Abstract
Purpose
Patients with persistent/recurrent cervical cancer following platinum-based chemotherapy have limited therapeutic options. The Gynecologic-Oncology-Group conducted a phase II trial to assess efficacy and tolerability of nivolumab, an immune checkpoint inhibitor, in persistent/recurrent cervical carcinoma.
Patients and Methods
Key eligibility criteria included persistent/recurrent cervical cancer, failure of prior systemic therapy and ECOG PS 0–1. Nivolumab 3 mg/kg was given IV every 2 wk until disease progression or intolerable toxicity. Response was assessed every 8 wk for 6 months and every 12 wk thereafter. The primary endpoints were objective response as assessed by RECIST 1.1. The study used a 2-stage group sequential design. PD-L1 expression was evaluated in tumor specimens by immunohistochemistry (IHC) using a combined-positive-score (CPS) cutoff of ≥1%.
Results
Of 26 enrolled patients with persistent/recurrent cervical cancer, 25 were evaluable for response/toxicity with a median age of 45. 36% had ECOG PS of 1, and 100% had received one prior systemic chemotherapy regimen. PD-L1 expression (≥ 1%) was identified in 77.3% of tumor samples. As of 03/05/19, all patients were off study treatment; median follow-up for survival status was 32 months (range, 2–41.5). There were 21 (84%) patients with a treatment-related adverse event (TRAE) and most were grade 1–2. Six (24%) patients had grade 3 TRAEs with 1 discontinuing nivolumab due to hepatic toxicity. No grade 5 TRAEs occurred, and 2 patients had grade 4 TRAEs. One confirmed partial response (4%; 90% CI, 0.4%−22.9%), duration of response 3.8 months. Thirty-six percent of patients had stable disease (SD) (9/25; 90% CI, 20.2%−54.4%); the median duration of SD was 5.7 months (range, 3.5–12.7). Estimated PFS and OS at 6 months were 16% and 78.4%, respectively.
Conclusion
Single agent nivolumab exhibited low antitumor activity and an acceptable safety profile in patients with persistent/recurrent cervical cancer previously treated with platinum-based chemotherapy.
Keywords: Cervical neoplasms, nivolumab, PD-1, immunotherapy, immune-checkpoint inhibitors
INTRODUCTION
Cervical cancer accounted for 13,240 new cases and 4,170 deaths related to the disease in the United States in 2018 [1]. The standard treatment for recurrent/metastatic cervical cancer is a combination of paclitaxel and cisplatin or paclitaxel, cisplatin and bevacizumab. These treatment approaches are associated with a median survival of approximately one to 1.5 years [2,3]. Once patients progress after initial therapy for recurrent or metastatic disease, options are limited.
HPV DNA is detected in over 99% of cervical cancer specimens and a large portion of these tumors are associated with high risk HPV types 16 and 18. E6 and E7 play a major role in the transformation of HPV-infected cervical keratinocytes. These viral antigens are consistently expressed in HPV-associated neoplasms and may represent ideal targets for cervical cancer immunotherapy [4–7]. However, despite the great potential of immunotherapeutic approaches to treat chemotherapy/radiation resistant cervical cancer, tumor immunity is hindered by the expression of a series of cell surface molecules known as immune checkpoints in the suppressive tumor microenvironment [8]. Compelling evidence indicates that B7 molecules (i.e., B7–1/CD80, B7–2/CD86, B7-H1/PDL1, B7-H2/L-ICOS, B7-DC, B7-H3 and B7-H4) and their ligands (i.e., CTLA-4, CD28, PD-1, ICOS) not only provide crucial positive signals to stimulate and support T-cell activation, but can also offer negative signals that control and suppress potentially protective T-cell responses against spontaneously arising and virally-induced human tumors [8]. Expression of these molecules on the surface of cervical tumor cells, tumor associated macrophages (TAM) and/or dendritic cells (DC), may attenuate or abrogate the ability of the immune system to successfully eliminate antigenic (i.e., virus-infected) tumors such as cervical cancer [8,9].
