Purpose:

Although pituitary adenoma is classified as benign, Cushing disease is associated with significant morbidity due to the numerous sequelae of elevated cortisol levels. Successful therapy for Cushing disease remains elusive due to high rates of treatment-refractory recurrence. The frequent emergence of lymphocytic hypophysitis following checkpoint blockade for other cancers, as well as the expression of PD-L1 on pituitary adenomas, suggest a role for immunotherapy.

Experimental Design:

This study confirms PD-L1 expression on functioning pituitary adenomas and is the first to evaluate the efficacy of checkpoint blockade (anti–PD-L1) therapy in a preclinical model of Cushing disease.

Results:

Herein, treatment with anti–PD-L1 was successful in reducing adrenocorticotropic hormone plasma levels, decreasing tumor growth, and increasing survival in our model. Furthermore, tumor-infiltrating T cells demonstrated a pattern of checkpoint expression similar to other checkpoint blockade–susceptible tumors.

Conclusions:

This suggests that immunotherapy, particularly blockade of the PD1/PD-L1 axis, may be a novel therapeutic option for refractory Cushing disease. Clinical investigation is encouraged.

Translational Relevance

This study confirms PD-L1 expression on functioning pituitary adenomas and is the first to evaluate the efficacy of checkpoint blockade (anti–PD-L1) therapy in a preclinical model of Cushing disease. Although pituitary adenoma is classified as benign, Cushing disease is associated with significant morbidity due to the numerous sequelae of elevated cortisol levels. Successful therapy for Cushing disease remains elusive due to high rates of treatment-refractory recurrence. The frequent emergence of lymphocytic hypophysitis following checkpoint blockade for other cancers, as well as the expression of PD-L1 on pituitary adenomas, suggest a role for immunotherapy. Herein, treatment with anti–PD-L1 was successful in reducing adrenocorticotropic hormone plasma levels, decreasing tumor growth, and increasing survival in our model. Furthermore, tumor-infiltrating T-cells demonstrated a pattern of checkpoint expression similar to other checkpoint blockade–susceptible tumors. This suggests that immunotherapy, particularly blockade of the PD1/PD-L1 axis, may be a novel therapeutic option for refractory Cushing disease. Clinical investigation is encouraged.

Pituitary adenomas are among the most common intracranial tumors, occurring in up to 20% of the general population (1). While classified as benign, as many as 25%–55% of pituitary adenomas are invasive, exhibiting rapid growth patterns and posttreatment recurrence (2). Many of these tumors are associated with significant morbidity given their proximity to critical nerves and blood vessels (2). In addition, a variety of “functioning” pituitary adenomas secrete supraphysiologic levels of hormones, resulting in profound systemic effects that reflect the hormone(s) elaborated. One example is the adrenocorticotropic hormone (ACTH)-secreting pituitary adenoma, which stimulates the production and release of cortisol by the adrenal glands, resulting in Cushing disease. Cushing disease, in turn, is associated with various sequelae, including morbid weight gain, metabolic abnormalities such as diabetes and osteoporosis, immune-deficiency, reproductive dysfunction, and cardiovascular complications, among others (3–5).

Control of Cushing disease remains elusive. Trans-sphenoidal resection is the first-line intervention, but long-term follow-up reveals recurrence rates between 15% and 66% at 5–10 years (5–7). Upon recurrence, repeat surgical resection and medical therapy are variably effective, and radiation is a viable option, but can be limited by proximity to critical structures (7, 8). There is a significant need for additional, more effective adjuvant treatment options in this disease. Immunotherapy, and checkpoint blockade in particular, has gained acceptance in various cancers, but remains untried in Cushing disease and other pituitary tumors (9–17). A common target of checkpoint blockade is the PD-1/PD-L1 axis, which restricts the effector phase of the T-cell response (9). The binding of PD-L1 (on tumor or other microenvironment cells) to the PD-1 receptor on activated T cells inhibits the cytotoxic antitumor function of T cells, (18–20) while blockade of this interaction permits a perpetuated T-cell response (9). Several therapies targeting this pathway have achieved FDA approval and have shown marked success in metastatic solid tumors such as melanoma and non–small cell lung cancer (17, 21). It is interesting to note that lymphocytic hypophysitis, T-cell–based inflammation within the pituitary gland, is a common side-effect of checkpoint blockade treatment (22, 23). This immune-related adverse event provides compelling evidence that checkpoint blockade can and does stimulate an immune response readily within the pituitary gland.

PD-L1 expression on pituitary adenomas has been previously characterized, with highest expression found on functional adenomas, although few data on expression by ACTH-secreting tumors are available (6, 24, 25). In addition, pituitary adenomas demonstrate a lymphocytic infiltrate (25), while T cells are otherwise rare in the normal pituitary (23). Given the presence of T-cell infiltrates within the tumor and expression of coinhibitory ligands in the tumor microenvironment, pituitary adenomas may be susceptible to an appropriate checkpoint blockade strategy. We therefore sought to better characterize PD-L1 expression on ACTH-secreting adenomas and determine whether blockade of the PD-1/PD-L1 axis could be utilized to target these tumors and improve outcomes in Cushing disease. In this study, we corroborate the expression of PD-L1 on human pituitary adenomas (including those secreting ACTH), as well as an ACTH-secreting murine adenoma cell line. We employ a novel in vivo murine model for Cushing disease and demonstrate antitumor efficacy using anti–PD-L1 in both subcutaneous and intracranial tumor models.

Clinical studies and specimen processing

Studies were conducted in accordance with the provisions of the Declaration of Helsinki and the Good Clinical Practice guidelines of the Internatinoal Conference on Harmonisation. All studies were performed with approval by the Duke University Institutional Review Board. Patient specimens were collected following appropriate consent. A total of 67 human pituitary tumor tissue specimens were identified from the Duke University surgical pathology archives. These formalin-fixed paraffin-embedded (FFPE) samples were used for IHC staining. For human studies, antibodies to PD-L1 (790-4905, Ventana Medical Systems, Inc), ACTH (ab74976, Abcam), and CD3 (RM-9107-S, Thermo Fisher Scientific) were used. Steps 1-23 of the RUO Discovery universal DISCOVERY ULTRA staining module were followed for anti–PD-L1 Staining (Ventana Medical Systems, Inc) by our pathology department. All samples were reviewed by a senior pathologist (R.E. McLendon).

Mice

Female C57Bl/6 were purchased from Charles River Laboratories. Female C57L/J mice and male A/HeJ were purchased from Jackson Laboratories and bred to create LAF1 hybrid mice with in-house colony expansion breeding. All mice were used at 6–12 weeks of age. Animals were maintained under specific pathogen-free conditions at the Cancer Center Isolation Facility (CCIF) of Duke University Medical Center (Durham, NC). All mouse studies were conducted with approval from the Duke University Insitutitional Animal Care and Use Committee (IUCAC).

