We explored the association between liver metastases, tumor CD8+ T-cell count, and response in patients with melanoma or lung cancer treated with the anti-PD-1 antibody, pembrolizumab. The melanoma discovery cohort was drawn from the phase I Keynote 001 trial, whereas the melanoma validation cohort was drawn from Keynote 002, 006, and EAP trials and the non–small cell lung cancer (NSCLC) cohort from Keynote 001. Liver metastasis was associated with reduced response and shortened progression-free survival [PFS; objective response rate (ORR), 30.6%; median PFS, 5.1 months] compared with patients without liver metastasis (ORR, 56.3%; median PFS, 20.1 months) P ≤ 0.0001, and confirmed in the validation cohort (P = 0.0006). The presence of liver metastasis significantly increased the likelihood of progression (OR, 1.852; P < 0.0001). In a subset of biopsied patients (n = 62), liver metastasis was associated with reduced CD8+ T-cell density at the invasive tumor margin (liver metastasis+ group, n = 547 ± 164.8; liver metastasis group, n = 1,441 ± 250.7; P < 0.016). A reduced response rate and shortened PFS was also observed in NSCLC patients with liver metastasis [median PFS, 1.8 months; 95% confidence interval (CI), 1.4–2.0], compared with those without liver metastasis (n = 119, median PFS, 4.0 months; 95% CI, 2.1–5.1), P = 0.0094. Thus, liver metastatic patients with melanoma or NSCLC that had been treated with pembrolizumab were associated with reduced responses and PFS, and liver metastases were associated with reduced marginal CD8+ T-cell infiltration, providing a potential mechanism for this outcome. Cancer Immunol Res; 5(5); 417–24. ©2017 AACR.

Antibodies that block binding between programmed death 1 (PD-1) and its ligands, PD-L1 or PD-L2, have shown marked clinical activity in many malignancies, including metastatic melanoma (1–7), non–small cell lung cancer (NSCLC; refs. 8–11), and other cancers (12). The diversity of different cancers in which PD-1/PD-L1–directed therapies have shown efficacy has emphasized that the biological importance of PD-1 on activated, tumor-associated T cells (13–15) transcends histologic subtype. However, specific inter- and even intrapatient features define the distinct nature of a given tumor's immune microenvironment that can modulate the likelihood of benefit from PD-1/PD-L1 blockade.

The presence of a T-cell infiltrate and PD-L1 expression on tumor and tumor stroma represents a stratification factor that has shown predictive value in various cancer types (4, 16, 17). It has been noted, though, that PD-L1 expression is only modestly predictive of response. Tumor CD8+ T-cell infiltration at the invasive margin has been shown to be predictive of response in melanoma (18). Less attention has been paid to the clinical variables that may impact responsiveness to PD-1/PD-L1 blockade and may provide insight into characteristics of both host and tumor that ultimately shape the tumor microenvironment. One criticism of efforts in predictive modeling for immunotherapy focused on single-assay biomarkers, such as PD-L1 expression, has been the lack of integration of clinical variables into the models and consequently the reduced usefulness of these models (19).

Recent reports have suggested an association between the presence of lung metastases and clinical benefit with pembrolizumab (3). Conversely, although not contradictorily, our group previously noted that the presence of liver metastases was associated with poor prognosis in an initial subset of melanoma patients receiving pembrolizumab (20). In this study, we sought to determine the relationship between metastatic pattern, organ-specific differential T-cell infiltration, and treatment outcome in patients treated with pembrolizumab.

Study design

Between December 2011 and October 2013, 223 patients with melanoma were treated with pembrolizumab at University of California, San Francisco (UCSF, San Francisco, CA), University of California, Los Angeles (UCLA; Los Angeles, CA), or the Angeles Clinic as part of KEYNOTE-001 (ClinicalTrials.gov NCT01295827). This trial was a large phase I clinical trial that examined the safety and efficacy of pembrolizumab in patients with multiple solid tumor malignancies. An independent validation cohort was comprised of 113 patients treated with pembrolizumab between February 2013 and September 2015 at UCSF or at the University Hospital of Zürich (Zürich, Switzerland) on Keynotes 002 and 006, and Merck EAP (ClinicalTrials.gov identifiers, P002: NCT 01704287, P006: NCT01866319, EAP: NCT02083484). A comparison cohort of 165 patients with advanced non–small cell lung cancer treated with pembrolizumab at UCSF, MSKCC, or UCLA as part of KEYNOTE-001 (ClinicalTrials.gov NCT01295827) was also examined.

