lau J, et al. DCDT2980S, an anti-CD22- monomethyl auristatin E antibody-drug conjugate, is a potential treatment for non-Hodgkin lymphoma. Molecular cancer therapeutics. 2013;12:1255-65. 23. Doronina SO, Toki BE, Torgov MY, Mendelsohn BA, Cerveny CG, Chace DF, et al. Development of potent monoclonal antibody auristatin conjugates for cancer therapy. Nature biotechnology. 2003;21:778-84. 24. Sussman D, Smith LM, Anderson ME, Duniho S, Hunter JH, Kostner H, et al. SGN- LIV1A: a novel antibody-drug conjugate targeting LIV-1 for the treatment of metastatic breast cancer. Molecular cancer therapeutics. 2014;13:2991-3000. 25. Okeley NM, Miyamoto JB, Zhang X, Sanderson RJ, Benjamin DR, Sievers EL, et al. Intracellular activation of SGN-35, a potent anti-CD30 antibody-drug conjugate. Clinical cancer research : an official journal of the American Association for Cancer Research. 2010;16:888-97. 26. Francisco JA, Cerveny CG, Meyer DL, Mixan BJ, Klussman K, Chace DF, et al. cAC10- vcMMAE, an anti-CD30-monomethyl auristatin E conjugate with potent and selective antitumor activity. Blood. 2003;102:1458-65. 27. Doronina SO, Mendelsohn BA, Bovee TD, Cerveny CG, Alley SC, Meyer DL, et al. Enhanced activity of monomethylauristatin F through monoclonal antibody delivery: effects of linker technology on efficacy and toxicity. Bioconjugate chemistry. 2006;17:114-24. 28. Zeisberger SM, Odermatt B, Marty C, Zehnder-Fjallman AH, Ballmer-Hofer K, Schwendener RA. Clodronate-liposome-mediated depletion of tumour-associated macrophages: a new and highly effective antiangiogenic therapy approach. British journal of cancer. 2006;95:272-81. 29. McDonagh CF, Kim KM, Turcott E, Brown LL, Westendorf L, Feist T, et al. Engineered anti-CD70 antibody-drug conjugate with increased therapeutic index. Molecular cancer therapeutics. 2008;7:2913-23. 30. Stein EM, Stein A, Walter RB, Fathi AT, Lancet JE, Kovacsovics TJ, et al. Interim Analysis of a Phase 1 Trial of SGN-CD33A in Patients with CD33-Positive Acute Myeloid Leukemia (AML). Blood. 2014;124:623-.

