Dual Checkpoint Blockade in a Neuroendocrine Carcinoma With Dual PD-L1/PD-L2 Amplification and High Tumor Mutational Burden

INTRODUCTION

Neuroendocrine neoplasms (NENs) comprise a heterogeneous group of rare neoplasms that most commonly arise in the GI tract, bronchopulmonary tree, and pancreas.¹ With the exception of the US Food and Drug Administration (FDA) approvals of the programmed cell death-1 (PD-1) inhibitor nivolumab in refractory small-cell lung cancer and the programmed death-ligand 1 (PD-L1) inhibitor avelumab in metastatic Merkel cell carcinoma, the antitumor efficacy of immune checkpoint inhibitors in NENs remains largely unexplored.^2,3

We present the case of a treatment-refractory, metastatic neuroendocrine carcinoma (NEC) of unknown primary in which next-generation sequencing (NGS) showed amplifications in PD-L1 and PD-L2 along with a high tumor mutational burden (TMB). The patient began therapy with the combination of nivolumab and ipilimumab (cytotoxic T-lymphocyte associated protein-4 or CTLA-4 inhibitor) in the Southwest Oncology Group (SWOG)-directed DART (S1609) trial (ClinicalTrials.gov identifier: NCT02834013) and had a profound response to treatment.

CASE PRESENTATION

A 53-year-old woman initially presented with pelvic pain and left lower extremity neuropathy with imaging that showed left supraclavicular, left retrocrural, and retroperitoneal lymphadenopathy; left gluteal masses; and left hydronephrosis. A left inguinal, soft tissue biopsy specimen and retroperitoneal lymph node biopsy specimen both showed high-grade NEC with Ki-67 > 90% (Fig 1). The patient initially started treatment with cisplatin with etoposide. After 4 cycles, surveillance imaging showed progressive disease along with a new osseous metastasis in the L4 verte-bral body. The patient’s treatment was subsequently transitioned to carboplatin with irinotecan and the lesion at L4 was palliatively radiated. The patient completed 4 total cycles of carboplatin/irinotecan, and surveillance imaging again showed progressive disease. Archival tumor tissue was sent to Perthera, Inc. (McLean, VA) for precision-matched therapeutic options based on multiplatform profiling whereby NGS was performed by FoundationOne (Foundation Medicine, Cambridge, MA) as previously validated⁴ and proteomic analysis by immunohistochemistry was performed by Caris Life Sciences (Phoenix, AZ) using commercially available antibodies as previously described.^5-7 The relevant molecular profiling results are shown in Table 1.

FIG 1.

Computed tomography-guided core needle biopsy specimen of an enlarged retroperitoneal lymph node showed evidence of a high-grade neuroendocrine carcinoma on (A, B) hematoxylin and eosin staining (A, ×40; B, ×60) with (C-F) positivity for CK7 (C, ×40), TTF-1 (D, ×40), synaptophysin (E, ×40), and Ki-67 > 90% (F, ×40). Overall, the histologic and morphologic features were suggestive of a small-cell neuroendocrine carcinoma.

TABLE 1.

Next-Generation DNA Sequencing and Immunohistochemistry From Archival Tumor Tissue

On the basis of the patient’s amplifications in PD-L1 and PD-L2 and her high TMB, she was enrolled in the SWOG-directed DART (S1609) trial and dual checkpoint blockade was begun with the combination of nivolumab 240 mg intravenously every 2 weeks and ipilimumab 1 mg/kg intravenously every 6 weeks. She received nivolumab/ipilimumab for 8 months followed by maintenance nivolumab every 2 weeks for an additional 3 months until her treatment was discontinued for grade 3 colitis. After 8 months of receiving therapy, the patient’s NEC showed a sustained partial response (Fig 2). She has received no treatment for the past 7 months and surveillance scans have shown stable disease (SD).

FIG 2.

Computed tomography scans from (A) June 2017 and (B) February 2018 after the patient started dual checkpoint blockade therapy with nivolumab plus ipilimumab. Arrows indicate decrease in size of a left para-aortic lymph node. Mild decreases in the size of the retroperitoneal lymphadenopathy and a left pelvic sidewall mass were also observed. There were also findings of stable sclerotic osseous lesions and a left supraclavicular lymph node.