In the discovery and validation melanoma cohorts, as well as the NSCLC comparison cohort, pembrolizumab was administered intravenously at 2 or 10 mg/kg every 2 or 3 weeks as described previously (3). Efficacy was determined in the melanoma cohorts by RECIST v 1.1 using CT imaging at 12 and 16 weeks after the first infusion, and every 12 weeks thereafter (21). Progression-free survival (PFS) was calculated from the date of randomization to the date of progression or death. In the NSCLC cohort, efficacy was determined using immune-related response criteria, and scans were repeated every 9 weeks (10).

Available efficacy and immunologic data as of July 1, 2015, were included in all the analyses. The efficacy analysis included two endpoints: (i) best overall response was defined as the best tumor response from the start of treatment to the time of disease progression or death; and (ii) PFS was defined as the interval between the date of enrollment and the date of progression or death (or the last date of clinic visit where the patient was known not to have had radiological or clinical progression). Best overall response was determined from investigator-reported data according to RECIST 1.1 criteria.

Tumor sample procurement

Melanoma patients underwent an optional biopsy before starting treatment. Of these, 61 samples were available for IHC analysis. Biopsy collection and analyses were approved by IRBs 11-003066 (UCLA) and 13-12246 (UCSF).

IHC staining

Slides were stained with hematoxylin and eosin, S100, CD8, PD-1, and PD-L1 at the UCLA Clinical IHC Laboratory as described previously (18). All stained slides were evaluated in a blinded fashion by one dermatopathologist and one investigator trained to identify the features of melanoma. S100, an established melanoma marker, was used to define the histologic tumor margin. Slides were examined for the presence of CD8, PD-1, and PD-L1 at the invasive tumor margin as described previously (18).

Digital image acquisition and analysis

All slides were scanned at an absolute magnification of ×200 (resolution of 0.5 μm/pixel). An algorithm was designed on the basis of pattern recognition that quantified immune cells. Image analysis based on RGB (red, green, blue) spectra was used to detect all cells by counterstaining with hematoxylin (blue), and DAB or fast red. The imaging analysis algorithm calculated the density of CD8, PD-1, and PD-L1–positive cells (cells/mm2).

Statistical analyses

Demographic and clinical characteristics were compared using the Kruskal–Wallis or Student t test for age and continuous variables, and χ2 or Fisher exact test for categorical variables. Ninety-five percent confidence intervals (CI) were computed using the Wald confidence limits for the binomial proportion. Proportional hazard Cox regression was used to determine the association of demographic and clinical variables with response and PFS. The full model included terms for metastatic location, age, gender, previous targeted therapy, BRAF status, baseline tumor burden, lactate dehydrogenase (LDH) level, and primary site of melanoma. PFS curves were constructed with the Kaplan–Meier method separately stratified by primary site of melanoma and metastatic location. Analyses were performed using SAS V9.4 (SAS Institute Inc.), SPSS V22 (IBM Corp.), and GraphPad Prism v6 (GraphPad Software, Inc.). All tests were two-sided with P values <0.05 considered statistically significant.

Baseline clinical characteristics according to metastatic pattern

Table 1 shows the baseline characteristics of the melanoma patients in the discovery and validation cohorts stratified by metastatic pattern. Variables that were significantly associated with best overall response included gender, LDH concentration, prior ipilimumab therapy, and metastatic site (Supplementary Fig. S1). Pembrolizumab dosing and schedule were not included in the analysis, because multiple reports have independently examined this issue and confirmed the lack of association (3, 7, 22–24).

Table 1.

Baseline demographic and clinical characteristics of patients with and without liver metastases in discovery and validation cohort populations

