Tumor-associated PD-L1 Expression Research Articles

A Preclinical Assessment of 89Zr-atezolizumab Identifies a Requirement for Carrier Added Formulations Not Observed with 89Zr-C4

The swell of experimental imaging technologies to noninvasively measure immune checkpoint protein expression presents the opportunity for rigorous comparative studies toward identifying a gold standard. 89Zr-atezolizumab is currently in man, and early data show tumor targeting but also abundant uptake in several normal tissues. Therefore, we conducted a reverse translational study both to understand if tumor to normal tissue ratios for 89Zr-atezolizumab could be improved and to make direct comparisons to 89Zr-C4, a radiotracer that we showed can detect a large dynamic range of tumor-associated PD-L1 expression. PET/CT and biodistribution studies in tumor bearing immunocompetent and nu/nu mice revealed that high specific activity 89Zr-atezolizumab (∼2 μCi/μg) binds to PD-L1 on tumors but also results in very high uptake in many normal mouse tissues, as expected. Unexpectedly, 89Zr-atezolizumab uptake was generally higher in normal mouse tissues compared to 89Zr-C4 and lower in H1975, a tumor model with modest PD-L1 expression. Also unexpectedly, reducing the specific activity at least 15-fold suppressed 89Zr-atezo uptake in normal mouse tissues but increased tumor uptake to levels observed with high specific activity 89Zr-C4. In summary, these data reveal that low specific activity 89Zr-atezo may be necessary for accurately measuring PD-L1 in the tumor microenvironment, assuming a threshold can be identified that preferentially suppresses binding in normal tissues without reducing binding to tumors with abundant expression. Alternatively, high specific activity approaches like 89Zr-C4 PET may be simpler to implement clinically to measure the broad dynamic range of PD-L1 expression known to manifest among tumors.

PD-L2 Expression in Human Tumors: Relevance to Anti-PD-1 Therapy in Cancer.

Purpose: Tumor-associated PD-L1 expression is predictive of clinical response to PD-1-directed immunotherapy. However, PD-L1-negative patients may also respond to PD-1 checkpoint blockade, suggesting that other PD-1 ligands may be relevant to the clinical activity of these therapies. The prevalence of PD-L2, the other known ligand of PD-1, and its relationship to response to anti-PD-1 therapy were evaluated.Experimental Design: PD-L2 expression was assessed in archival tumor tissue from seven indications using a novel immunohistochemical assay. In addition, relationships between clinical response and PD-L2 status were evaluated in tumor tissues from patients with head and neck squamous cell carcinoma (HNSCC) with recurrent or metastatic disease, treated with pembrolizumab.Results: PD-L2 expression was observed in all tumor types and present in stromal, tumor, and endothelial cells. The prevalence and distribution of PD-L2 correlated significantly with PD-L1 (P = 0.0012-<0.0001); however, PD-L2 was detected in the absence of PD-L1 in some tumor types. Both PD-L1 and PD-L2 positivity significantly predicted clinical response to pembrolizumab on combined tumor, stromal and immune cells, with PD-L2 predictive independent of PD-L1. Response was greater in patients positive for both PD-L1 and PD-L2 (27.5%) than those positive only for PD-L1 (11.4%). PD-L2 status was also a significant predictor of progression-free survival (PFS) with pembrolizumab independent of PD-L1 status. Longer median times for PFS and overall survival were observed for PD-L2-positive than PD-L2-negative patients.Conclusions: Clinical response to pembrolizumab in patients with HNSCC may be related partly to blockade of PD-1/PD-L2 interactions. Therapy targeting both PD-1 ligands may provide clinical benefit in these patients. Clin Cancer Res; 23(12); 3158-67. ©2017 AACR.

Abstract 1478: Anti-EMP2 IgG1 combined with anti-PD1/PDL1 antibodies synergistically reduce tumor load in animal models of breast cancer