Cell lines

Cell lines studied included murine ATT-20/D16v.2 ACTH-secreting pituitary adenoma, SMA-560 malignant glioma, CT-2A malignant glioma, GL261 malignant glioma, KR158B malignant glioma, E0771 breast medullary adenocarcinoma, B16F10 melanoma, and Lewis Lung Carcinoma (LLC). ATT/D16v.2 cells are syngeneic in LAF1 hybrid mice. SMA-560 cells are syngeneic on the VM/Dk mouse background, while all others are syngeneic in C57BL/6 mice. All cell lines have been authenticated by using NIST published 9 species-specific short-tandem repeat markers to establish genetic profiles. Interspecies contamination check for human, mouse, rat, African green monkey, and Chinese hamster was also performed for each cell line. The CellCheck Mouse PlusTM cell line authentication and Mycoplasma spp. testing services were provided by IDEXX Laboratories. All cell lines have been tested negative for Mycoplasma spp. None are among the ICLAC database of commonly misidentified cell lines. ATT/D16v.2 cells were grown in vitro in DMEM (Gibco Thermo Fisher Scientific) with 2 mmol/L l-glutamine, 4.5 mg/mL glucose (Gibco Thermo Fisher Scientific) containing 10% FBS (Gemini Bio-Products) and 1% 200 mmol/L l-glutamine 100× (Gibco Thermo Fisher Scientific). SMA560, CT2A, B16, and LLC cells were grown in vitro in DMEM with 2 mmol/L l-glutamine and 4.5 mg/mL glucose (Gibco Thermo Fisher Scientific) containing 10% FBS (Gemini Bio-Products). E0771 cells were grown in vitro in RPMI1640 (Gibco Thermo Fisher Scientific) containing 10% FBS plus 1% HEPES (Gibco Thermo Fisher Scientific). Cells were harvested in the logarithmic growth phase. For subcutaneous implantation, 107 ATT-20/D16v.2 cells were delivered in a total volume of 200 μL per mouse into the subcutaneous tissues of the left flank of A/HeJ x C57L/J F1 (LAF1) mice. For intracranial (IC) implantation, ATT-20/D16v.2 cells in PBS were mixed 1:1 with 3% methylcellulose and loaded into a 250 μL Syringe (Hamilton). The needle was positioned 2 mm to the right of the bregma and 4 mm below the surface of the skull at the coronal suture using a stereotactic frame. A total of 5 × 104 ATT-20/D16v.2 cells were delivered in a total volume of 5 μL per mouse.

Tissue processing

For tumor growth models, area under the curve (AUC) was estimated using the trapezoidal rule based on data starting from the first day of treatment. AUC was then standardized (sAUC) by number of days followed for each animal to allow for animals that dropped out to be appropriately included in the analysis. For subcutaneous tumor-bearing mice, humane endpoints included tumor size greater than 20 mm in one dimension, 2,000 mm3 in total volume, or tumor ulceration or necrosis. Subcutaneous tumors were harvested, minced, and incubated in 100 U/mL collagenase IV (Sigma-Aldrich) and 0.1 mg/mL DNAse I (Roche Diagnostics) in RPMI supplemented with 10% FBS for 20–30 minutes in a Stomacher machine set at normal speed, and washed through 70 μm nylon Cell Strainers (Falcon; BD Biosciences) in PBS with 2% FBS. Cells were immediately stained and subsequently analyzed by flow cytometry. Murine pituitary glands were identified and removed, processed in RPMI, minced into single-cell suspensions, cell-strained, counted, stained with antibodies, and analyzed via flow cytometry. Blood samples were directly labeled with antibodies and red blood cells subsequently lysed using RBC Lysis Buffer (eBioscience) or BD Pharm Lyse (BD Biosciences).

Flow cytometry

Murine-specific antibodies were purchased from BD Biosciences: (CD3: 145-2C11; PD-L1/CD247: MIH5; TIM-3: RMT3-23; PD-1: J43; LAG-3: C9B7W) and from BioLegend: (CD-45: 30F-11, CD3: 2C11, B220: RA3-6B2, CD11b: M1/70, CD11c: HL3, NK1.1 PK136). Appropriate isotype controls were used when applicable. The LIVE/DEAD Fixable Yellow Dead Cell Stain Kit (Thermo Fisher Scientific) was used to exclude dead cells. Cells were analyzed with an LSRFortessa (BD Biosciences) and data were analyzed with FlowJo Software (TreeStar).

ELISA

In the subcutaneous tumor-bearing model, 107ATT-20/D16v.2 cells were injected subcutaneously into the left flank of LAF1 mice on day 0. At day 1, prior to tumor implantation, mice underwent 200 μL retro-orbital bleed EDTA tubes (MCVT200, Kent Scientific Corporation) with an added 20 μL Aprotinin (RK-APRO, Phoenix Pharmaceuticals, Inc). This blood mixture was then centrifuged at 4°C at 2,000 × g for 15 minutes and aliquots of plasma were stored at −80°C. At day 38, mice underwent an additional 200 μL retro-orbital bleed, which was processed and stored as described previously. ACTH concentrations were measured by Enzyme Immunoassay Kit (Phoenix Pharmaceuticals Inc.). The manufacturer's specifications were followed accordingly to determine plasma ACTH concentrations.

In vivo antibody injection

In the subcutaneous tumor-bearing model, 107 ATT-20/D16v.2 cells were injected subcutaneously into the left flank of LAF1 mice on day 0. Following tumor implantation, mice were treated intraperitoneally (ip) with 200 μg of either anti–PD-L1 (clone 10F.9G2; BioXCell) or isotype control antibody (Rat IgG2b; BioXCell) every 3 days starting on day 3 for 12 total doses until day 36. In our intracranial tumor-bearing model, 5 × 104 D16v.2 cells were injected intracranially at day 0. Beginning on day 3, mice were treated intraperitoneally with either 200 μg of anti–PD-L1 (10F.9G2; BioXCell) or PBS (Gibco Thermo Fisher Scientific) control every 3 days starting on day 3 for 15 total doses until day 45.

Human ACTH-secreting pituitary adenomas express PD-L1 and are infiltrated by T cells

Increased levels of PD-L1 by IHC on various types of human pituitary adenomas have been reported, with expression found to be highest among functional adenomas (6). Levels of PD-L1 on the ACTH-secreting tumors constituting Cushing disease, however, are not well characterized. To assess PD-L1 levels in Cushing disease and compare with other adenoma subtypes, 67 human pituitary tumor surgical specimens were obtained from the Duke University surgical pathology archives for analysis, and IHC was performed on the formalin-fixed paraffin-embedded (FFPE) tissue samples, following deparaffinization. The median age for these patients was 49 (range 16–81 years of age), and 51% were male (34/67; Table 1). Tissue samples were taken from both nonfunctioning (11/67—16%) and functioning (56/67—84%) pituitary adenomas. Hormonal positivity for these samples were 42% (28/67) for ACTH, 30% (20/67) for growth hormone (GH), and 28% (19/67) for prolactin (a number of tumors were polyfunctional). Tumors were newly diagnosed in 75% (50/67) of samples, and recurrent in 25% (17/67) of samples.

Table 1.

Patient demographics for surgical specimens obtained from Duke University surgical pathology archives.