Discovery (n = 223)Validation (n = 113)
CharacteristicsLiver − (n = 151)Liver + (n = 72)Liver − (n = 77)Liver + (n = 36)
Median age – y (range) 64 (26–95) 65 (19–77) 61 (21–91) 61.5 (28–87) 
Sex – n (%) 
 Female 53 (35%) 25 (35%) 27 (35%) 13 (36%) 
 Male 98 (65%) 47 (65%) 50 (65%) 23 (64%) 
ECOG performance status – n (%) 
 0 116 (77%) 49 (68%) 63 (82%) 20 (56%) 
 1 35 (23%) 23 (32%) 14 (18%) 16 (44%) 
LDH level – n (%) 
 Elevated (>199) 58 (38%) 43 (60%) 49 (64%) 27 (75%) 
 Normal (≤199) 93 (62%) 29 (40%) 28 (36%) 9 (25%) 
Primary site of melanoma – n (%) 
 Cutaneous 125 (83%) 47 (65.3%) 66 (85.7%) 30 (83.|$\bar{3}$|%) 
 Mucosal 14 (9%) 5 (6.9%) 2 (2.6%) 3 (8.|$\ \bar{3}$|%) 
 Unknown 10 (7%) 1 (1.4%) 9 (11.7%) 3 (8.|$\ \bar{3}$|%) 
 Uveal 2 (1%) 19 (26.4%) 0 (0%) 0 (0%) 
Previous targeted therapy – n (%) 
 No 119 (79%) 57 (79%) 57 (74%) 27 (75%) 
 Yes 32 (21%) 15 (21%) 20 (26%) 9 (25%) 
Previous ipilimumab – n (%) 
 No 83 (55%) 37 (51%) 12 (16%) 6 (17%) 
 Yes 68 (45%) 35 (49%) 65 (84%) 30 (83%) 
BRAF V600E status – n (%) 
 Mutated 41 (27%) 15 (21%) 25 (32%) 8 (22%) 
 Wild type 110 (73%) 57 (79%) 52 (68%) 28 (78%) 
Brain metastasis – n (%) 
 No 119 (79%) 65 (90%) 59 (77%) 27 (75%) 
 Yes 32 (21%) 7 (10%) 18 (23%) 9 (25%) 
Lung metastasis – n (%) 
 No 73 (48%) 42 (58%) 27 (35%) 15 (42%) 
 Yes 78 (52%) 30 (42%) 50 (65%) 21 (58%) 
Discovery (n = 223)Validation (n = 113)
CharacteristicsLiver − (n = 151)Liver + (n = 72)Liver − (n = 77)Liver + (n = 36)
Median age – y (range) 64 (26–95) 65 (19–77) 61 (21–91) 61.5 (28–87) 
Sex – n (%) 
 Female 53 (35%) 25 (35%) 27 (35%) 13 (36%) 
 Male 98 (65%) 47 (65%) 50 (65%) 23 (64%) 
ECOG performance status – n (%) 
 0 116 (77%) 49 (68%) 63 (82%) 20 (56%) 
 1 35 (23%) 23 (32%) 14 (18%) 16 (44%) 
LDH level – n (%) 
 Elevated (>199) 58 (38%) 43 (60%) 49 (64%) 27 (75%) 
 Normal (≤199) 93 (62%) 29 (40%) 28 (36%) 9 (25%) 
Primary site of melanoma – n (%) 
 Cutaneous 125 (83%) 47 (65.3%) 66 (85.7%) 30 (83.|$\bar{3}$|%) 
 Mucosal 14 (9%) 5 (6.9%) 2 (2.6%) 3 (8.|$\ \bar{3}$|%) 
 Unknown 10 (7%) 1 (1.4%) 9 (11.7%) 3 (8.|$\ \bar{3}$|%) 
 Uveal 2 (1%) 19 (26.4%) 0 (0%) 0 (0%) 
Previous targeted therapy – n (%) 
 No 119 (79%) 57 (79%) 57 (74%) 27 (75%) 
 Yes 32 (21%) 15 (21%) 20 (26%) 9 (25%) 
Previous ipilimumab – n (%) 
 No 83 (55%) 37 (51%) 12 (16%) 6 (17%) 
 Yes 68 (45%) 35 (49%) 65 (84%) 30 (83%) 
BRAF V600E status – n (%) 
 Mutated 41 (27%) 15 (21%) 25 (32%) 8 (22%) 
 Wild type 110 (73%) 57 (79%) 52 (68%) 28 (78%) 
Brain metastasis – n (%) 
 No 119 (79%) 65 (90%) 59 (77%) 27 (75%) 
 Yes 32 (21%) 7 (10%) 18 (23%) 9 (25%) 
Lung metastasis – n (%) 
 No 73 (48%) 42 (58%) 27 (35%) 15 (42%) 
 Yes 78 (52%) 30 (42%) 50 (65%) 21 (58%) 

Abbreviation: ECOG, Eastern Cooperative Oncology Group.

Multivariate analysis of prognostic factors including metastatic site

On the basis of the univariate analysis described above, we constructed a multivariate model including the entire melanoma population (Fig. 1). In the multivariate analysis, female gender, elevated LDH, ECOG > 0, and the presence of liver metastasis were all significantly correlated with worse PFS (Fig. 1). Other significant variables in multivariable analysis included prior ipilimumab treatment, whereas the history of brain metastasis, BRAF status, and prior targeted therapy were not significant.

Figure 1.

Multivariable analysis of pretreatment prognostic factors. ORs were calculated for PFS in each subgroup. The statistical significance and the CIs are depicted on the right columns. The dotted vertical line designates the PFS for the entire cohort.

Figure 1.

Multivariable analysis of pretreatment prognostic factors. ORs were calculated for PFS in each subgroup. The statistical significance and the CIs are depicted on the right columns. The dotted vertical line designates the PFS for the entire cohort.