DISCUSSION

Immune checkpoint inhibitors have undergone rapid development and implementation into the treatment paradigms for a increasing number of malignancies.⁸ However, not all patients with cancer treated with PD-(L)1/CTLA-4 inhibitors achieve benefit, and efforts to study predictors of response to checkpoint blockade have recently identified potential biomarkers, including, but not limited to, presence of tumor-infiltrating lymphocytes (TILs), microsatellite instability (MSI), TMB, PD-L1 expression, and the gut microbiome.^8,9

Two of these biomarkers directly relate to the case presented here and provide plausible explanations for the clinical response observed in a tumor type for which checkpoint blockade has not been widely implemented. PD-L1 expression has been among the most extensively studied predictive biomarkers whose presence has since been required for FDA-labeled use of PD-1 inhibitors in several advanced solid tumors.^10-13 Aside from small-cell NECs (including lung) and Merkel cell carcinoma, which have the highest PD-L1 expression, several NEN subtypes have demonstrated PD-L1 and PD-L2 expression that may predict response to immunotherapy.^14,15 Furthermore, higher grade has been associated with significantly more PD-L1 expression in GI neuroendocrine tumors (NETs).¹⁶

Notably, in a recent large series of > 100,000 patient samples, PD-L1 amplifications were identified in 0.7% of cases that included > 100 types of solid tumors, though a small proportion (< 5%) of these were from NENs (virtually all were large-cell NECs).¹⁷ In our case, the tumor was positive for PD-L1 and PD-L2 amplification. Clinical and preclinical evidence has begun to support that genomic amplification of PD-L1, which resides on chromosome 9p24.1, results in strikingly increased expression of PD-L1 that renders tumors susceptible to immune checkpoint blockade.^18-20 Somatic amplification of the genes encoding PD-L1 and PD-L2 (CD274 and PDCD1LG2, respectively) on chromosome 9p24.1 formed the basis for a PD-1 blockade strategy, which has resulted in the FDA approval of nivolumab for treatment of Hodgkin lymphoma.²¹

TMB has recently undergone prospective validation as a predictor of response to dual checkpoint blockade.²² In a randomized, controlled trial of patients with lung cancer (CheckMate-586), there was no association between TMB and PD-L1 expression level, but patients with high TMB (≥ 10 mutations/Mb) treated with dual checkpoint blockade had a significantly longer median progression-free survival (PFS) of 7.2 versus. 5.5 months, when compared with chemotherapy (hazard ratio [HR], 0.58; 95% CI, 0.41 to 0.81; P < .001). Patients with low TMB (< 10 mutations/Mb) treated with dual checkpoint inhibition had a shorter median PFS of 3.2 versus 5.5 months when compared with chemotherapy (HR, 1.07; 95% CI, 0.84 to 1.35), albeit not significant. Historically, TMB has been highest in Merkel cell carcinoma and large-cell NECs (4.3 and 9.9 median mutations/Mb, respectively), but lowest in unknown primary undifferentiated NENs (2.7 median mutations/Mb).²³ The patient we report on had a TMB of 14 mutations/Mb, which is higher than the previously described averages by Chalmers et al²³ and also stratifies her into the high TMB category of CheckMate-586. We hypothesize this could be another reason for her response to treatment with nivolumab/ipilimumab.

There is growing support to suggest that checkpoint blockade has activity in heavily pretreated patients with advanced NETs.^24,25 Although these studies provide support that advanced NETs can derive benefit from checkpoint blockade, additional investigation is needed to identify the subsets of patients with tumor characteristics conducive for immune response to checkpoint inhibitors. The treatment-refractory NEC in the case reported here demonstrated a relatively high TMB with dual amplification of PD-L1 and PD-L2 that, altogether, represents a unique patient subset that could be sought out as candidates for checkpoint blockade. It is worthwhile to note that molecular profiling in this case was performed on archival tumor tissue whereby immunotherapy was administered several months (7 months) after the original tumor biopsy. It is likely that the molecular profile on this patient’s real-time tissue would differ from archival tumor tissue, given that selective pressure from prior treatment contributes to tumor evolution.²⁶ Whether tumor evolution and the forces that dynamically drive such changes in molecular heterogeneity can result in a molecular profile predictive of response to checkpoint blockade deserves more investigation.

Unfortunately, the cumulative evidence thus far suggests that only a limited number of NENs demonstrate a tumor microenvironment that is favorable for successful checkpoint blockade, including PD-L1 positivity, MSI, higher mutational load, and presence of TILs.²⁷ In essence, checkpoint blockade will have to overcome the majority of immunologically silent or tolerant NENs to represent a viable anticancer therapy in this arena. This concept is increasingly being recognized as evidenced by the increasing number of ongoing trials investigating synergistic combinations on a backbone of checkpoint blockade (Table 2).