CharacteristicsPatients (n)Percentage
Age (median) 49  
Gender 
 Male 33 49 
 Female 34 51 
Classification 
 Nonfunctioning 11 16.4 
 Functioning 56 83.6 
 PRL 19 28.4 
 GH 20 29.9 
 ACTH 28 41.8 
Recurrence status 
 Recurrent 17 25.3 
 Newly diagnosed 50 74.6 
CharacteristicsPatients (n)Percentage
Age (median) 49  
Gender 
 Male 33 49 
 Female 34 51 
Classification 
 Nonfunctioning 11 16.4 
 Functioning 56 83.6 
 PRL 19 28.4 
 GH 20 29.9 
 ACTH 28 41.8 
Recurrence status 
 Recurrent 17 25.3 
 Newly diagnosed 50 74.6 

A representation of positive staining (defined as >1% cellular reactivity) is shown in Fig. 1A, with placental tissue serving as a positive control for PD-L1. In general, the reactivity pattern was cytoplasmic and clustered in groups of cells. In addition, T-cell infiltration was confirmed, examining the presence of CD3 in human pituitary adenomas; a representative image of CD3 staining by IHC is shown in Fig. 1B. Our results confirmed that expression of PD-L1 was higher on functional, as compared with nonfunctional, pituitary adenomas. Scattered positive immunoreactivity for PD-L1 was found in 22% (15/67) of the tumors overall. None of the nonfunctioning pituitary adenomas expressed PD-L1 (Fig. 1C). Ultimately, 14% (4/28) of ACTH-secreting samples were positive for PD-L1. PD-L1 positivity in GH- and prolactin-secreting samples was 45% (9/20) and 37% (7/19), respectively (Fig. 1D). The proportion of PD-L1–positive tumors trended higher among primary tumors (12/50—24%) than among recurrent adenomas (3/17—17.6%), although the difference was not significant (Fig. 1E).

Figure 1.

PD-L1 and CD3 expression on human ACTH-secreting pituitary adenomas (n = 67). A, Representative IHC staining of human pituitary adenoma samples demonstrating scattered positive immunoreactivity for PD-L1 (right) with placenta serving as positive PD-L1 control (left). PD-L1 positivity was defined as >1% staining. B, Representative IHC staining for CD3 demonstrating a T-cell infiltrate in an ACTH-secreting adenoma. C–E, PD-L1 positivity on: C, Functioning (Func) vs. nonfunctioning (Non-Func) adenomas (P = 0.047; Fisher exact test one-sided); D, GH-producing (GH), prolactin-producing (Pro), ACTH-producing (ACTH), and Non-Func adenomas (descriptive); and E, Primary versus recurrent adenomas (P = 0.4309; Fisher exact test one-sided). F–H, Mean CD3 positivity per high-powered field (HPF) on: F, Functioning (Func) vs. nonfunctioning (Non-Func) adenomas (descriptive); G, GH-producing (GH), prolactin-producing (Pro), ACTH-producing (ACTH), and Non-Func adenomas. H, Primary versus recurrent adenomas.

Figure 1.

PD-L1 and CD3 expression on human ACTH-secreting pituitary adenomas (n = 67). A, Representative IHC staining of human pituitary adenoma samples demonstrating scattered positive immunoreactivity for PD-L1 (right) with placenta serving as positive PD-L1 control (left). PD-L1 positivity was defined as >1% staining. B, Representative IHC staining for CD3 demonstrating a T-cell infiltrate in an ACTH-secreting adenoma. C–E, PD-L1 positivity on: C, Functioning (Func) vs. nonfunctioning (Non-Func) adenomas (P = 0.047; Fisher exact test one-sided); D, GH-producing (GH), prolactin-producing (Pro), ACTH-producing (ACTH), and Non-Func adenomas (descriptive); and E, Primary versus recurrent adenomas (P = 0.4309; Fisher exact test one-sided). F–H, Mean CD3 positivity per high-powered field (HPF) on: F, Functioning (Func) vs. nonfunctioning (Non-Func) adenomas (descriptive); G, GH-producing (GH), prolactin-producing (Pro), ACTH-producing (ACTH), and Non-Func adenomas. H, Primary versus recurrent adenomas.

Close modal

After confirming the presence of CD3+ T cells, we showed that CD3+ counts per HPF were greater in functioning (6.76) compared with nonfunctioning adenomas (1.45; Fig. 1F). Immunoreactivity for CD3 was highest in GH-secreting samples (9.94/HPF), followed by ACTH (4.60/HPF), and prolactin-secreting (4.18/HPF) tumors (Fig. 1G). CD3+ infiltrates were more prevalent in recurrent tumors (9.84/HPF) compared with primary adenomas (3.98/HPF; Fig. 1H).

PD- L1 expression is recapitulated on murine ACTH-secreting pituitary adenoma

PD-L1 expression was assessed in vitro on the murine ACTH-secreting pituitary adenoma cell line (ATT/D16v.2) via flow cytometry (Fig. 2A), and compared with other murine tumor cell lines, including B16 melanoma, E0771 breast adenocarcinoma, Lewis Lung Carcinoma (LLC), GL261 glioma, CT2A glioma, SMA560 glioma, and KR158B glioma (Fig. 2B). The most prominent expression of PD-L1 was observed on B16 melanoma and ATT/D16v.2 adenoma.

Figure 2.

PD-L1 expression on murine ACTH-secreting pituitary adenoma cell line. A, PD-L1 level on the ATT20/D16v2 pituitary adenoma cell line, as analyzed by flow cytometry. B, Comparative PD-L1 expression across various murine tumor cell lines. C, Following subcutaneous implantation and subsequent harvest of ATT20/D16v2, H&E staining confirmed histology consistent with pituitary adenoma. ++ signifies strongly positive; ± signifies weakly positive to negative; − signifies negative. D, Representative IHC staining demonstrates diffuse PD-L1 positivity among harvested adenoma following growth in vivo. E, Expression of PD-L1 on adenoma grown in vivo compared with normal mouse pituitary gland, as analyzed by flow cytometry.

Figure 2.

PD-L1 expression on murine ACTH-secreting pituitary adenoma cell line. A, PD-L1 level on the ATT20/D16v2 pituitary adenoma cell line, as analyzed by flow cytometry. B, Comparative PD-L1 expression across various murine tumor cell lines. C, Following subcutaneous implantation and subsequent harvest of ATT20/D16v2, H&E staining confirmed histology consistent with pituitary adenoma. ++ signifies strongly positive; ± signifies weakly positive to negative; − signifies negative. D, Representative IHC staining demonstrates diffuse PD-L1 positivity among harvested adenoma following growth in vivo. E, Expression of PD-L1 on adenoma grown in vivo compared with normal mouse pituitary gland, as analyzed by flow cytometry.

Close modal

To examine whether PD-L1 levels on ATT/D16v.2 were maintained in vivo, 107 cells were implanted into the left flank of syngeneic LAF1 mice. Subcutaneous (sc) implantation was performed initially to permit sufficient tumor growth for analysis. Tumors were allowed to progress for up to eight weeks and then were harvested for profiling. Hematoxylin and eosin (H&E) staining confirmed histology consistent with pituitary adenoma (Fig. 2C). PD-L1 expression was confirmed on the harvested in vivo tumor by IHC (representative positive staining shown in Fig. 2D). Likewise, PD-L1 levels on harvested tumors were compared by flow cytometry to levels on normal pituitary harvested from host LAF1 mice. PD-L1 expression was present on the harvested adenoma, but absent on the normal pituitary gland (Fig. 2E).

Establishment of a murine model of Cushing disease

To establish a heterotopic murine model of Cushing disease, 107 ACTH-secreting ATT/D16v.2 cells were implanted into the flank of LAF1 mice. The first objective was to identify and confirm ACTH production by the ATT/D16v.2 cell line in vivo. Following subcutaneous growth for approximately 4 weeks, tumor was harvested, and ACTH expression was demonstrated via IHC (Fig. 3A). Serum levels of ACTH were measured by ELISA and compared at baseline and 8 weeks posttumor implantation, with anticipated rises in serum ACTH levels accompanying tumor growth (Fig. 3B). Expected weight gain also accompanied increases in tumor volume. An approximate 30% increase in weight from baseline was observed by week 5 in these mice (Fig. 3C). Gross phenotypic changes related to Cushing syndrome after subcutaneous implantation of the ATT/D16v.2 cell line are depicted in Fig. 3D.