Because these negative signals in multiple human solid tumors have been shown to be largely provided by PD-1 or programmed death-1, blockade of PD-1/PD-L1 co-inhibitory pathways by novel monoclonal antibodies may represent an effective therapeutic approach to reverse immune suppression while inducing tumor-specific immunity in cervical cancer patients. Consistent with this hypothesis, the activity of pembrolizumab (Keytruda), a fully-human antibody targeting the inhibitory receptor PD1 expressed on activated T-cells has recently been evaluated in KEYNOTE-028 and KEYNOTE-158 trials in women with recurrent cervical cancer [10,11]. In KEYNOTE-028, where by study eligibility criteria all enrolled patients were PD-L1 positive (i.e., expression in ≥1% of tumor or stroma cells by immunohistochemistry), the objective response rate (ORs) was 17% (4/24), as measured by standard RECIST criteria, with 3 patients demonstrating stable disease and 16 progressive disease as best response [10,11]. In KEYNOTE-158, objective response rate (ORs) was 13.3% (13/98), as measured by standard RECIST criteria with 17 patients having stable disease as best response. All responses occurred among patients with PD-L1–positive tumors. Most of these responses were durable and treatment was well tolerated with a manageable toxicity profile in both studies. Based on these results (i.e. tumor response rate and durability of response), the FDA recently approved pembrolizumab (Keytruda) for the treatment of patients with recurrent or metastatic cervical cancer following disease progression on or after chemotherapy whose tumors express programmed death ligand 1 (PD-L1) (combined positive score [CPS] ≥1) as determined by an FDA-approved test.
Similar to pembrolizumab, nivolumab (Opdivo) is a fully human monoclonal immunoglobulin G4 (IgG4) antibody (HuMAb) that is specific for human programmed death-1 (PD-1, cluster of differentiation 279 [CD279]) cell surface membrane receptor. NRG/GOG conducted a phase II trial of single-agent nivolumab in patients with persistent or recurrent cervical carcinoma. The primary endpoints of this study were the frequency of objective tumor response, and the frequency and severity of adverse events as assessed by CTCAE v4. The secondary objective was to estimate the distributions of progression-free survival (PFS) and overall survival (OS).
MATERIALS AND METHODS
Eligibility
Eligibility criteria included patients with persistent, recurrent or metastatic squamous cell carcinoma, adenosquamous carcinoma or adenocarcinoma of the cervix with documented disease progression following chemotherapy; one prior systemic chemotherapeutic regimen for management of persistent, recurrent or metastatic disease; measurable disease and at least one target lesion as defined by RECIST 1.1; a GOG performance status of 0 or 1; recovery from effects of recent surgery, radiotherapy or chemotherapy; and adequate hematologic (absolute neutrophil count ≥1,500/μL and platelets ≥100,000/μL), renal (serum creatinine ≤1.5x the institutional upper limit of normal [ULN]); hepatic (serum bilirubin ≤1.5x ULN, and both AST and alkaline phosphatase ≤2.5x ULN) laboratory values. Histologic documentation of the original primary tumor was required with a pathology report. The study received local institutional review board approval at participating institutions and all patients provided authorization permitting the release of personal health information and gave informed consent according to institutional and federal guidelines before enrollment.
PD-L1 Expression
Tumor and immune cell PD-L1 expression was assessed in pretreatment (archival) formalin-fixed paraffin embedded tumor biopsy specimens of 22 evaluable patients by immunohistochemistry using rabbit monoclonal anti-human PD-L1 antibody clone E1L3N (Cell Signaling Technologies) at 1:100 dilution on the Leica Bond platform. Tumor cell PD-L1 expression was defined as the percentage of tumor cells exhibiting membrane staining at any intensity. Combined Positive Score (CPS) was defined as the number of PD-L1 positive cells (including tumor cells, lymphocytes, and macrophages) divided by the total number of viable tumor cells × 100 for a post hoc analysis. Samples with a combined-positive-score (CPS) cutoff of 1% or more were defined as positive.