TABLE 2.

Ongoing Clinical Trials Investigating Immune Checkpoint Inhibitors in Neuroendocrine Neoplasms

AUTHOR CONTRIBUTIONS

Conception and design: Jacob J. Adashek, Richard Tuli, Young K. Chae Razelle Kurzrock

Provision of study material or patients: David Frishberg, Alexandra Gangi

Collection and assembly of data: Jun Gong, Jacob J. Adashek, David Frishberg, Michelle Guan, Veronica R. Placencio-Hickock, Andrew E. Hendifar

Data analysis and interpretation: Jun Gong, Sandip Patel, Jacob J. Adashek, David Frishberg, Alexandra Gangi, Gillian Gresham, Richard Tuli, Razelle Kurzrock, Andrew E. Hendifar

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Jun Gong

Honoraria: Amgen, Astellas Pharma, Clinical Congress Consultants

Consulting or Advisory Role: Amgen, Astellas Pharma, Clinical Congress Consultants

Sandip Patel

Consulting or Advisory Role: Eli Lilly, Novartis, Bristol-Myers Squibb, AstraZeneca/MedImmune, Nektar, Compugen, Illumina

Speakers' Bureau: Merck, Boehringer Ingelheim

Research Funding: Bristol-Myers Squibb (Inst), Pfizer (Inst), Roche (Inst), Amgen (Inst), AstraZeneca/MedImmune (Inst), Fate (Inst), Merck (Inst)

David Frishberg

Stock and Other Ownership Interests: Amgen (I)

Consulting or Advisory Role: Ultragenyx (I), Replimmune (I)

Travel, Accommodations, Expenses: Ultragenyx (I), Replimmune (I)

Veronica R. Placencio-Hickok

Patents, Royalties, Other Intellectual Property: Patent 17814046.3-1111 PCT/US2017037558. Sensitization of tumors to therapies through endoglin antagonism (Inst)

Richard Tuli

Consulting or Advisory Role: AstraZeneca

Research Funding: AstraZeneca, AbbVie

Patents, Royalties, Other Intellectual Property: Patent pending: Methods of Treating Gastrointestinal Malignancies, Application No. 16/256,840 filed: January 24, 2019, inventor (Inst)

Young K. Chae

Consulting or Advisory Role: Foundation Medicine, Boehringer Ingelheim, Biodesix, Counsyl, AstraZeneca, Guardant Health, Takeda, Roche, Immuneoncia, Hanmi, Eli Lilly, Tempus

Speakers' Bureau: Roche, Merck, AstraZeneca, Eli Lilly

Research Funding: AbbVie, Bristol-Myers Squibb, Lexent Bio, Freenome, Biodesix

Travel, Accommodations, Expenses: Hanmi

Razelle Kurzrock

Leadership: CureMatch, CureMetrix

Stock and Other Ownership Interests: CureMatch, IDbyDNA, Soluventis

Honoraria: Roche, EUSA Pharma, NeoGenomics Laboratories, Biocom, NeoMed Therapeutics, Advanced Therapeutics, LEK, AACR, Chugai Pharma USA, Wiley

Consulting or Advisory Role: Actuate Therapeutics, Loxo, XBiotech, Neo-Med, Roche, Gaido, Soluventis, Pfizer, Merck

Speakers' Bureau: Roche

Research Funding: Guardant Health (Inst), Sequenom (Inst), Merck Serono (Inst), Genentech (Inst), Pfizer (Inst), Foundation Medicine (Inst), Incyte (Inst), Konica Minolta (Inst), Grifols (Inst), OmniSeq (Inst), Debiopharm Group (Inst), Boehringer Ingelheim (Inst)

Travel, Accommodations, Expenses: Roche, EUSA Pharma, NeoGenomics Laboratories, Biocom, NeoMed Therapeutics, Advanced Therapeutics, LEK, AACR, Chugai Pharma USA, Wiley

Andrew E. Hendifar

Consulting or Advisory Role: Novartis, Ipsen, Perthera, Celgene, AbbVie

Research Funding: Ipsen

Travel, Accommodations, Expenses: Halozyme

No other potential conflicts of interest were reported.