Close modal

PFS and objective response rate, by metastatic pattern

Having identified the significance of liver metastasis in the multivariate analysis of the entire melanoma population, we examined the relationship between liver metastasis and PFS and objective response rate (ORR) in the discovery cohort. In the discovery cohort (Fig. 2A and B), the outcome for the liver metastases groups was worse than the outcome of patients without liver involvement. For example, the median PFS was 5.1 months for patients with liver metastasis, whereas it was 20.1 months for patients without liver metastasis. This difference was statistically significant (P < 0.0001). This effect was confirmed in the validation cohort (Fig. 3A and B); patients with liver metastasis had a median PFS of 2.7 months versus a median PFS of 18.5 months for patients without liver involvement. This difference was statistically significant in the validation cohort as well (P = 0.0006).

Figure 2.

PFS and response rate in patients with melanoma treated with pembrolizumab. A and B, The discovery cohort (n = 223). A, Kaplan–Meier PFS curve; patients with liver metastasis are shown in gray, whereas those without liver metastasis are shown in black. B, Waterfall plots for patients without and with liver metastasis.

Figure 2.

PFS and response rate in patients with melanoma treated with pembrolizumab. A and B, The discovery cohort (n = 223). A, Kaplan–Meier PFS curve; patients with liver metastasis are shown in gray, whereas those without liver metastasis are shown in black. B, Waterfall plots for patients without and with liver metastasis.

Close modal
Figure 3.

PFS and response rate in the validation cohort (n = 113) patients with melanoma treated with pembrolizumab. A, Kaplan–Meier PFS curve with liver metastasis (gray) and those without liver metastasis (black). B, Waterfall plot for patients in the validation cohort. PFS was calculated from the time of randomization. Log-rank analysis was used to calculate the differences between these groups.

Figure 3.

PFS and response rate in the validation cohort (n = 113) patients with melanoma treated with pembrolizumab. A, Kaplan–Meier PFS curve with liver metastasis (gray) and those without liver metastasis (black). B, Waterfall plot for patients in the validation cohort. PFS was calculated from the time of randomization. Log-rank analysis was used to calculate the differences between these groups.

Close modal

Tumor margin CD8+ T-cell count and response to pembrolizumab, by metastatic pattern

Having identified the importance of metastatic pattern to response, we investigated the relationship between the presence of preexisting tumor-associated T-cell infiltrates and metastatic pattern with quantitative IHC analysis of CD8, PD-1, and PD-L1 expression at the invasive margin in samples obtained from 61 patients before treatment (Fig. 4). Fewer CD8+ T cells were found in the nonresponder group when compared with the responder group (responder group n = 25, nonresponder group n = 35, P < 0.0001, Fig. 4A). The CD8+ T-cell count at the invasive margin was also significantly lower in the liver metastasis+ group versus the liver metastasis group (liver metastases group n = 22, mean count 547 ± 164.8; no liver metastases group, n = 40, mean count 1441 ± 250.7; P < 0.016; Fig. 4B). In the same tumor samples, PD-1 and PD-L1 expression by IHC was not significantly different in the liver metastases+ cohort as compared with the liver metastases cohort (Fig. 4C and D). Figure 4E shows examples of CD8, PD-1, and PD-L1 expression in samples obtained from distant metastatic tumors in terms of the presence or absence of liver metastases and response to pembrolizumab.

Figure 4.

IHC analysis of CD8, PD-1, and PD-L1 in samples obtained from patients before pembrolizumab, according to response and metastatic pattern. A, CD8+ T-cell density by response category (responder group n = 25, nonresponder group n = 35; P < 0.0001). B, CD8+ T-cell density by liver metastasis status (liver metastasis+ group n = 22, mean count 547 ± 164.8; liver metastases group n = 40, mean count 1,441 ± 250.7; P < 0.016). C and D, PD-1 and PD-L1, respectively, expression by liver metastasis. NS, not significant. E, Examples of CD8, PD-1, and PD-L1 expression in samples obtained from distant tumors in terms of metastatic location and response. Magnification, ×20. Serial cut tissue sections (2 μm) were stained for S100, CD8, PD-1, and PD-L1. **, <0.016; ****, <0.0001.

Figure 4.

IHC analysis of CD8, PD-1, and PD-L1 in samples obtained from patients before pembrolizumab, according to response and metastatic pattern. A, CD8+ T-cell density by response category (responder group n = 25, nonresponder group n = 35; P < 0.0001). B, CD8+ T-cell density by liver metastasis status (liver metastasis+ group n = 22, mean count 547 ± 164.8; liver metastases group n = 40, mean count 1,441 ± 250.7; P < 0.016). C and D, PD-1 and PD-L1, respectively, expression by liver metastasis. NS, not significant. E, Examples of CD8, PD-1, and PD-L1 expression in samples obtained from distant tumors in terms of metastatic location and response. Magnification, ×20. Serial cut tissue sections (2 μm) were stained for S100, CD8, PD-1, and PD-L1. **, <0.016; ****, <0.0001.