Figure 3.

Establishment of a murine model of Cushing disease. A, Representative IHC staining confirming heterogeneous diffuse immunoreactivity for ACTH among harvested AtT20/D16v2 adenoma following growth in vivo. B, Increased plasma ACTH levels observed in tumor-bearing (TB) mice 8 weeks following subcutaneous (SC) implantation of the AtT20/D16v2 adenoma (n = 5) compared with baseline plasma measurements in naïve mice (n = 4), as measured by ELISA (P = 0.0178; unpaired t test). C, Increased weight gain accompanying increases in tumor volume as observed in TB mice following SC implantation of the AtT20/D16v2 adenoma line. D, Gross phenotypic changes (weight gain, fat deposition) consistent with Cushing disease in mice bearing SC pituitary adenoma tumor.

Figure 3.

Establishment of a murine model of Cushing disease. A, Representative IHC staining confirming heterogeneous diffuse immunoreactivity for ACTH among harvested AtT20/D16v2 adenoma following growth in vivo. B, Increased plasma ACTH levels observed in tumor-bearing (TB) mice 8 weeks following subcutaneous (SC) implantation of the AtT20/D16v2 adenoma (n = 5) compared with baseline plasma measurements in naïve mice (n = 4), as measured by ELISA (P = 0.0178; unpaired t test). C, Increased weight gain accompanying increases in tumor volume as observed in TB mice following SC implantation of the AtT20/D16v2 adenoma line. D, Gross phenotypic changes (weight gain, fat deposition) consistent with Cushing disease in mice bearing SC pituitary adenoma tumor.

Close modal

Characterization of T-cell infiltrates within ACTH-secreting adenoma

The presence of intratumoral, as opposed to peritumoral, T cells was confirmed via IHC (Fig. 4A). We then focused specifically on T cells and further characterized phenotypic markers of exhaustion. TIL exhaustion, marked by the upregulation of costimulatory molecules such as LAG-3, TIM-3, and PD-1, has proven particularly limiting for checkpoint blockade strategies in cancer, as we and others have shown (26, 27). More specifically, an increased expression of a combination of PD-1, Tim-3, and Lag-3 on TIL has been shown to correlate inversely with tumor response to checkpoint blockade (26, 28). We therefore examined the presence of all three markers on TILs and in blood harvested from tumor-bearing and naïve mice. A representative flow cytometry plot is shown in Fig. 4B. The combination of PD-1, TIM-3, and LAG-3 were present to a greater extent in TIL than in the blood of either tumor-bearing or naïve mice (Fig. 4C). PD-1, TIM-3, and LAG-3 were present in modest levels on TILs in ATT/D16v2 and resemble levels of immunogenic tumors (B16 melanoma, LLC). A heatmap comparing levels of each exhaustion marker against TIL from several tumor models is shown in Fig. 4D.

Figure 4.

Characterization of tumor–immune infiltrates. A, Representative IHC staining of CD3+ cells in ATT20/D16v2 adenoma. B, Representative flow cytometry plot of immune checkpoints on adenoma TIL. C, Quantified levels of PD-1, Tim-3, and Lag-3 on T cells taken from naïve mouse blood (n = 3), tumor-bearing (TB) mouse blood (n = 5), or TILs infiltrating subcutaneous (SC) ATT20/D16v2 adenomas (n = 5), as assessed by flow cytometry. D, Heatmap comparing the same checkpoints on CD3+ TILs taken from SC AtT20/D16v2 adenomas, CT2A gliomas, LLC, or B16 melanoma (n = 4/group). Each column represents an individual mouse.

Figure 4.

Characterization of tumor–immune infiltrates. A, Representative IHC staining of CD3+ cells in ATT20/D16v2 adenoma. B, Representative flow cytometry plot of immune checkpoints on adenoma TIL. C, Quantified levels of PD-1, Tim-3, and Lag-3 on T cells taken from naïve mouse blood (n = 3), tumor-bearing (TB) mouse blood (n = 5), or TILs infiltrating subcutaneous (SC) ATT20/D16v2 adenomas (n = 5), as assessed by flow cytometry. D, Heatmap comparing the same checkpoints on CD3+ TILs taken from SC AtT20/D16v2 adenomas, CT2A gliomas, LLC, or B16 melanoma (n = 4/group). Each column represents an individual mouse.

Close modal

Treatment with anti–PD-L1 restricts tumor growth and ACTH production in a model of Cushing disease

Encouraged by our TIL characterization, we used the heterotopic murine model of Cushing disease established above to evaluate the antitumor efficacy of treatment with the anti–PD-L1 antibody (10F.9G2; BioXCell). Initial use of a heterotopic flank tumor model permitted assessment of treatment effect on tumor volumes, specifically. Following subcutaneous tumor implantation, mice were treated intraperitoneally with either anti–PD-L1 or isotype control antibody (Rat IgG2b; BioXCell) every 3 days starting on day 3 for 12 total doses (through day 36). Treatment with anti–PD-L1 resulted in significant inhibition of tumor growth (Fig. 5A), and likewise precipitated a decrease in levels of serum ACTH at 8 weeks posttreatment compared with untreated tumor-bearing mice (Fig. 5B). A subset of mice achieved complete tumor regressions.

Figure 5.

Treatment with anti–PD-L1 prevents tumor growth and ACTH production in a heterotopic Cushing disease model. A, Following subcutaneous (SC) implantation of AtT20/D16v2 adenomas, treatment with anti–PD-L1 (n = 8) significantly restricted tumor growth compared with treatment with isotype control (n = 16; P = 0.0002; exact Wilcoxon of sAUC). B, Serum ACTH levels measured by ELISA demonstrate decreased ACTH levels following anti–PD-L1 treatment (P < 0.0001; one-way ANOVA; Naïve n = 20; TB + control n = 14; TB + anti-PDL1 n = 7). Results are pooled across multiple experiments. C, Nude (athymic) mice implanted SC with ATT20/D16v2 tumors also develop a Cushingoid phenotype. D, When treated with anti-PDL1 antibody, nude tumor-bearing mice (n = 7) demonstrate a more modest restriction of tumor growth compared with controls (n = 8; P = 0.0401; exact Wilcoxon of sAUC).

Figure 5.

Treatment with anti–PD-L1 prevents tumor growth and ACTH production in a heterotopic Cushing disease model. A, Following subcutaneous (SC) implantation of AtT20/D16v2 adenomas, treatment with anti–PD-L1 (n = 8) significantly restricted tumor growth compared with treatment with isotype control (n = 16; P = 0.0002; exact Wilcoxon of sAUC). B, Serum ACTH levels measured by ELISA demonstrate decreased ACTH levels following anti–PD-L1 treatment (P < 0.0001; one-way ANOVA; Naïve n = 20; TB + control n = 14; TB + anti-PDL1 n = 7). Results are pooled across multiple experiments. C, Nude (athymic) mice implanted SC with ATT20/D16v2 tumors also develop a Cushingoid phenotype. D, When treated with anti-PDL1 antibody, nude tumor-bearing mice (n = 7) demonstrate a more modest restriction of tumor growth compared with controls (n = 8; P = 0.0401; exact Wilcoxon of sAUC).