Abstract Purpose: Despite significant advances in detection and medicine, the incidence and mortality due to breast cancer world-wide is still unacceptably high. Our recent work reveals that tetraspan protein epithelial membrane protein-2 (EMP2) is up-regulated in up to 70% of patients with invasive breast cancer and its expression is further augmented in metastatic lesions. New data suggests that EMP2 expression correlate with an increase in tumor associated PDL1 expression. We thus hypothesized that EMP2 expression regulated tumor immunogenicity through either direct control of PDL1 expression or MHC Class I processing. Recently, to determine the therapeutic potential of EMP2, our group created a panel of anti-EMP2 antibodies and new data shows that tumors treated with anti-EMP2 IgG1 show an increase in immune infiltrates. We thus proposed that EMP2 positive tumors will be amendable to anti-PD1/PDL1 therapy. Methods:A panel of breast cancer cell lines, both TNBC as well as hormone positive cell lines, were transfected with a vector to overexpress EMP2 or to reduce its levels. To determine if EMP2 expression influenced the expression of immune associated genes, RNAseq was performed. In addition, PDL1 protein expression was measured using standard immunohistochemistry or western blot analysis. Immune cell populations in various xenografts models were quantitated using standard immunohistochemistry. To determine the efficacy of combination anti-EMP2 and anti-PD1 therapy, syngeneic 4T1/firefly luciferase mouse models were created in BALB/c mice. Briefly, cells were injected into the mammary fat pad of female BALB/c mice, and once they approached 100 mm3, were injected IP with a)10 mg/kg dose twice a week of anti-EMP2 antibody, b) 5mg/kg twice a week with anti-PD-1 antibody (BioXCell), c) 10mg/kg ant-EMP2 and 5mg/kg of anti-PD-1 antibody twice a week, or d) saline control. Results: In multiple breast cell lines, EMP2 levels correlated with an increase in PDL1 expression, and treatment with anti-EMP2 IgG1 increased the number of infiltrating leukocytes. Using syngeneic mouse models, tumors treated with a combination of anti-EMP2 and anti-PD-1 antibodies grew slowly and had the smallest total volume compared to all other treatment groups. Anti-EMP2 IgG1 treatment was superior in reducing overall tumor volume compared to anti-PD-1 antibody treatment alone. Conclusions: Our preliminary data strongly supports the synergistic efficacy of anti-EMP2 IgG1 and anti-PD-1 antibody in reducing tumor load, suggesting that using anti-EMP2 treatment with an anti-PD1/PDL1 antibody may improve survival without inducing systemic toxicity as seen with chemotherapy. We predict that this is the direct result of EMP2 expression altering the immunogenicity of the tumor and in support of this, RNAseq data and western blot data suggest that EMP2 levels regulate PDL1 and MHC associated gene expression. Citation Format: Negin Ashki, Jessica Tsui, Madhuri Wadehra. Anti-EMP2 IgG1 combined with anti-PD1/PDL1 antibodies synergistically reduce tumor load in animal models of breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1478.

Abstract OT1-01-07: nab-paclitaxel (nab-P) plus nivolumab (Nivo) in human epidermal growth factor receptor 2 (HER2)–negative recurrent metastatic breast cancer (MBC)

Abstract Background: Nivo is an inhibitory antibody against programmed death receptor-1 (PD-1), a regulator of antitumor immunity. Nivo is approved for treatment of unresectable or metastatic melanoma and disease progression (PD) following ipilimumab or, in BRAF V600 mutation–positive melanoma, following a BRAF inhibitor, and for metastatic squamous non-small cell lung cancer (NSCLC) following PD during or after platinum-based chemotherapy. Nivo and other immune checkpoint inhibitors are also being investigated in other tumor-types. nab-P is a novel taxane formulation and does not require prophylaxis with immunosuppressive steroids. It has demonstrated superior efficacy over control regimens in phase III studies of MBC, pancreatic cancer, and NSCLC. This open-label, 6-arm, multicenter phase I trial will evaluate the safety of Nivo with nab-P in 3 cancer types (2 arms/disease): MBC, advanced NSCLC (+ carboplatin), and advanced pancreatic cancer (± gemcitabine). The study design for the MBC portion is described below. Methods: Eligibility criteria include histologically/cytologically confirmed HER2-negative MBC; 1 prior chemotherapy for MBC, including an anthracycline unless clinically contraindicated; no relapse &amp;lt; 12 months after taxane adjuvant therapy; measurable disease by RECIST v1.1; ECOG performance status 0-1; adequate organ function; and preexisting peripheral neuropathy grade &amp;lt; 2. Patients (pts) with MBC will be treated in 2 arms: nab-P 100 mg/m2 on days 1, 8, and 15 of each 28-day cycle plus Nivo 3 mg/kg on days 1 and 15 starting at cycle 3 or nab-P 260 mg/m2 on day 1 of each 21-day cycle plus Nivo 5 mg/kg on day 15 starting at cycle 3. Pts will be treated until PD or allowed to continue treatment beyond RECIST v1.1–defined PD if they continue to meet study eligibility; do not have rapid PD or clinical deterioration or unacceptable toxicities; and can benefit from continuation of study treatment in the treating physician's opinion and will not delay an imminent intervention to prevent serious complications of PD. The primary endpoints of the study are the number of pts with dose-limiting toxicities (DLTs) in each treatment arm (part 1) and the percentage of pts with grade 3/4 treatment-emergent adverse events (TEAEs) or treatment discontinuation due to a TEAE (parts 1 and 2). Part 1 of the study will assess whether the starting dose of Nivo is deemed safe (≤ 1 DLT in 6 pts); otherwise, the Nivo dose will be de-escalated and assessed in a new cohort at the next lower dose level. The Nivo dose in combination with nab-P deemed safe in a treatment arm may be further assessed in part 2 of the study, with enrollment expanded to an additional ∼ 14 pts/arm (total of 20 Nivo-treated pts/arm). Secondary study endpoints include TEAEs leading to dose reduction, delay, interruption, or treatment discontinuation; progression-free survival; overall survival; disease control rate; overall response rate; and duration of response (per RECIST v1.1). Exploratory endpoints include tumor-associated PD-L1 expression, modulation of immune activation in the tumor and peripheral blood in response to Nivo treatment, Nivo serum levels, and development of anti-globulin antibodies. ClinicalTrials.gov identifier NCT02309177. Citation Format: Waterhouse D, Gutierrez M, Bekaii-Saab T, DeRosa W, Wainberg Z, George B, Duval Fraser C, Ko A, Pierce DW, Stergiopoulos S, Soliman H. nab-paclitaxel (nab-P) plus nivolumab (Nivo) in human epidermal growth factor receptor 2 (HER2)–negative recurrent metastatic breast cancer (MBC). [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr OT1-01-07.