31. Phillips AC, Boghaert ER, Vaidya KS, Mitten MJ, Norvell S, Falls HD, et al. ABT-414, an Antibody-Drug Conjugate Targeting a Tumor-Selective EGFR Epitope. Molecular cancer therapeutics. 2016. 32. Hu S, Xu-Monette ZY, Balasubramanyam A, Manyam GC, Visco C, Tzankov A, et al. CD30 expression defines a novel subgroup of diffuse large B-cell lymphoma with favorable prognosis and distinct gene expression signature: a report from the International DLBCL Rituximab-CHOP Consortium Program Study. Blood. 2013;121:2715-24. 33. Campuzano-Zuluaga G, Cioffi-Lavina M, Lossos IS, Chapman-Fredricks JR. Frequency and extent of CD30 expression in diffuse large B-cell lymphoma and its relation to clinical and biologic factors: a retrospective study of 167 cases. Leukemia & lymphoma. 2013;54:2405-11. 34. Slack GW, Steidl C, Sehn LH, Gascoyne RD. CD30 expression in de novo diffuse large B-cell lymphoma: a population-based study from British Columbia. British journal of haematology. 2014;167:608-17. 35. Coiffier B, Thieblemont C, de Guibert S, Dupuis J, Ribrag V, Bouabdallah R, et al. A phase II, single-arm, multicentre study of coltuximab ravtansine (SAR3419) and rituximab in patients with relapsed or refractory diffuse large B-cell lymphoma. British journal of haematology. 2016;173:722-30. 36. Morschhauser F, Flinn I, Advani RH, Diefenbach CS, Kolibaba K, Press OW, et al. Updated Results of a Phase II Randomized Study (ROMULUS) of Polatuzumab Vedotin or Pinatuzumab Vedotin Plus Rituximab in Patients with Relapsed/Refractory Non-Hodgkin Lymphoma. Blood. 2014;124:4457-. 37. Jeffrey SC, Burke PJ, Lyon RP, Meyer DW, Sussman D, Anderson M, et al. A potent anti-CD70 antibody-drug conjugate combining a dimeric pyrrolobenzodiazepine drug with site- specific conjugation technology. Bioconjugate chemistry. 2013;24:1256-63. 38. Leek RD, Lewis CE, Whitehouse R, Greenall M, Clarke J, Harris AL. Association of macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma. Cancer research. 1996;56:4625-9. 39. Steidl C, Lee T, Shah SP, Farinha P, Han G, Nayar T, et al. Tumor-associated macrophages and survival in classic Hodgkin's lymphoma. The New England journal of medicine. 2010;362:875-85. 40. Tan KL, Scott DW, Hong F, Kahl BS, Fisher RI, Bartlett NL, et al. Tumor-associated macrophages predict inferior outcomes in classic Hodgkin lymphoma: a correlative study from the E2496 Intergroup trial. Blood. 2012;120:3280-7. 41. Kridel R, Steidl C, Gascoyne RD. Tumor-associated macrophages in diffuse large B-cell lymphoma. Haematologica. 2015;100:143-5. 42. Riihijarvi S, Fiskvik I, Taskinen M, Vajavaara H, Tikkala M, Yri O, et al. Prognostic influence of macrophages in patients with diffuse large B-cell lymphoma: a correlative study from a Nordic phase II trial. Haematologica. 2015;100:238-45. 43. Nam SJ, Go H, Paik JH, Kim TM, Heo DS, Kim CW, et al. An increase of M2 macrophages predicts poor prognosis in patients with diffuse large B-cell lymphoma treated with rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone. Leukemia & lymphoma. 2014;55:2466-76. 44. Cai QC, Liao H, Lin SX, Xia Y, Wang XX, Gao Y, et al. High expression of tumor- infiltrating macrophages correlates with poor prognosis in patients with diffuse large B-cell lymphoma. Medical oncology. 2012;29:2317-22.

45. Wada N, Zaki MA, Hori Y, Hashimoto K, Tsukaguchi M, Tatsumi Y, et al. Tumour- associated macrophages in diffuse large B-cell lymphoma: a study of the Osaka Lymphoma Study Group. Histopathology. 2012;60:313-9. 46. Ruffell B, Coussens LM. Macrophages and therapeutic resistance in cancer. Cancer cell. 2015;27:462-72. 47. Oflazoglu E, Stone IJ, Gordon KA, Grewal IS, van Rooijen N, Law CL, et al. Macrophages contribute to the antitumor activity of the anti-CD30 antibody SGN-30. Blood. 2007;110:4370-2. 48. Desnoyers LR, Vasiljeva O, Richardson JH, Yang A, Menendez EE, Liang TW, et al. Tumor-specific activation of an EGFR-targeting probody enhances therapeutic index. Science translational medicine. 2013;5:207ra144. 49. Ries CH, Cannarile MA, Hoves S, Benz J, Wartha K, Runza V, et al. Targeting tumor- associated macrophages with anti-CSF-1R antibody reveals a strategy for cancer therapy. Cancer cell. 2014;25:846-59.

Figure Legend

Figure 1: Antigen-independent antitumor activity correlates with intratumoral MMAE release in

L-428 xenograft. A. Tumor growth of L-428 tumors, after the treatment of αCD30-vcMMAE or hIgG-vcMMAE (both ADCs given at 2 mg/kg, once ip), n=8 per group. Using two-way ANOVA test p<0.0001 (untreated vs. hIgG1-vsMMAE); p=0.87 (hIgG1-vcMMAE vs. αCD30- vcMMAE).B. In a parallel cohort of animals, L-428 tumors were collected and intratumoral

MMAE concentration was measured by LC/MS following ADC treatment, n=3 per time point, p=0.06 by two-way ANOVA test. The efficacy study has been repeated at least twice at 3 mg/kg dose level, while the LC/MS measurement was performed once as shown in B.