Close modal

Patients with liver metastases also had significantly lower densities of CD8+ T cells in distant nonliver metastases. We obtained archived tumor samples that represented 35 patients with confirmed melanoma metastases in the liver and in nonliver sites, but were never treated with pembrolizumab (Supplementary Fig. S2). We analyzed the presence of CD8+ T cells in the nonliver biopsies from these patients. CD8 expression in the nonliver biopsies of the distant metastases group was comparable with the liver metastases+ group (liver metastases+ group, 546.9 ± 164.8; distant metastases of liver metastases+ group, 479.1 ± 98.49, P = 0.7079) and significantly lower when compared with the liver metastases group (liver metastases group, 1,441 ± 250.7; P ≤ 0.0001; distant metastases group, P = 35; liver metastases group, n = 40, liver metastases+ group, n = 22).

PFS and metastatic pattern in NSCLC

Given the association between liver metastasis and the outcome of pembrolizumab treatment in melanoma, we investigated whether this relationship could be seen in NSCLC. As with melanoma, there is a significant body of data on patients treated with pembrolizumab. In this tumor type, the PFS was also significantly reduced in patients with liver metastasis (n = 46, median PFS 1.82 months; 95% CI, 1.36–2.02) compared with those without liver metastasis (n = 119, median PFS 4.03 months; 95% CI, 2.12–5.09), P = 0.0094 (Kaplan–Meier curves for PFS, Fig. 5A). ORRs were also lower in the liver metastases+ population compared with the liver metastases population (waterfall plots, Fig. 5B).

Figure 5.

PFS and response rate in patients with NSCLC treated with pembrolizumab. A, Kaplan–Meier PFS curve; patients with liver metastasis are shown in gray, whereas those without liver metastasis are shown in black. B, Waterfall plots for patients without and with liver metastasis.

Figure 5.

PFS and response rate in patients with NSCLC treated with pembrolizumab. A, Kaplan–Meier PFS curve; patients with liver metastasis are shown in gray, whereas those without liver metastasis are shown in black. B, Waterfall plots for patients without and with liver metastasis.

Close modal

In this report, we investigated the clinical characteristics of nonresponders to pembrolizumab. In melanoma patients, we discovered that liver metastasis was independently predictive of reduced response and poor outcome. In a separate validation cohort, this relationship was confirmed. This effect is not due to advanced stage (M1C) alone (ORR for M1C, 45.5%) or due to the site of origin of melanoma (uveal melanoma patients were excluded from the validation cohort). Additional factors noted to be significantly associated with adverse outcome in the multivariable analysis included female gender, elevated LDH, and prior ipilimumab treatment.

The presence of liver metastases was associated with fewer infiltrating CD8+ T cells at the invasive margin in distant tumors, a cellular signature that correlates with response to PD-1. We extended this observation in a set of patients who had cutaneous metastasis as well as liver metastasis. In this set of tumors, the cutaneous metastases also had depleted marginating CD8+ T cells, suggesting that the effect of liver metastasis was systemic.

In a comparison cohort of patients with NSCLC, the presence of liver metastasis was associated with decreased likelihood of response to pembrolizumab. Although pembrolizumab has less activity in terms of ORR and PFS in NSCLC compared with melanoma (10, 11), the same pattern of reduced efficacy was seen in patients with liver metastasis.

The decreased probability of response to pembrolizumab seen in patients with liver metastasis can have several explanations. Liver-induced peripheral tolerance represents a well-established but poorly understood phenomenon that was initially described in the setting of orthotopic liver transplantation. Unlike heart or kidney allografts, liver allografts are accepted spontaneously in mice, rats, pigs, and even in humans, often without the need for histocompatibility or even in some instances, immunosuppression (25–27). In addition, liver allografts confer on the recipient tolerance to other transplanted organs from the same donor, suggesting that the transplanted liver can induce systemic immune tolerance (26, 28). Multiple mechanisms have been put forward to explain liver-induced systemic tolerance, including incomplete activation of CD8+ T cells (29–32), trapping and deletion of activated CD8+ cells (33, 34), poor CD4+ T-cell activation (35), and Kupffer cells promoting activation of regulatory T cells (30). In addition, it appears that viral pathogens, in particular hepatitis C virus (HCV) and lymphocytic choriomeningitis virus, may exploit mechanisms of liver tolerance to evade antiviral CD8 responses, including the direct upregulation of PD-L1 on myeloid-derived Kupffer cells by HCV (36). Mechanistic studies using animal models may help to distinguish between these possibilities in the future. It is also possible that other unexamined variables may explain the findings we describe. Other studies have shown that baseline tumor size (37), tumor aneuploidy (38), tumor mutation burden (39), intestinal microbial flora (40), and tumor wnt pathway signaling (41) can all affect response to checkpoint inhibitors. It is certainly possible that these could be confounding factors in terms of response. Although these mechanistic studies and additional confirmatory studies are ongoing, the presence of liver metastasis should not be used to exclude patients from PD-1 therapy. Indeed, the response rate even in this group of patients exceeds the response rate reported for other therapies.