Treatment
Patients were treated with 4 doses of IV nivolumab (3 mg/kg every 2 weeks), followed by an additional 42 doses 3 mg/kg every 2 weeks for a maximum of 46 doses until disease progression or adverse effects prohibit therapy. Toxicity was monitored with history, physical examination, and laboratory assessment before each treatment cycle. Adverse events were defined and graded according to National Cancer Institute Common Terminology Criteria version 4.0 (CTCAE v4).
Evaluations
Response on nivolumab was assessed according to RECIST (1.1 version), either by computed tomography or magnetic resonance imaging at baseline, and every 8 weeks for the first 6 months, then every 12 weeks thereafter until the determination of disease progression. Therapy was discontinued if there was disease progression, unacceptable toxicity, receipt of other anticancer therapy, or voluntary withdrawal.
Statistical Design
The primary objective was to assess the anti-tumor activity of nivolumab measured by objective tumor response (partial or complete). The null hypothesis was that the probability of objective tumor response is 5% or less, determined from historical single-agent GOG-0127 studies that had a similar population of patients to this trial with minimal to modest activity; the alternative hypothesis was the complement of the null, but at least a 20% improvement in the probability of objective tumor response is clinically interesting for further investigation. An optimal flexible 2-stage design by Chen and Ng was chosen for this study to evaluate the null hypothesis [17]. Stage I targeted 12 eligible and evaluable patients with an accrual range from 8 and 15; if 1 or more out of the 12 patients had objective tumor response, the study would advance to the second stage. Stage II targeted an overall 22 eligible and evaluable patients with a permitted accrual ranging from 18 to 25. If 3 or more out of the 25 cumulative eligible and evaluable patients at the end of stage II responded, then the study regimen would be considered worthy for further investigation. Chen and Ng design gave this study an overall 90% statistical power to detect a 20% improvement in the probability of objective tumor response, an average 10% type I error rate, and an average 65% probability of early termination when the regimen was uninteresting.
PFS was defined as the duration of time from study entry to time of progression or death, whichever occurs first. PFS was ensored on the date of last contact in patients who were alive and had not progressed. OS was defined as the duration of time from study entry to death or the date of last contact in patients who were alive. Kaplan-Meier method was used to estimate the distributions for PFS and OS. The relationship of PD-L1 expression in tumor specimen with objective response was explored by exact Spearman’s rank correlation tests.
RESULTS
PD-L1 expression
Among the 25 eligible and evaluable patients, 22 patients had sufficient tumor tissue to be tested for PD-L1 expression including 15 patients with primary tumor tissue, 3 patients tested in primary metastatic tumor tissue, 1 patient tested recurrent primary tumor tissue and 3 patients tested in recurrent metastatic tumor tissue (Table 2). Membrane staining ≥ 1% in tumor cells was identified in 14 (63.6%) of the patients while 16 (72.7%) of the patients demonstrated PD-L1 expression in immune cells (Table 1). When PD-L1 expression was assessed on both tumor cells and tumor-associated immune cells (i.e., CPS of 1% or greater), a total of 17 patients (77.3%) had quantifiable PD-L1 expression (Table 1).