Close modal

To evaluate whether this effect was a T-cell–dependent phenomenon, subcutaneous tumors were also implanted into nude (athymic) mice. Mice were then treated intraperitoneally with either anti–PD-L1 or isotype control antibody over the same time course. Tumor-bearing nude mice likewise developed a Cushingoid phenotype, marked by weight gain and increased fat deposits (Fig. 5C). Treatment with anti–PD-L1 still produced a modest restriction to tumor growth, (Fig. 5D) suggesting that although present and contributory, T cells were not necessary for efficacy.

Anti–PD-L1 retains efficacy in the intracranial environment

We next examined whether anti–PD-L1′s capacity to restrict tumor growth would be maintained when tumors were implanted intracranially. For reasons of practicality, tumors were placed in the right frontal lobe rather than within the sella. Stereotactic introduction of 5 × 104 ATT/D16v.2 cells to the base of the right frontal lobe led to unincumbered intracranial tumor growth, resulting in 100% mortality after approximately 3 weeks. This proved a fairly stringent, aggressive model, as it was more lethal than the human condition. Mice with intracranial ATT/D16v.2 were then administered anti–PD-L1 or control antibody every 3 days, beginning on day 3 following tumor implantation. The median overall survival among the treatment cohort was extended to 29.5 days (95% CI: 15, not estimable), compared with a median survival of 21.5 days (95% CI, 15.0–24.0) in the control group. In addition, treatment with anti–PD-L1 enhanced long-term survival, as the 30-day survival estimate in the treated group was 50% (95% CI, 18.4%–75.3%) and 60-day survival estimate was 40% (95% CI, 12.3%–67%) while in the control group, survival was 0% at both time points (Fig. 6A).

Figure 6.

PD-L1 is expressed in the tumor microenvironment and treatment with anti–PD-L1 promotes survival in an IC model of pituitary adenoma. A, Survival following IC implantation of ATT20/D16v2 adenomas into immunocompetent mice; treatment with anti–PD-L1 or isotype control. B, Characterization of CD45+ immune cells in the tumor microenvironment. C, PD-L1 MFI on tumor cells and myeloid cells in the IC tumor microenvironment. D, Representative flow cytometry of PD-L1 expression on tumor cells (top: CD45; forward-scatterhi) and myeloid cells (bottom: CD45+; CD11b+).

Figure 6.

PD-L1 is expressed in the tumor microenvironment and treatment with anti–PD-L1 promotes survival in an IC model of pituitary adenoma. A, Survival following IC implantation of ATT20/D16v2 adenomas into immunocompetent mice; treatment with anti–PD-L1 or isotype control. B, Characterization of CD45+ immune cells in the tumor microenvironment. C, PD-L1 MFI on tumor cells and myeloid cells in the IC tumor microenvironment. D, Representative flow cytometry of PD-L1 expression on tumor cells (top: CD45; forward-scatterhi) and myeloid cells (bottom: CD45+; CD11b+).

Close modal

Given our earlier observation in nude mice that T cells were dispensible to treatment efficacy (Fig. 5D), we opted to characterize the immune infiltrates within the more relevant intracranial tumor microenvironment, to better determine which population(s) might be mediating the therapeutic effect. The vast majority of tumor-infiltrating CD45+ cells proved to be myeloid cells expressing the CD11b marker, although T-cell infiltrates were also present as expected (Fig. 6B). Furthermore, when examined by flow cytometry, myeloid cells demonstrated particularly high levels of PD-L1 expression, even when compared with tumor cells themselves. (Fig. 6C, representative flow cytometry plot depicted in Fig. 6D).

Over the past decade, the emerging success of immunotherapeutic strategies has altered the approach to treating various cancers. Among such strategies, immune checkpoint blockade has perhaps made the greatest gains, with targeting of the PD-1/PD-L1 axis evolving as the frequently preferred approach. Clinical trials with anti–PD-1/PD-L1 have shown promising results against a variety of malignancies, including non–small cell lung cancer, renal cell, melanoma, ovarian, and bladder cancer (17, 21, 29–32). While checkpoint blockade is now FDA-approved for various tumors, this study is the first to systematically explore its employment against pituitary adenomas.

Prior studies have demonstrated PD-L1 expression on human pituitary adenomas (6, 24). Mei and colleagues identified PD-L1 at both the transcript and protein level across 48 human pituitary adenomas. They found that PD-L1 expression was higher in functioning tumors, in line with our own results. Wang and colleagues assessed 191 human pituitary adenomas by IHC, and found that PD-L1 stained positively among 36.6% of samples. They also found that positivity was more frequent among functioning adenomas, with positive staining in 58.8% of functioning tumors compared with 34.3% of nonfunctioning adenomas. This group observed that PD-L1 expression was significantly correlated with expression of GH and prolactin, but not ACTH, although 17 of 28 ACTH-producing tumors did express PD-L1 (25).

In concordance with these studies, we found PD-L1 expression to be highest among functional pituitary adenomas, with the highest expression in GH-producing adenomas, followed by those producing prolactin and ACTH. Subtler differences regarding the proportions staining positively in each study might be explained by technical discrepancies, sample sizes employed, and defined criteria for positivity.

PD-L1 expression is often viewed as a potential biomarker for response to checkpoint blockade. In some cancers, it has been shown to correlate positively with treatment response to the PD-1 blockade (i.e., nivolumab; refs. 29–33), while other studies have not found the same correlation (31). Given the PD-L1 expression we observed on ACTH-secreting adenomas, however, we chose to begin with direct targeting of PD-L1, avoiding some of this controversy and simplifying our approach. We were further encouraged by the detection of a lymphocytic infiltrate in pituitary adenomas (including our own), suggesting the presence of an antitumor effector arm that might be licensed with anti–PD-L1 therapy (6, 25). Given the mechanism of action for checkpoint blockade therapy, the minimum requirement is the presence of both ligand and receptor for whichever inhibitory axis is targeted (e.g., PD-1 and PD-L1 present in high levels when treating with nivolumab), as well as T-cell infiltrates within the tumor. Therefore, immunogenic “hot” tumors with a high degree of T-cell infiltration and expression of PD-L1 on the tumor cell surface are most capable of responding to checkpoint blockade therapy. This is juxtaposed with immunologically “cold” tumors that are resistant to immunotherapy based on their dearth of immune infiltrates. Our findings suggest that ACTH-secreting pituitary adenomas may respond to checkpoint blockade therapy targeting the PD-1/PD-L1 axis given the presence of CD3+ TILs and expression of PD-L1 on the tumor.

Prior groups have also created Cushing disease murine models (34, 35). For our use, we employed an ACTH-secreting pituitary adenoma cell line (ATT-20/D16v2), which prior studies have confirmed maintains properties of anterior pituitary cells (36, 37). We passaged this line in vivo to select for a penetrant subclone and confirmed surface expression of PD-L1 both in vitro and in vivo. Four weeks after subcutaneous tumor implantation, average body weight had increased by more than 20%, and mice had developed a Cushingoid phenotype that included increased subcutaneous fat deposits and marked obesity. These changes were accompanied by a 3-fold increase in ACTH concentration in the plasma. Treating these mice with a PD-L1–blocking antibody significantly slowed tumor growth and led to a decrease in serum ACTH levels. Importantly, our findings also demonstrated an antitumor effect of PD-L1-blockade that extended intracranially, providing context for translation.