M.D. Hellmann reports receiving other commercial research support from BMS and Genentech and is a consultant/advisory board member for AstraZeneca, BMS, Genentech, Janssen, Merck, and Novartis. O. Hamid has received speakers bureau honoraria from Amgen, BMS, Genentech, and Novartis, is a consultant/advisory board member for Amgen, BMS, Merck, Novartis, and Roche. M.A. Gubens is a consultant/advisory board member for BMS. J.M. Taube reports receiving a commercial research grant from Bristol-Myers Squibb and is a consultant/advisory board member for Astra Zeneca, Bristol-Myers Squibb, and Merck. E.J. Taylor is the chief executive officer at Naked Biome. B. Chmielowski has received speakers bureau honoraria from Genentech and Janssen, is a consultant/advisory board member for Amgen, Astellas, BMS, Eisai, Genentech, Immunocore, Lilly, and Merck. R. Dummer is a consultant/advisory board member for BMS and MSD. S.M. Goldinger is a consultant/advisory board member for BMS, MSD, Novaertis, and Roche. No potential conflicts of interest were disclosed by the other authors.

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NCI or the NIH. The funders and sponsors had no role in the design and conduct of the study, collection, management, analysis, and interpretation of the data, preparation, review, or approval of the manuscript, and decision to submit the manuscript for publication.

Conception and design: P.C. Tumeh, M.D. Hellmann, C.L. Harview, J.M. Taube, M.F. Krummel, P.N. Munster, B. Chmielowski, R.H. Pierce, A. Daud

Development of methodology: P.C. Tumeh, M.D. Hellmann, K.K. Tsai, M. Rosenblum, P.J. Sanchez, N. Banayan, R.H. Pierce, A. Daud

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): P.C. Tumeh, M.D. Hellmann, O. Hamid, K.K. Tsai, K.L. Loo, M.A. Gubens, M. Rosenblum, C.L. Harview, J.M. Taube, N. Handley, N. Khurana, A. Nosrati, A. Tucker, E.V. Sosa, P.J. Sanchez, N. Banayan, J.C. Osorio, J. Chang, I.P. Shintaku, P.D. Boasberg, P.N. Munster, A.P. Algazi, B. Chmielowski, R. Dummer, S.M. Goldinger, E.B. Garon, A. Daud

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.C. Tumeh, M.D. Hellmann, O. Hamid, K.K. Tsai, K.L. Loo, M.A. Gubens, C.L. Harview, J.M. Taube, N. Handley, M.F. Krummel, A. Tucker, P.J. Sanchez, N. Banayan, J.C. Osorio, D.L. Nguyen-Kim, J. Chang, A.P. Algazi, B. Chmielowski, R. Dummer, T.R. Grogan, D. Elashoff, J. Hwang, S.M. Goldinger, E.B. Garon, R.H. Pierce, A. Daud

Writing, review, and/or revision of the manuscript: P.C. Tumeh, M.D. Hellmann, O. Hamid, K.K. Tsai, M.A. Gubens, M. Rosenblum, J.M. Taube, N. Handley, A. Nosrati, M.F. Krummel, A. Tucker, E.V. Sosa, D.L. Nguyen-Kim, E.J. Taylor, A.P. Algazi, B. Chmielowski, R. Dummer, T.R. Grogan, D. Elashoff, S.M. Goldinger, E.B. Garon, R.H. Pierce, A. Daud

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P.C. Tumeh, M.D. Hellmann, K.L. Loo, C.L. Harview, N. Khurana, A. Nosrati, A. Tucker, E.V. Sosa, J. Chang, P.N. Munster, B. Chmielowski, S.M. Goldinger, A. Daud

Study supervision: P.C. Tumeh, A. Daud

P.C. Tumeh is a Damon Runyon Clinical Investigator and was in part supported by the Damon Runyon Cancer Research Foundation (CI-79-15), NIH (K08 AI091663 and UL1TR000124), the Jonsson Comprehensive Cancer Center (JCCC), Kure-It Cancer Research, and STOP Cancer. A. Daud was supported by the Helen Diller Comprehensive Cancer Center and the Amoroso and Cook Fund. E.B. Garon was supported by NIH (R01 CA208403).