Abstract A90: Tumor associated PDL1 expression pattern in microsopically tumor positive sentinel lymph nodes (SLN) in patients with melanoma

Abstract Background: Anti-PD1/PDL1 immunotherapy has demonstrated unprecedented clinical antitumor activity in patients with metastatic melanoma supporting planned testing as adjuvant therapy in patients with operable disease at high risk of recurrence and death. Characterization of PDL1 expression within clinically/radiologically negative but microscopically tumor positive sentinel lymph nodes is critical to our understanding of the relevance of this immune checkpoint pathway for adjuvant therapy. Methods: Cut sections (5 µm) of sentinel lymph node from 25 patients reported positive for melanoma micrometastasis were included. H&amp;E staining was performed, followed by PD-L1 immunohistochemical (IHC) staining using a preliminary IHC assay with anti-PD-L1 antibody clone 22C3. Slides from 2 patients were also stained using an anti-HMB45/MelA protocol to better ascertain the presence and/or localization of melanoma lesions in the tissue in order to facilitate interpretation of the PD-L1 staining in those samples. The slides were separately evaluated by 2 pathologists. Samples containing metastatic melanoma lesions were scored separately for PD-L1 expression in intratumoral (including along tumor periphery but with clear tumor cell labeling) and peritumoral (expression external to tumor nodule in immediately surrounding tissue) locations. For intratumoral signal, attempts were made to classify the expression as tumor cell associated (T), non-tumor cell associated (NT), or both (T/NT). Scores were assigned using a 0-5 semiquantitative scale assessing prevalence of positive cells where 0 = negative, 1 = minimal or rare, 2 = low, 3 = moderate, 4 = high, and 5 = very high. Samples where melanin content was too high to confidently assign a PD-L1 score were specifically noted. Results: Twenty patients where metastatic melanoma presence in regional sentinel nodes was confirmed by H&amp;E review of the cut sections were included in the final analysis of PDL1 expression in the target population. Table 1 summarizes the study results. Except for samples where melanin content was too high to confidently assign a PD-L1 score, all SLNs examined demonstrated evidence of intratumoral and/or peritumoral PD-L1 expression. Conclusions: PD-L1 expression is readily detectable intratumorally and/or peritumorally within melanoma micrometastases in the SLN. These results support the testing of a therapeutic role for PD1/PDL1 inhibition in the adjuvant therapy setting, targeting melanoma micrometastases. Patient with SLN+ Tumor Associated Expression PD-L1 Score / Tumor ---- PD-L1 Score / Peritumoral 1. 1 T, very weak ---- 1 2. 4, predominantly T, peripheral ---- 1 3. melanin confounds ---- 2 4. 0 ---- 2 (adjacent subcapsular sinus) 5. 4 T/NT---- 2 6. 2, predominantly NT ---- 3 7. 0 ---- 3 (adjacent subcapsular sinus) 8. 2.5, predominantly NT ---- 3 9. 4.5, predominantly T, weak ---- 2 (adjacent sinus) 10. melanin confounds ---- 0 11. 1 T ---- 2.5 12. 0* ---- 2 13. 1 T ---- 1 14. 2.5 NT ---- 2 15. 2 predominantly NT ---- 3 16. 1 T ---- 2 17. melanin confounds ---- 0 18. 1 ---- 1 19. 1 NT ---- 2.5 20. 2 T, melanin confounds ---- 2 *Very few tumor cells. In one region where a few tumor cells were present, a dense area of PD-L1 positivity was present directly adjacent but was impossible to say if all were non-tumor; this is interpreted as most likely. Citation Format: Ahmad A. Tarhini, Jennifer A. Yearley, Christopher Gibson, Uma N. M. Rao, Cindy Sander, John M. Kirkwood. Tumor associated PDL1 expression pattern in microsopically tumor positive sentinel lymph nodes (SLN) in patients with melanoma. [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy: A New Chapter; December 1-4, 2014; Orlando, FL. Philadelphia (PA): AACR; Cancer Immunol Res 2015;3(10 Suppl):Abstract nr A90.