Our study is the first to demonstrate efficacy for checkpoint blockade against pituitary adenoma in a preclinical model. Clinical use for adenoma has also not been reported. There has been a single case report in which a patient with resection- and chemotherapy-refractory pituitary carcinoma was treated with a combination of ipilimumab and nivolumab. This patient achieved a radiographic response in both her sellar tumor as well as among extracranial metastases, suggesting the intracranial compartment and pituitary gland are not barriers to effective checkpoint blockade (38).

The use of checkpoint inhibitors against intracranially situated lesions continues to be an area of investigation. While therapies targeting PD-1 have failed in clinical trials for glioblastoma, (39) there have been reported successes against intracranial metastases (40–43). Recently, a multicenter phase II study of ipilimumab + nivolumab for intracranial metastatic melanoma demonstrated intracranial response rates of 57%, comparable with the response rate seen extracranially (21). This suggests that the primary obstacle to checkpoint blockade is unlikely to be access to the intracranial compartment. There is indeed evidence that checkpoint blockade antibodies penetrate the blood–brain barrier (BBB; ref. 44), although it is less understood that BBB and tumor penetration are requisites for success. Certainly, activated T cells cross the BBB, and this may or may not be the more crucial component to therapeutic response. Regardless, portions of the pituitary are situated outside the BBB, perhaps obviating the issue of drug access for these tumors specifically.

It is salient to note that lymphocytic hypophysitis, T-cell–based inflammation within the pituitary gland, emerges as a common side effect of treatment with checkpoint blockade in clinical trials for other cancers (22, 23). It is most common with anti–CTLA-4 (ipilimumab), but has also been shown with anti–PD-1 treatment (nivolumab; ref. 45). While problematic, this provides compelling evidence that checkpoint blockade can and does stimulate an immune response readily within the pituitary gland, perhaps obviating concerns to the contrary. Our own data showing 40% long-term survival in an intracranial Cushing disease model treated with anti–PD-L1 provide further optimism for this approach despite the intracranial locale.

The efficacy of checkpoint blockade against intracranial tumors, and cancer more generally, has been variably restricted by tumor-imposed T-cell dysfunction among TIL. Exhaustion, marked frequently by the upregulation of alternative immune checkpoints, has proved particularly limiting (26–28, 46). Our group has shown, for instance, that gliomas exhibit particularly severe patterns of TIL exhaustion, whether in orthotopic or heterotopic models, and likewise prove resistant to checkpoint monoblockade therapy. Conversely, checkpoint-sensitive tumors such as melanoma and lung cancers demonstrate milder TIL exhaustion signatures, as expression of PD-1, TIM-3, and LAG-3 function to modulate T-cell activation at homeostasis, and activated T cells are necessary for an antitumor response (26).

A specific mode of T-cell dysfunction arising in the setting of chronic, suboptimal antigen exposure, exhaustion is a transcriptional program often manifested by the expression of multiple coinhibitory molecules such as PD-1, TIM3, and LAG3 on the T-cell surface (26, 47). In our study, pituitary adenoma TILs expressed these molecules, but at modest levels typical of immunogenic tumors, such as melanoma and lung carcinoma, which are responsive to checkpoint blockade therapy. Very few TILs expressed multiple coinhibitory molecules, which would have indicated more severe exhaustion and typified by therapy-resistant tumors such as glioma (26).

The responsiveness to PD-L1 therapy in our studies, however, may be mediated through multiple immune cell types, as modest treatment responses were still observed in athymic mice lacking T cells. In keeping with a study by Lu and colleagues, we observed that the myeloid compartment constituted the largest fraction of the pituitary adenoma microenvironment (48). Furthermore, myeloid cells demonstrated high PD-L1 expression. This likewise raises the question as to whether PD-L1 expression in the tumor or its microenvironment is more relevant to therapeutic success. Such questions are likely to be best answered by clinical studies in patients, as the single cell line employed here creates little variability in tumor PD-L1 expression across mice.

One limitation of our study is the implantation of the murine ATT20/D16v.2 cell line in the right hemisphere given the technical restraints of direct sellar implantation. This method is imperfect as a true anatomical recapitulation of Cushing disease. Furthermore, as space in the intracranial compartment is limited, tumor growth is likewise limited. Mice succumb to tumor mass effect and systemic cachexia before tumors become large enough to secrete sufficient hormone to cause the Cushingoid phenotype, making it an imperfect model for patients with Cushing disease. However, the intracranial positioning allowed for demonstration of therapeutic efficacy against a tumor situated behind the BBB. Furthermore, these tumors in the IC compartment proved 100% lethal, which although not representative of their human counterparts, provided a seemingly more aggressive, difficult-to-treat model. Coupled with the experiments showing efficacy against subcutaneous tumors (which are not size-limited) and decreases to the associated ACTH levels, these experiments provide proof of concept for using anti–PD-L1 to treat the ACTH-driven morbidity that occurs in patients with Cushing disease from a pituitary tumor.

Future directions, then, include initiation of clinical trials targeting the PD-1/PD-L1 axis in patients with recurrent or treatment-refractory functional pituitary adenomas. Although we model Cushing disease in these studies, high PD-L1 expressors such as GH-secreting tumors may also be viable targets for therapy. Likewise, as the importance of PD-L1 for dictating responses to treatment remains indeterminate, clinical studies may ultimately reveal a role in nonfunctioning tumors where complete eradication is less a goal than mitigating mass effect. While all recurrent/refractory pituitary adenomas present challenges, however, the greatest area of need does perhaps surround ACTH-producing tumors. Because of the endocrinopathies associated with Cushing disease, patients are at increased risk of cardiovascular complications, metabolic abnormalities, and psychiatric manifestations, all of which may lead to devastating morbidity and higher mortality (49, 50). The lack of viable treatment options poses a great need for novel therapies, and our study suggests checkpoint blockade may be a promising strategy. Further preclinical study can be aimed at investigating other checkpoint blockade modalities such as anti–PD-1, anti–CTLA-4, or the combination.

No potential conflicts of interest were disclosed.

Conception and design: H.R. Kemeny, A.A. Elsamadicy, S.H. Farber, V. Tsvankin, R.E. McLendon, P.E. Fecci

Development of methodology: H.R. Kemeny, A.A. Elsamadicy, S.H. Farber, V. Tsvankin, R.E. McLendon, P.E. Fecci

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): H.R. Kemeny, A.A. Elsamadicy, S.H. Farber, C.D. Champion, P. Chongsathidkiet, K.I. Woroniecka, S.H. Shen, K.E. Rhodin, J. Everitt, R.E. McLendon

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H.R. Kemeny, A.A. Elsamadicy, S.H. Farber, C.D. Champion, S.J. Lorrey, P. Chongsathidkiet, S.H. Shen, K.E. Rhodin, V. Tsvankin, L. Sanchez-Perez, P. Healy, R.E. McLendon, P.J. Codd

Writing, review, and/or revision of the manuscript: H.R. Kemeny, S.H. Farber, C.D. Champion, S.J. Lorrey, P. Chongsathidkiet, S.H. Shen, K.E. Rhodin, V. Tsvankin, P. Healy, R.E. McLendon, P.J. Codd, I.F. Dunn, P.E. Fecci

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H.R. Kemeny, S.J. Lorrey, X. Cui