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.
Brahmer
JR
,
Drake
CG
,
Wollner
I
,
Powderly
JD
,
Picus
J
,
Sharfman
WH
, et al
Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates
.
J Clin Oncol
2010
;
28
:
3167
75
.
2.
Brahmer
JR
,
Tykodi
SS
,
Chow
LQM
,
Hwu
W-J
,
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
.
3.
Hamid
O
,
Robert
C
,
Daud
A
,
Hodi
FS
,
Hwu
W-J
,
Kefford
R
, et al
Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma
.
N Engl J Med
2013
;
369
:
134
44
.
4.
Topalian
SL
,
Hodi
FS
,
Brahmer
JR
,
Gettinger
SN
,
Smith
DC
,
McDermott
DF
, et al
Safety, activity, and immune correlates of anti-PD-1 antibody in cancer
.
N Engl J Med
2012
;
366
:
2443
54
.
5.
Wolchok
JD
,
Kluger
H
,
Callahan
MK
,
Postow
MA
,
Rizvi
NA
,
Lesokhin
AM
, et al
Nivolumab plus ipilimumab in advanced melanoma
.
N Engl J Med
2013
;
369
:
122
33
.
6.
Robert
C
,
Thomas
L
,
Bondarenko
I
,
O'Day
S
,
Weber
J
,
Garbe
C
, et al
Ipilimumab plus dacarbazine for previously untreated metastatic melanoma
.
N Engl J Med
2011
;
364
:
2517
26
.
7.
Robert
C
,
Ribas
A
,
Wolchok
JD
,
Hodi
FS
,
Hamid
O
,
Kefford
R
, et al
Anti-programmed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial
.
Lancet
2014
;
384
:
1109
17
.
8.
Antonia
S
,
Goldberg
SB
,
Balmanoukian
A
,
Chaft
JE
,
Sanborn
RE
,
Gupta
A
, et al
Safety and antitumour activity of durvalumab plus tremelimumab in non-small cell lung cancer: a multicentre, phase 1b study
.
Lancet Oncol
2016
;
17
:
299
308
.
9.
Borghaei
H
,
Paz-Ares
L
,
Horn
L
,
Spigel
DR
,
Steins
M
,
Ready
NE
, et al
Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer
.
N Engl J Med
2015
;
373
:
1627
39
.
10.
Garon
EB
,
Rizvi
NA
,
Hui
R
,
Leighl
N
,
Balmanoukian
AS
,
Eder
JP
, et al
Pembrolizumab for the treatment of non-small-cell lung cancer
.
N Engl J Med
2015
;
372
:
2018
28
.
11.
Herbst
RS
,
Baas
P
,
Kim
D-W
,
Felip
E
,
Pérez-Gracia
JL
,
Han
J-Y
, et al
Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial
.
Lancet Lond Engl
2016
;
387
:
1540
50
.
12.
Callahan
MK
,
Postow
MA
,
Wolchok
JD
. 
Targeting T cell co-receptors for cancer therapy
.
Immunity
2016
;
44
:
1069
78
.
13.
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
.
14.
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
.
15.
Pardoll
DM
. 
The blockade of immune checkpoints in cancer immunotherapy
.
Nat Rev Cancer
2012
;
12
:
252
64
.
16.
Daud
AI
,
Hamid
O
,
Ribas
A
,
Hodi
FS
,
Hwu
W-J
,
Kefford
R
, et al
Abstract CT104: antitumor activity of the anti-PD-1 monoclonal antibody MK-3475 in melanoma(MEL): correlation of tumor PD-L1 expression with outcome
.
Cancer Res
2014
;
74
(
19 Suppl
):
CT104
.
17.
Herbst
RS
,
Soria
J-C
,
Kowanetz
M
,
Fine
GD
,
Hamid
O
,
Gordon
MS
, et al
Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients
.
Nature
2014
;
515
:
563
7
.
18.
Tumeh
PC
,
Harview
CL
,
Yearley
JH
,
Shintaku
IP
,
Taylor
EJM
,
Robert
L
, et al
PD-1 blockade induces responses by inhibiting adaptive immune resistance
.
Nature
2014
;
515
:
568
71
.
19.
Blank
CU
,
Haanen
JB
,
Ribas
A
,
Schumacher
TN
. 
CANCER IMMUNOLOGY. The “cancer immunogram.”
Science
2016
;
352
:
658
60
.
20.
Tsai
KK
,
Loo
K
,
Khurana
N
,
Algazi
AP
,
Hwang
J
,
Sanchez
R
, et al
Clinical characteristics predictive of response to pembrolizumab in advanced melanoma
.
J Clin Oncol
33
, 
2015
(
suppl; abstr 9031
).
21.
Eisenhauer
EA
,
Therasse
P
,
Bogaerts
J
,
Schwartz
LH
,
Sargent
D
,
Ford
R
, et al
New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1)