Figure 2: Membrane permeability of payload and antigen-independent antitumor activity

A. L-428 tumors were treated with targeting αCD30-vcMMAE and αCD30-vcMMAF ADCs, or non-binding hIgG-vcMMAE and hIgG-vcMMAF ADCs. All ADCs were given once ip at 3 mg/kg, n=5 per group, p<0.001 (hIgG-vcMMAE vs. hIgG-vcMMAF) by two-way ANOVA test.

B. Structure of MMAE and MAMF. LC/MS measurement of intratumoral released payload concentration in L-428 tumors collected three days after treatment of indicated ADCs. n=3 for hIgG-vcMMAE and n=2 for all other groups. Experiment shown in A has been repeated once more.

Figure 3: TAMs bind to hIgG-vcMMAE in L-428 xenograft. A. IHC analysis of F4/80 (left, red), anti-auristatin (middle, brown), or both antibodies (right) of L-428 tumors collected 48 hours post hIgG-vcMMAE treatment (3 mg/kg once ip). Images represent staining performed on tumor sections from two individual mice. Scale bar: 50µm. B. Flow cytometry analysis of

F4/80, anti-auristatin, or both antibodies in L-428 tumors 48 hours post hIgG-vcMMAE treatment (3 mg/kg once ip). This experiment has been performed twice.

Figure 4: TAM infiltration correlates with Ag-independent ADC activity in vivo. A. IHC staining of F4/80 (brown) on L-428, KM-H2, and BR620 tumors. Scale bar: 100µm. B. Tumor growth after the treatment of hIgG-vcMMAE in L-428 tumor model (3mg/kg once ip, n=5,

P<0.001), KM-H2 xenografts (3mg/kg once ip, n=5, p=0.004 comparing untreated vs. hIgG- vcMMAE treated tumors.), and BR620 PDX tumors (3mg/kg, q4dx4 ip, n=5, p<0.001). C. IHC staining of F4/80 on DOHH-2, SU-DHL-8, and Karpas 299 tumors. D. Tumor growth after the treatment of indicated ADCs in DOHH-2, SU-DHL-8, and Karpas 299 tumors. All ADCs given at 3 mg/kg once ip., n=5 per group. Two-way ANOVA test was performed to compare the untreated group and hIgG-vcMMAE treated group. p=0.7(DOHH-2), p=0.65(SU-DUL-8), and p=0.01(Karpas-299). L-428, KM-H2, and Karpas 299 xenograft studies have been repeated at least three times. BR620, DOHH-2, and SU-DHL-8 xenograft studies have been repeated twice.

Figure 5: Fc receptor-ADC interaction is required for TAM-mediated ADC activity. A. Growth responses of L-428 tumors after the treatment of hIgG-vcMMAE and hIgG1V1-vcMMAE. Both

ADCs were given at 3 mg/kg once ip, n=5 per group. B. Growth kinetics of KM-H2 tumors after the treatment of hIgG-vcMMAE and hIgG1V1-vcMMAE (3 mg/kg once ip), n=5 per group. C.

Tumor growth of BR620 patient derived xenograft models after the treatment of hIgG-vcMMAE and hIgG1V1-vcMMAE (3 mg/kg, q4dX4 ip), n=5 per group. D. KM-H2 xenograft that were either untreated, received hIgG1-vcMMAE, or received hIVIg 24 hours prior to hIgG1- vcMMAE. hIVIg was given at 10 mg/kg iv; hIgG1-vcMMAE was given at 3 mg/kg once ip, n=5 per group. L-428 studies were repeated separately two more times. KM-H2 and BR-620 studies were performed once, respectively.

Figure 1. Antitumor activity correlates with extent of intratumoral MMAE release in L-428 xenograft

) 3

m 1 0 0 0 u n tre a te d

m (

 C D 3 0 -v c M M A E

e 8 0 0

h Ig G -v c M M A E

m u

l 6 0 0

o 4 0 0

m u

t 2 0 0

n a

e 0

M 0 5 1 0 1 5

B )

M 2 5 0

(

E 2 0 0 A U C

A  C D 3 0 -v c M M A E 9 0 7 M 1 5 0 h Ig G -v c M M A E

M 6 6 9

l a

r 1 0 0

o m

u 5 0

a r

t 0 n I 0 5 1 0 1 5 D a y s p o s t d o s e

Figure 2: Membrane permeability of payload and antigen-independent

activity )