Tumor associated PD-L1 expression pattern in microscopically tumor positive sentinel lymph nodes in patients with melanoma.

BackgroundCharacterization of PD-L1 expression within clinically/radiologically negative but microscopically tumor positive sentinel lymph nodes (SLN) is important to our understanding of the relevance of this immune checkpoint pathway for adjuvant therapy.MethodsPatients included had primary cutaneous melanoma, Breslow thickness of 2.01–4.0 or >4 mm with or without tumor ulceration (T3a, T3b, T4a, T4b). All patients had microscopically tumor positive SLN. Hematoxylin and eosin (H&E) staining was performed, followed by PD-L1 immunohistochemical (IHC) staining using a preliminary IHC assay with anti-PD-L1 antibody clone 22C3. The slides were separately evaluated by two pathologists (JY and CG). Samples containing metastatic melanoma lesions were scored separately for PD-L1 expression in intratumoral and peritumoral locations, by utilizing two scoring methods.ResultsTwenty-four patients where metastatic melanoma presence in the SLN was confirmed by H&E review of the cut sections were included in the final analysis of PD-L1 expression. SLN tumor size ranged from 1 to 2 mm. For three patients, the melanin content was too high to confidently assign a PD-L1 score. For the remaining 21 patients, all had some evidence of either intratumoral or peritumoral PD-L1 expression. The frequency of intratumoral tumor-associated PD-L1 expression was: 0 % of tumor cells (3 pts, 14 %); <1 % (5 pts, 24 %); 1–10 % (6 pts, 29 %) and >10 % (7 pts, 33 %).ConclusionsTumor-associated PD-L1 expression is readily detectable within melanoma micrometastases in the SLN of the majority of patients. These results support the testing of a therapeutic role for PD1/PD-L1 inhibition in the adjuvant setting, targeting melanoma micrometastases.

Abstract 249: Assessment of PD-L1 expression and tumor-associated lymphocytes in pediatric cancer tissues