.
Eur J Cancer
2009
;
45
:
228
47
.
22.
Daud
AI
,
Ribas
A
,
Robert
C
,
Hodi
FS
,
Wolchok
JD
,
Joshua
AM
, et al
Long-term efficacy of pembrolizumab (pembro; MK-3475) in a pooled analysis of 655 patients (pts) with advanced melanoma (MEL) enrolled in KEYNOTE-001
.
J Clin Oncol
33
, 
2015
(
suppl; abstr 9005
).
23.
Ribas
A
,
Puzanov
I
,
Dummer
R
,
Schadendorf
D
,
Hamid
O
,
Robert
C
, et al
Pembrolizumab versus investigator-choice chemotherapy for ipilimumab-refractory melanoma (KEYNOTE-002): a randomised, controlled, phase 2 trial
.
Lancet Oncol
2015
;
16
:
908
18
.
24.
Robert
C
,
Schachter
J
,
Long
GV
,
Arance
A
,
Grob
JJ
,
Mortier
L
, et al
Pembrolizumab versus ipilimumab in advanced melanoma
.
N Engl J Med
2015
;
372
:
2521
32
.
25.
Calne
RY
. 
Mechanisms in the acceptance of organ grafts
.
Br Med Bull
1976
;
32
:
107
12
.
26.
Calne
RY
,
Sells
RA
,
Pena
JR
,
Davis
DR
,
Millard
PR
,
Herbertson
BM
, et al
Induction of immunological tolerance by porcine liver allografts
.
Nature
1969
;
223
:
472
6
.
27.
Cantor
HM
,
Dumont
AE
. 
Hepatic suppression of sensitization to antigen absorbed into the portal system
.
Nature
1967
;
215
:
744
5
.
28.
Calne
RY
. 
Immunological tolerance – the liver effect
.
Immunol Rev
2000
;
174
:
280
2
.
29.
Qian
S
,
Lu
L
,
Fu
F
,
Li
Y
,
Li
W
,
Starzl
TE
, et al
Apoptosis within spontaneously accepted mouse liver allografts: evidence for deletion of cytotoxic T cells and implications for tolerance induction
.
J Immunol
1997
;
158
:
4654
61
.
30.
Jenne
CN
,
Kubes
P
. 
Immune surveillance by the liver
.
Nat Immunol
2013
;
14
:
996
1006
.
31.
Limmer
A
,
Ohl
J
,
Kurts
C
,
Ljunggren
H-G
,
Reiss
Y
,
Groettrup
M
, et al
Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance
.
Nat Med
2000
;
6
:
1348
54
.
32.
Limmer
A
,
Ohl
J
,
Wingender
G
,
Berg
M
,
Jüngerkes
F
,
Schumak
B
, et al
Cross-presentation of oral antigens by liver sinusoidal endothelial cells leads to CD8 T cell tolerance
.
Eur J Immunol
2005
;
35
:
2970
81
.
33.
Crispe
IN
. 
Hepatic T cells and liver tolerance
.
Nat Rev Immunol
2003
;
3
:
51
62
.
34.
John
B
,
Crispe
IN
. 
Passive and active mechanisms trap activated CD8+ T cells in the liver
.
J Immunol
2004
;
172
:
5222
9
.
35.
Wang
J-CE
,
Livingstone
AM
. 
Cutting edge: CD4+ T cell help can be essential for primary CD8+ T cell responses in vivo
.
J Immunol
2003
;
171
:
6339
43
.
36.
Tu
Z
,
Pierce
RH
,
Kurtis
J
,
Kuroki
Y
,
Crispe
IN
,
Orloff
MS
. 
Hepatitis C virus core protein subverts the antiviral activities of human kupffer cells
.
Gastroenterology
2010
;
138
:
305
14
.
37.
Ribas
A
,
Hamid
O
,
Daud
A
,
Hodi
FS
,
Wolchok
JD
,
Kefford
R
, et al
Association of pembrolizumab with tumor response and survival among patients with advanced melanoma
.
JAMA
2016
;
315
:
1600
9
.
38.
Davoli
T
,
Uno
H
,
Wooten
EC
,
Elledge
SJ
. 
Tumor aneuploidy correlates with markers of immune evasion and with reduced response to immunotherapy
.
Science
2017
;
355
:
pii:eaaf8399
.
39.
Rizvi
NA
,
Hellmann
MD
,
Snyder
A
,
Kvistborg
P
,
Makarov
V
,
Havel
JJ
, et al
Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer
.
Science
2015
;
348
:
124
8
.
40.
Sivan
A
,
Corrales
L
,
Hubert
N
,
Williams
JB
,
Aquino-Michaels
K
,
Earley
ZM
, et al
Commensal Bifidobacterium promotes antitumor immunity and facilitates anti-PD-L1 efficacy
.
Science
2015
;
350
:
1084
9
.
41.
Spranger
S
,
Bao
R
,
Gajewski
TF
. 
Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity
.
Nature
2015
;
523
:
231
5
.