A. 3

m 1 0 0 0 u n tre a te d m

( h Ig G -v c M M A E

e 8 0 0  C D 3 0 -v c M M A E m

u h Ig G -v c M M A F l 6 0 0

o  C D 3 0 -v c M M A F

r R x

o 4 0 0

m u

t 2 0 0

n a

e 0

M 0 1 0 2 0 3 0 4 0 5 0 D a y s p o s t im p la n t

MMAE MMAF

Intratumoral released drug (nM) αCD30-vcMMAE MMAE 260 αCD30-vcMMAF MMAF 185 hIgG-vcMMAE MMAE 145 hIgG-vcMMAF MMAF 119

Figure 3. Tumor infiltrating macrophages bind to ADC in L-428 xenograft

A. F4/80 anti-auristatin F4/80+anti-auristatin

B. F4/80 anti-auristatin F4/80+anti-auristatin

Figure 4: TAM infiltration correlates with Ag-independent ADC activity in vivo

A L-428 KM-H2 BR620

) )

L-428 ) KM-H2 BR620

3 m

m 1 0 0 0

1 0 0 0 m 1 0 0 0 u n tre a te d m

m u n tre a te d

(

( h Ig G -v c M M A E

u n tre a te d

h Ig G -v c M M A E e

e e

h Ig G -v c M M A E m m

m  C D 3 0 -v c M M A E

o v

v 5 0 0

5 0 0 v

5 0 0

o o

Rx o

m m

R x u

a a

e 0 e 0 e 0

M 0 2 0 4 0 6 0 M 0 2 0 4 0 6 0 8 0 M 1 0 2 0 3 0 4 0 D a y s p o s t d o s e D a y s p o s t im p la n t D a y s p o s t im p la n t C DOHH-2 SU-DHL-8 Karpas 299

D DOHH2 SU-DHL8 Karpas 299

)

m 3 1 0 0 0 3

u n tre a te d

m m

m 1 5 0 0 1 2 0 0

(

h Ig G -v c M M A E m

(

A b 1 -v c M M A E e

m u

l 1 0 0 0 8 0 0

o o

v 5 0 0 u n tre a te d

r h Ig G -v c M M A E

o o 5 0 0 o 4 0 0

m a C D 3 0 -v c M M A E m

u n tre a te d m

h Ig G -v c M M A E

n n

A b 1 -v c M M A E n

e e

e 0 0 0

M M M 0 5 1 0 1 5 0 5 1 0 1 5 0 1 0 2 0 3 0 4 0 D a y s p o s t d o s e D a y s p o s t d o s e D a y s p o s t d o s e

Figure 5. Fc-ADC interaction is required for TAM-mediated ADC activity

A L-428 C BR620

U n tre a te d

) 1 0 0 0

h Ig G -v c M M A E 3

) 1 0 0 0 U n tre a te d

3 m

m h Ig G 1 V 1 -v c M M A E h Ig G -v c M M A E

m 8 0 0

(

( e

h Ig G 1 V 1 -v c M M A E

e m

u 6 0 0

r V 4 0 0

5 0 0

m u

m 2 0 0

T u

n e

a 0

e M

M 0 0 2 0 4 0 6 0 0 2 0 4 0 6 0 8 0 1 0 0 D a y s p o s t im p la n t D a y s p o s t im p la n t B KM-H2 D KM-H2

) 1 0 0 0 ) 3 U n tre a te d

3 U n tre a te d m

h Ig G -v c M M A E m m

( 1 0 0 0

8 0 0 m h Ig G -v c M M A E

(

e h Ig G 1 V 1 -v c M M A E e

m h Ig G -v c M M A E + h IV Ig

u l

6 0 0 u

r o

4 0 0 o 5 0 0 m

Rx m u

u Rx

n 2 0 0

e M 0 M 0 0 2 0 4 0 6 0 0 2 0 4 0 6 0 D a y s p o s t im p la n t D a y s p o s t im p la n t

Tumor associated macrophages can contribute to antitumor activity through Fc γR-mediated processing of antibody-drug conjugates

Fu Li, Michelle Ulrich, Mechthild Jonas, et al.

Mol Cancer Ther Published OnlineFirst March 24, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-17-0019

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2017/03/24/1535-7163.MCT-17-0019.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/early/2017/04/19/1535-7163.MCT-17-0019. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.