Table 2:
Patient and tumor characteristics for all eligible and evaluable patients
| Characteristic | Nivolumab | |
|---|---|---|
| N | % | |
| Age (years) | ||
| 20–29 | 1 | 4.0 |
| 30–39 | 7 | 28.0 |
| 40–49 | 8 | 32.0 |
| 50–59 | 5 | 20.0 |
| 60–69 | 3 | 12.0 |
| 70–79 | 1 | 4.0 |
| Country of residence | ||
| United States | 23 | 92.0 |
| India | 1 | 4.0 |
| Russia | 1 | 4.0 |
| Ethnicity | ||
| Hispanic | 2 | 8.0 |
| Non-Hispanic | 22 | 88.0 |
| Unknown/Not Reported | 1 | 4.0 |
| Race | ||
| Black/African American | 5 | 20.0 |
| White | 18 | 72.0 |
| Unknown/Not Reported | 2 | 8.0 |
| Performance Status | ||
| 0 | 16 | 64.0 |
| 1 | 9 | 36.0 |
| Histology | ||
| Adenocarcinoma, Unsp. | 6 | 24.0 |
| Adenosquamous | 4 | 16.0 |
| Squamous Cell Carcinoma | 15 | 60.0 |
| FIGO stage at diagnosis | ||
| I | 9 | 36.0 |
| II | 6 | 24.0 |
| III | 5 | 20.0 |
| IV | 5 | 20.0 |
| Prior Radiation | ||
| Yes | 23 | 92.0 |
| No | 2 | 8.0 |
| Prior Immunotherapy | ||
| Yes | 2 | 8.0 |
| No | 23 | 92.0 |
| Prior Surgery | ||
| Yes | 17 | 68.0 |
| No | 8 | 32.0 |
| AE present at study entry | ||
| Yes | 24 | 96.0 |
| No | 1 | 4.0 |
| Taking concomitant medications at study entry | ||
| Yes | 23 | 92.0 |
| No | 2 | 8.0 |
| Platinum-based prior chemo | ||
| Yes | 22 | 88.0 |
| No | 3 | 12.0 |
| Prior radiation with platinum-based concurrent or adjuvant chemo | ||
| Yes | 23 | 92.0 |
| No | 2 | 8.0 |
| Tumor specimen type | ||
| Primary tumor | 15 | 60.0 |
| Primary metastatic tumor | 3 | 12.0 |
| Recurrent primary tumor | 1 | 4.0 |
| Recurrent metastatic tumor | 3 | 12.0 |
| Not available | 3 | 12.0 |
| Total | 25 | 100.0 |
Table 1:
PD-L1 expression by IHC in eligible and evaluable patients with tumor specimens
| Response by RECIST 1.1 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| PD-L1 expression (%) | Increase disease | Partial response | Stable disease | Indeterminate | Total | |||||
| N | % | N | % | N | % | N | % | N | % | |
| PD-L1 tumor cell (%) | ||||||||||
| 0 | 4 | 36.4 | 0 | 0 | 2 | 25.0 | 2 | 100.0 | 8 | 36.4 |
| >= 1% | 7 | 63.6 | 1 | 100.0 | 6 | 75.0 | 0 | 0 | 14 | 63.6 |
| PD-L1 immune cell (%) | ||||||||||
| 0 | 3 | 27.3 | 0 | 0 | 2 | 25.0 | 1 | 50.0 | 6 | 27.3 |
| >= 1% | 8 | 72.7 | 1 | 100.0 | 6 | 75.0 | 1 | 50.0 | 16 | 72.7 |
| PD-L1 CPS (%) | ||||||||||
| 0 | 3 | 27.3 | 0 | 0 | 1 | 12.5 | 1 | 50.0 | 5 | 22.7 |
| >= 1% | 8 | 72.7 | 1 | 100.0 | 7 | 87.5 | 1 | 50.0 | 17 | 77.3 |
| Total | 11 | 50.0 | 1 | 4.5 | 8 | 36.4 | 2 | 9.1 | 22 | 100.0 |
Patient characteristics
From May 2015 through June 2016, 26 patients were enrolled onto this study (Table 2). One patient did not receive any study treatment leaving 25 eligible and evaluable patients for efficacy and adverse events. The patients and disease characteristics for all eligible and evaluable patients are summarized in Table 2. The median age was 45 years. Most patients were non-Hispanic or Latino ethnicity, of the white race, with country of residence in United States, or with performance status of 0. The most common histology was squamous cell carcinoma. Twenty percent of patients had FIGO stage III and another 20% had stage IV at initial diagnosis. Over 90% of patients had prior radiation therapy. Among the patients with prior radiation therapy, all of them had platinum-based concurrent or adjuvant chemotherapy. Ninety-two percent of patients did not have prior immunotherapy. All 25 eligible and evaluable patients received one prior systemic chemotherapy regimen, with eighty-eight percent of the patients receiving platinum-based prior chemotherapy.