Abstract PD-1 signaling in the tumor microenvironment dampens immune responses to cancer and blocking this axis induces anti-tumor effects in several malignancies. Some studies have demonstrated increased efficacy of PD-1 blockade when tumor cells express PD-L1. Clinical studies of PD-1 blockade have not yet been conducted in pediatric patients and little is known regarding PD-L1 expression in common childhood cancers. We characterized PD-L1 expression and Tumor Associated Immune Cells (TAIC, lymphocytes and macrophages) in common pediatric cancers. Whole slide sections (n = 91) and tissue microarrays (n = 365) were evaluated by IHC for PD-L1 expression using the 28-8 anti-PD-L1 mAb in an automated Dako assay. PD-L1 expression was considered positive when at least 1% of tumor cells analyzed demonstrated plasma membrane staining. TAIC were also assessed for expression of PD-L1, and a subset of 60 tumors were assessed for CD3, CD4, CD8, CD45RO, PD-1, and FoxP3 expression on TAIC. Nine percent (N = 40) of evaluable tumors expressed PD-L1. Highest frequency histotypes comprised non-Hodgkin lymphoma (80%, 8/10), glioblastoma multiforme (30%, 6/20), and neuroblastoma (14%, 17/118). No PD-L1 staining was observed in Ewing sarcoma (0/20) or medulloblastoma (0/40). Despite a relatively low frequency of PD-L1+ pediatric cancers, the majority contained TAIC (73%, 334/456). Of these TAIC+ tumors, 73% of samples contained lymphocytes only, 25% contained lymphocytes and macrophages and 3% contained macrophages only. Twenty-one percent of TAIC+ samples demonstrated PD-L1 expression on TAIC, mostly comprising PD-L1 expression on macrophages (65%, 60/92), while lymphocytes expressed PD-L1 in only 7% (22/324). Taken together, PD-L1 was expressed in tumor and/or TAIC in 20% (90/456). Sixty samples were further analyzed to characterize TAIC, with 77% of the samples demonstrating infiltration of CD8+ T cells, most of which expressed CD45RO. Fox-P3 and PD-1 were expressed in 42% and 47% of the samples respectively. In summary, this provides the most comprehensive analysis of PD-L1 expression in pediatric associated cancer to date. Results show that a subset of pediatric cancers demonstrate tumor associated PD-L1 expression, while a much larger fraction demonstrate infiltration with tumor associated lymphocytes. Further preclinical and clinical investigation will define the predictive nature of PD-L1 expression in childhood cancers, but available data is not sufficient to exclude enrollment to clinical trials based on PD-L1 status. Citation Format: Robbie G. Majzner, Jason S. Simon, Joseph F. Grosso, Daniel Martinez, Bruce Pawel, Mariarita Santi-Vincini, Melinda S. Merchant, Poul Sorensen, Crystal L. Mackall, John M. Maris. Assessment of PD-L1 expression and tumor-associated lymphocytes in pediatric cancer tissues. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 249. doi:10.1158/1538-7445.AM2015-249

Immunotherapy for malignant gliomas.

Cancer immunotherapy aims to harness the innate ability of the immune system to recognize and destroy malignant cells. Immunotherapy for malignant gliomas is an emerging field that promises the possibility of highly specific and less toxic treatment compared to conventional chemotherapy. In addition, immunotherapy has the added benefit of sustained efficacy once immunologic memory is induced. Although there are numerous therapeutic agents that boost general immune function and facilitate improved antitumor immunity, to date, immunotherapy for gliomas has focused primarily on active vaccination against tumor-specific antigens. The results of numerous early phase clinical trials demonstrate promising results for vaccine therapy, but no therapy has yet proven to improve survival in a randomized, controlled trial. The major barrier to immunotherapy in malignant gliomas is tumor-induced immunosuppression. The mechanisms of immunosuppression are only now being elucidated, but clearly involve a combination of factors including regulatory T cells, tumor-associated PD-L1 expression, and CTLA-4 signaling. Immunomodulatory agents have been developed to combat these immunosuppressive factors and have demonstrated efficacy in other cancers. The future of glioma immunotherapy likely lies in a combination of active vaccination and immune checkpoint inhibition.

The Expression and Biological Significance of PD-L1 on Lung Cancer Cell Lines.

Tumor-associated PD-L1 expression was recently shown to promote T-cell apoptosis and proposed as a potential mechanism of immune evasion by tumors. On the basis of the ability of tumor-associated PD-L1 to mediate activated T-cell death, it is likely that manipulation of the PD-L1 pathway at defined time points during the development of the T-cell antitumor immune response can enhance the efficacy of T-cell-based immunotherapy. Here, the levels of expression of PD-L1 on lung cancer cell lines and its role in interaction of CTL and target cells was investigated. Human PBMC derived DCs were loaded with apoptotic tumor cells and stimulated by CD40 mAb (5C11). Tumor specific CTL was generated in vitro by autologous T cells co-cultured with mature DCs. Expression of PD-L1 on lung cancer cell lines H1299 and A549 were analyzed by FCM. JAM assay was used to detect the cytolytic activity of CTL with or without blocking PD-L1 by PD-L1 mAb respectively. The concentrations of IFN-gamma in supernatants from distinct groups were analyzed by ELISA. Tumor cells-loaded mature DCs could induce the generation of the tumor specific CTL. Expression of PD-L1 was low on A549 cell, but high on H1299 cell. Blockade of PD-L1 on A549 could not improve cytolytic effect of CTL on target cells and IFN-gamma production, but fragmentation of H1299 cells and IFN-gamma production were significantly enhanced by the combination of PD-L1 mAb and CTL. Expression of PD-L1 on lung cancer cell line can decrease the cytolytic effect of CTL on target cells.