Study supervision: V. Tsvankin, L. Sanchez-Perez, P.E. Fecci

This research was supported by a Duke Health Scholars Award (to P.E. Fecci).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Asa
SL
,
Ezzat
S
. 
The pathogenesis of pituitary tumors
.
Annu Rev Pathol
2009
;
4
:
97
126
.
2.
Di Ieva
A
,
Rotondo
F
,
Syro
LV
,
Cusimano
MD
,
Kovacs
K
. 
Aggressive pituitary adenomas–diagnosis and emerging treatments
.
Nat Rev Endocrinol
2014
;
10
:
423
35
.
3.
Pivonello
R
,
Isidori
AM
,
De Martino
MC
,
Newell-Price
J
,
Biller
BM
,
Colao
A
. 
Complications of Cushing
' 
s syndrome: state of the art
.
Lancet Diabetes Endocrinol
2016
;
4
:
611
29
.
4.
Molitch
ME
. 
Diagnosis and treatment of pituitary adenomas: a review
.
JAMA
2017
;
317
:
516
24
.
5.
Patil
CG
,
Prevedello
DM
,
Lad
SP
,
Vance
ML
,
Thorner
MO
,
Katznelson
L
, et al
Late recurrences of Cushing
' 
s disease after initial successful transsphenoidal surgery
.
J Clin Endocrinol Metab
2008
;
93
:
358
62
.
6.
Mei
Y
,
Bi
WL
,
Greenwald
NF
,
Du
Z
,
Agar
NY
,
Kaiser
UB
, et al
Increased expression of programmed death ligand 1 (PD-L1) in human pituitary tumors
.
Oncotarget
2016
;
7
:
76565
76
.
7.
Bertagna
X
,
Guignat
L
. 
Approach to the Cushing
' 
s disease patient with persistent/recurrent hypercortisolism after pituitary surgery
.
J Clin Endocrinol Metab
2013
;
98
:
1307
18
.
8.
Feelders
RA
,
Hofland
LJ
. 
Medical treatment of Cushing
' 
s disease
.
J Clin Endocrinol Metab
2013
;
98
:
425
38
.
9.
Pardoll
DM
. 
The blockade of immune checkpoints in cancer immunotherapy
.
Nat Rev Cancer
2012
;
12
:
252
64
.
10.
Hodi
FS
,
O'Day
SJ
,
McDermott
DF
,
Weber
RW
,
Sosman
JA
,
Haanen
JB
, et al
Improved survival with ipilimumab in patients with metastatic melanoma
.
N Engl J Med
2010
;
363
:
711
23
.
11.
Wolchok
JD
,
Chiarion-Sileni
V
,
Gonzalez
R
,
Rutkowski
P
,
Grob
JJ
,
Cowey
CL
, et al
Overall survival with combined nivolumab and ipilimumab in advanced melanoma
.
N Engl J Med
2017
;
377
:
1345
56
.
12.
Callahan
MK
,
Kluger
H
,
Postow
MA
,
Segal
NH
,
Lesokhin
A
,
Atkins
MB
, et al
Nivolumab plus ipilimumab in patients with advanced melanoma: updated survival, response, and safety data in a phase I dose-escalation study
.
J Clin Oncol
2018
;
36
:
391
8
.
13.
Lynch
TJ
,
Bondarenko
I
,
Luft
A
,
Serwatowski
P
,
Barlesi
F
,
Chacko
R
, et al
Ipilimumab in combination with paclitaxel and carboplatin as first-line treatment in stage IIIB/IV non-small-cell lung cancer: results from a randomized, double-blind, multicenter phase II study
.
J Clin Oncol
2012
;
30
:
2046
54
.
14.
Hellmann
MD
,
Rizvi
NA
,
Goldman
JW
,
Gettinger
SN
,
Borghaei
H
,
Brahmer
JR
, et al
Nivolumab plus ipilimumab as first-line treatment for advanced non-small-cell lung cancer (CheckMate 012): results of an open-label, phase 1, multicohort study
.
Lancet Oncol
2017
;
18
:
31
41
.
15.
McDermott
DF
,
Sosman
JA
,
Sznol
M
,
Massard
C
,
Gordon
MS
,
Hamid
O
, et al
Atezolizumab, an anti-programmed death-ligand 1 antibody, in metastatic renal cell carcinoma: long-term safety, clinical activity, and immune correlates from a phase Ia study
.
J Clin Oncol
2016
;
34
:
833
42
.
16.
Motzer
RJ
,
Tannir
NM
,
McDermott
DF
,
Aren Frontera
O
,
Melichar
B
,
Choueiri
TK
, et al
Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma
.
N Engl J Med
2018
;
378
:
1277
90
.
17.
Brahmer
JR
,
Tykodi
SS
,
Chow
LQ
,
Hwu
WJ
,
Topalian
SL
,
Hwu
P
, et al
Safety and activity of anti-PD-L1 antibody in patients with advanced cancer
.
N Engl J Med
2012
;
366
:
2455
65
.
18.
Tsushima
F
,
Yao
S
,
Shin
T
,
Flies
A
,
Flies
S
,
Xu
H
, et al
Interaction between B7-H1 and PD-1 determines initiation and reversal of T-cell anergy
.
Blood
2007
;
110
:
180
5
.
19.
Butte
MJ
,
Keir
ME
,
Phamduy
TB
,
Sharpe
AH
,
Freeman
GJ
. 
Programmed death-1 ligand 1 interacts specifically with the B7-1 costimulatory molecule to inhibit T cell responses
.
Immunity
2007
;
27
:
111
22
.
20.
Taube
JM
,
Anders
RA
,
Young
GD
,
Xu
H
,
Sharma
R
,
McMiller
TL
, et al
Colocalization of inflammatory response with B7-h1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape
.
Sci Transl Med
2012
;
4
:
127ra37
.
21.
Tawbi
HA
,
Forsyth
PA
,
Algazi
A
,
Hamid
O
,
Hodi
FS
,
Moschos
SJ
, et al
Combined nivolumab and ipilimumab in melanoma metastatic to the brain
.
N Engl J Med
2018
;
379
:
722
30
.
22.
Caturegli
P
,
Di Dalmazi
G
,
Lombardi
M
,
Grosso
F
,
Larman
HB
,
Larman
T
, et al
Hypophysitis secondary to cytotoxic T-lymphocyte-associated protein 4 blockade: insights into pathogenesis from an autopsy series
.
Am J Pathol
2016
;
186
:
3225
35
.
23.
Iwama
S
,
De Remigis
A
,
Callahan
MK
,
Slovin
SF
,
Wolchok
JD
,
Caturegli
P
. 
Pituitary expression of CTLA-4 mediates hypophysitis secondary to administration of CTLA-4 blocking antibody
.
Sci Transl Med
2014
;
6
:
230ra45
.
24.
Wang
P
,
Wang
T
,
Yang
Y
,
Yu
C
,
Liu
N
,
Yan
C
. 
Detection of programmed death ligand 1 protein and CD8+ lymphocyte infiltration in plurihormonal pituitary adenomas: a case report and review of the literatures
.
Medicine
2017
;
96
:
e9056
.
25.
Wang
PF
,
Wang
TJ
,
Yang
YK
,
Yao
K
,
Li
Z
,
Li
YM
, et al
The expression profile of PD-L1 and CD8(+) lymphocyte in pituitary adenomas indicating for immunotherapy
.
J Neurooncol
2018
;
139