As of March 5, 2019, all patients were off study treatment; median follow-up for survival status was 32 months (range, 2 – 41.5). There was 1 confirmed PR (4%; 90% CI, 0.4%−22.9%), duration of response 3.8 months, and the PD-L1 expressions in her recurrent metastatic tumor specimen were 1% or greater (i.e., PD-L1 tumor cell 1% and PD-L1 immune cell 10%). Three additional patients demonstrated a reduction in the target lesions at the first 8-week CT scan: however, one patient had a new right inguinal lymph node at that 8-week scan and a new left paratracheal lymph node at 16-week CT scan; the other two patients demonstrated progression by RECIST 1.1 criteria at the time of the 16-week CT scans secondary to a new left supraclavicular lymph node or multiple new lesions plus enlargement of non-target lesion, respectively. These three patients had their primary tumor specimen tested for PD-L1 expressions. Two of them had 1% PD-L1 expression in tumor cell and immune cell, respectively: one had stable disease and the other had increase disease. The third patient had 1% PD-L1 expression in tumor cell and 0% in immune cell with stable disease. No significant correlation was found between PD-L1 expression and tumor response: all 2-sided p-values from Spearman’s correlation tests were great than 0.4 regardless whether PD-L1 expression was measured by IHC in tumor cell, immune cell or CPS in either primary tumor specimen or all of the tumor specimens. The waterfall plot showing distribution of the best percentage change in the sum of target lesion size from baseline for the individual patient is depicted in Figure 1. The proportion of patients with stable disease (SD) was 36% (9/25; 90% CI, 20.2%−54.4%); the median duration of SD was 5.7 months (range, 3.5–12.7). The estimates for median PFS and OS were 3.5 (90% CI: 1.9 – 5.1) and 14.5 months (90% CI: 8.3 – 26.8), respectively (Figure 2); the estimated PFS and OS at 6 months was 16% and 78.4%, respectively (Figure 2).
Figure 1.
A) Waterfall plot showing distribution of the best percentage change in the sum of target lesion size from baseline for an individual patient. The lines (−30 and + 20%) indicate the region with change from baseline that typically represent SD based on RECIST guidelines.
Figure 2.
Kaplan-Meier curves for progression-free survival and overall survival
Adverse Events
Adverse events are assessed by the NCI Common Terminology Criteria for Adverse events (CTCAE) version 4.0. Table 3 summarizes the distribution of patients by the highest grade of overall adverse events and by highest grade within each system organ class of adverse events, regardless of attribution. Overall, 21 (84%) pts experienced a treatment-related adverse event (TRAE), most of which were grade 1 and 2. Six (24 %) pts had grade 3 TRAEs, 1 of whom discontinued nivolumab due to hepatic toxicity. No grade 5 TRAEs occurred, and grade 4 TRAEs occurred in 2 pts: one patient had grade 4 blood bilirubin increased, another had grade 4 serum amylase increased. All reported adverse event terms that had at least one grade 1 or higher adverse event are included in Table 3 within their system organ class.
Table 3:
Distribution of patients by highest grade adverse event without regard to attribution
| Nivolumab (n=25) No. and (%) of Patients by Grade |
|||||
|---|---|---|---|---|---|
| System Organ Class | 1 | 2 | 3 | 4 | 5 |
| Overall Highest Grade | 3 | 6 | 12 | 3 | 0 |
| (12.0) | (24.0) | (48.0) | (12.0) | (0.0) | |
| Blood and lymphatic system disorders | 3 | 8 | 3 | 0 | 0 |
| (12.0) | (32.0) | (12.0) | (0.0) | (0.0) | |
| Cardiac disorders | 2 | 0 | 0 | 0 | 0 |
| (8.0) | (0.0) | (0.0) | (0.0) | (0.0) | |
| Ear and labyrinth disorders | 1 | 0 | 0 | 0 | 0 |
| (4.0) | (0.0) | (0.0) | (0.0) | (0.0) | |
| Endocrine disorders | 3 | 2 | 1 | 0 | 0 |
| (12.0) | (8.0) | (4.0) | (0.0) | (0.0) | |
| Eye disorders | 0 | 1 | 0 | 0 | 0 |
| (0.0) | (4.0) | (0.0) | (0.0) | (0.0) | |
| Gastrointestinal disorders | 12 | 3 | 5 | 0 | 0 |
| (48.0) | (12.0) | (20.0) | (0.0) | (0.0) | |
| General disorders and administration site conditions | 9 | 8 | 1 | 0 | 0 |
| (36.0) | (32.0) | (4.0) | (0.0) | (0.0) | |
| Hepatobiliary disorders | 0 | 0 | 1 | 0 | 0 |
| (0.0) | (0.0) | (4.0) | (0.0) | (0.0) | |
| Infections and infestations | 1 | 5 | 1 | 0 | 0 |
| (4.0) | (20.0) | (4.0) | (0.0) | (0.0) | |
| Injury, poisoning and procedural complications | 1 | 2 | 0 | 0 | 0 |
| (4.0) | (8.0) | (0.0) | (0.0) | (0.0) | |
| Investigations | 8 | 3 | 2 | 3 | 0 |
| (32.0) | (12.0) | (8.0) | (12.0) | (0.0) | |
| Metabolism and nutrition disorders | 8 | 1 | 5 | 0 | 0 |
| (32.0) | (4.0) | (20.0) | (0.0) | (0.0) | |
| Musculoskeletal and connective tissue disorders | 5 | 10 | 1 | 0 | 0 |
| (20.0) | (40.0) | (4.0) | (0.0) | (0.0) | |
| Neoplasms benign, malignant and unspecified | 0 | 0 | 2 | 0 | 0 |
| (0.0) | (0.0) | (8.0) | (0.0) | (0.0) | |
| Nervous system disorders | 9 | 3 | 1 | 0 | 0 |
| (36.0) | (12.0) | (4.0) | (0.0) | (0.0) | |
| Psychiatric disorders | 3 | 2 | 0 | 0 | 0 |
| (12.0) | (8.0) | (0.0) | (0.0) | (0.0) | |
| Renal and urinary disorders | 4 | 2 | 0 | 0 | 0 |
| (16.0) | (8.0) | (0.0) | (0.0) | (0.0) | |
| Reproductive system and breast disorders | 3 | 3 | 1 | 0 | 0 |
| (12.0) | (12.0) | (4.0) | (0.0) | (0.0) | |
| Respiratory, thoracic and mediastinal disorders | 9 | 2 | 0 | 0 | 0 |
| (36.0) | (8.0) | (0.0) | (0.0) | (0.0) | |
| Skin and subcutaneous tissue disorders | 7 | 0 | 0 | 0 | 0 |
| (28.0) | (0.0) | (0.0) | (0.0) | (0.0) | |
| Vascular disorders | 1 | 7 | 1 | 0 | 0 |
| (4.0) | (28.0) | (4.0) | (0.0) | (0.0) | |
DISCUSSION
Patients with recurrent, radiation/chemotherapy-resistant cervical cancer carry a poor prognosis. The development of innovative, effective therapies against recurrent, chemotherapy-resistant disease remains a high priority.
PD-1 and PD-L1 expression on cervical T cells and DCs, respectively, have been associated with high risk-HPV positivity and shown to increase in parallel with increasing cervical intraepithelial neoplasia (CIN) grade [12–14]. Furthermore, increased expression of PD-1 and its ligand PD-L1 correlates with impaired cell-mediated immunity in high-risk HPV-related CIN [12–14]. Finally, PD-1 is expressed in the striking majority of advanced stage cervical squamous and adenocarcinoma of the cervix (i.e. 90%) and by a vast number of infiltrating CD8 T cells [14], suggesting that blocking of PD-1 could have therapeutic potential in cervical cancer patients with recurrent and/or metastatic disease unresponsive to treatment [12–14]. Taken together, these studies strongly validate the importance of the PD-1-PD-L1 pathway for the treatment of patients harboring multiple human tumors, including virally-infected pre-neoplastic cervical lesions and frankly invasive cervical cancer.