:
89
95
.
26.
Woroniecka
K
,
Chongsathidkiet
P
,
Rhodin
KE
,
Kemeny
HR
,
Dechant
CA
,
Farber
SH
, et al
T cell exhaustion signatures vary with tumor type and are severe in glioblastoma
.
Clin Cancer Res
2018
;
24
:
4175
86
.
27.
Koyama
S
,
Akbay
EA
,
Li
YY
,
Herter-Sprie
GS
,
Buczkowski
KA
,
Richards
WG
, et al
Adaptive resistance to therapeutic PD-1 blockade is associated with upregulation of alternative immune checkpoints
.
Nat Commun
2016
;
7
:
10501
.
28.
Woroniecka
KI
,
Rhodin
KE
,
Chongsathidkiet
P
,
Keith
KA
,
Fecci
PE
. 
T-cell dysfunction in glioblastoma: applying a new framework
.
Clin Cancer Res
2018
;
24
:
3792
802
.
29.
Topalian
SL
,
Sznol
M
,
McDermott
DF
,
Kluger
HM
,
Carvajal
RD
,
Sharfman
WH
, et al
Survival, durable tumor remission, and long-term safety in patients with advanced melanoma receiving nivolumab
.
J Clin Oncol
2014
;
32
:
1020
30
.
30.
Powles
T
,
Eder
JP
,
Fine
GD
,
Braiteh
FS
,
Loriot
Y
,
Cruz
C
, et al
MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer
.
Nature
2014
;
515
:
558
62
.
31.
Gandini
S
,
Massi
D
,
Mandala
M
. 
PD-L1 expression in cancer patients receiving anti PD-1/PD-L1 antibodies: a systematic review and meta-analysis
.
Crit Rev Oncol Hematol
2016
;
100
:
88
98
.
32.
Taube
JM
,
Klein
A
,
Brahmer
JR
,
Xu
H
,
Pan
X
,
Kim
JH
, et al
Association of PD-1, PD-1 ligands, and other features of the tumor immune microenvironment with response to anti-PD-1 therapy
.
Clin Cancer Res
2014
;
20
:
5064
74
.
33.
Sabatier
R
,
Finetti
P
,
Mamessier
E
,
Adelaide
J
,
Chaffanet
M
,
Ali
HR
, et al
Prognostic and predictive value of PDL1 expression in breast cancer
.
Oncotarget
2015
;
6
:
5449
64
.
34.
Helseth
A
,
Siegal
GP
,
Haug
E
,
Bautch
VL
. 
Transgenic mice that develop pituitary tumors. A model for Cushing disease
.
Am J Pathol
1992
;
140
:
1071
80
.
35.
Furth
J
,
Gadsen
EL
,
Upton
AC
. 
ACTH secreting transplantable pituitary tumors
.
Proc Soc Exp Biol Med
1953
;
84
:
253
4
.
36.
Woods
MD
,
Shipston
MJ
,
Mullens
EL
,
Antoni
FA
. 
Pituitary corticotrope tumor (AtT20) cells as a model system for the study of early inhibition by glucocorticoids
.
Endocrinology
1992
;
131
:
2873
80
.
37.
Shipston
MJ
,
Antoni
FA
. 
Early glucocorticoid induction of calmodulin and its suppression by corticotropin-releasing factor in pituitary corticotrope tumor (AtT20) cells
.
Biochem Biophys Res Commun
1992
;
189
:
1382
8
.
38.
Lin
AL
,
Jonsson
P
,
Tabar
V
,
Yang
TJ
,
Cuaron
J
,
Beal
K
, et al
Marked response of a hypermutated adrenocorticotropic hormone-secreting pituitary carcinoma to ipilimumab and nivolumab
.
J Clin Endocrinol Metab
2018
;
103
:
3925
30
.
39.
Reardon
DA
,
Omuro
A
,
Brandes
AA
,
Rieger
J
,
Wick
A
,
Sepulveda
J
, et al
OS10.3 randomized phase 3 study evaluating the efficacy and safety of nivolumab vs. bevacizumab in patients with recurrent glioblastoma: CheckMate 143
.
Neuro Oncol
2017
;
19
:
iii21
.
40.
Nguyen
SM
,
Castrellon
A
,
Vaidis
O
,
Johnson
AE
. 
Stereotactic radiosurgery and ipilimumab versus stereotactic radiosurgery alone in melanoma brain metastases
.
Cureus
2017
;
9
:
e1511
.
41.
Kaidar-Person
O
,
Zagar
TM
,
Deal
A
,
Moschos
SJ
,
Ewend
MG
,
Sasaki-Adams
D
, et al
The incidence of radiation necrosis following stereotactic radiotherapy for melanoma brain metastases: the potential impact of immunotherapy
.
Anticancer Drugs
2017
;
28
:
669
75
.
42.
Long
GV
,
Atkinson
V
,
Lo
S
,
Sandhu
S
,
Guminski
AD
,
Brown
MP
, et al
Combination nivolumab and ipilimumab or nivolumab alone in melanoma brain metastases: a multicentre randomised phase 2 study
.
Lancet Oncol
2018
;
19
:
672
81
.
43.
Tawbi
HA-H
,
Forsyth
PAJ
,
Algazi
AP
,
Hamid
O
,
Hodi
FS
,
Moschos
SJ
, et al
Efficacy and safety of nivolumab (NIVO) plus ipilimumab (IPI) in patients with melanoma (MEL) metastatic to the brain: results of the phase II study CheckMate 204
.
J Clin Oncol
2017
;
35
:
9507
.
44.
Seyedin
SN
,
Schoenhals
JE
,
Lee
DA
,
Cortez
MA
,
Wang
X
,
Niknam
S
, et al
Strategies for combining immunotherapy with radiation for anticancer therapy
.
Immunotherapy
2015
;
7
:
967
80
.
45.
Okano
Y
,
Satoh
T
,
Horiguchi
K
,
Toyoda
M
,
Osaki
A
,
Matsumoto
S
, et al
Nivolumab-induced hypophysitis in a patient with advanced malignant melanoma
.
Endocr J
2016
;
63
:
905
12
.
46.
Koyama
S
,
Akbay
EA
,
Li
YY
,
Aref
AR
,
Skoulidis
F
,
Herter-Sprie
GS
, et al
STK11/LKB1 deficiency promotes neutrophil recruitment and proinflammatory cytokine production to suppress T-cell activity in the lung tumor microenvironment
.
Cancer Res
2016
;
76
:
999
1008
.
47.
Wherry
EJ
,
Ha
SJ
,
Kaech
SM
,
Haining
WN
,
Sarkar
S
,
Kalia
V
, et al
Molecular signature of CD8+ T cell exhaustion during chronic viral infection
.
Immunity
2007
;
27
:
670
84
.
48.
Lu
JQ
,
Adam
B
,
Jack
AS
,
Lam
A
,
Broad
RW
,
Chik
CL
. 
Immune cell infiltrates in pituitary adenomas: more macrophages in larger adenomas and more T cells in growth hormone adenomas
.
Endocr Pathol
2015
;
26
:
263
72
.
49.
Buliman
A
,
Tataranu
LG
,
Paun
DL
,
Mirica
A
,
Dumitrache
C
. 
Cushing disease: a multidisciplinary overview of the clinical features, diagnosis, and treatment
.
J Med Life
2016
;
9
:
12
8
.
50.
Lacroix
A
,
Feelders
RA
,
Stratakis
CA
,
Nieman
LK
. 
Cushing syndrome
.
Lancet
2015
;
386
:
913
27
.