The current study was designed by NRG/GOG to test the therapeutic benefit of single-agent nivolumab, a humanized anti-PD-1 antibody currently FDA-approved for the treatment of multiple human tumors including but not limited to melanoma, non-small cell lung cancer (NSCLC), renal cell carcinoma (RCC) and hepatocellular carcinoma, in the setting of treatment failure with one prior regimen for metastatic/recurrent cervical cancer. In this group of pretreated patients nivolumab showed limited activity, with a response rate of 4% (PR/CR) and 36% of patients demonstrating stable disease as best response. No significant correlation was found between PD-L1 expression and objective tumor response to nivolumab. A median OS of 14.5 months was found in this study (NRG GY002). It is not clear why nivolumab was unable to demonstrate a clinical activity similar to that detected in patients with recurrent cervical cancer treated with nivolumab in CheckMate 358 (i.e. 26% (5/19) CR/PR responses) [15] or similar to that of pembrolizumab (i.e. 13.3% and 17% response rate in KEYNOTE-158 and KEYNOTE-028, respectively), since all these studies used RECIST 1.1 criteria to evaluate clinical responses [10,11]. It is worth noting, however, that in the NRG-GY002 trial 3 additional patients demonstrated mixed clinical responses during nivolumab treatment (i.e. regression of target lesions with new enlargement of lymph nodes). Per RECIST 1.1 criteria no histological confirmation of metastatic disease was required/performed to confirm tumor progression in the setting of new lymphadenopathy. Previous immunotherapy trials with immune check point inhibitors have demonstrated that conventional RECIST 1.1 criteria may underestimate the benefit of nivolumab and/or pembrolizumab in up to 10% of treated patients [16]. Supporting this view, all these 3 patients with unconfirmed PR in the target lesions demonstrated a prolonged overall survival (i.e. 14.5, 31.6+ and 13.5 months) suggesting a potential clinical benefit from nivolumab treatment in a subset of patients.
Pembrolizumab has been recently added as a preferred regimen for second line option for treating PDL-1 positive cervical tumors (category 2A). Since previous studies demonstrated PDL-1 expression in the striking majority of advanced stage cervical tumors [14], as well as observation of clinical responses in cervical cancer patients regardless of PDL-1 status [10,11,15], in the current NRG GY002 trial, PDL-1 expression was not required for eligibility.
The safety profile of nivolumab has been studied in multiple previous reports [9–11,15,16]. The present study did not identify any new toxicities or an increased frequency of currently reported toxicities of nivolumab in cervical cancer patients with persistent or recurrent disease. Most of the toxicities were G1 and G2 adverse events with no grade 5 adverse event reported.
The results of this study have to be seen in light of some limitations. The first is related to the small sample size of our trial which may affect the reliability of our results. The second limitation concerns the use of RECISIT 1.1 instead of the immune-related response criteria (irRC), as standard method for the evaluation of tumor response [16]. As briefly discussed above, since RECISIT 1.1 was developed to evaluate response to cytotoxic agents in cancer clinical trials and cannot distinguish pseudo progression from progressive disease, similarly to other trials with immune check point inhibitors [16] it may have underestimated the benefit of immunotherapy treatment in some of our patients.
In conclusion, GY002 study indicates that single agent nivolumab has limited activity in patients with persistent or recurrent cervical cancer. Nivolumab was well tolerated in this patient population. Future trials evaluating synergistic combinations of immune check-point inhibitors may be necessary to increase the clinical response rate in cervical cancer patients.
RESEARCH HIGHLIGHTS.
Cervical cancers are infected with Human Papillomavirus (HPV) making these tumors potentially immunogenic
NRG-GY002 evaluated the activity of single agent nivolumab in persistent or recurrent cervical cancer patients
Nivolumab has an acceptable safety profile but low antitumor activity in cervical cancer
Combinations of immune check-point inhibitors may be necessary to increase clinical responses in cervical cancer
ACKNOWLEDGMENT
We would like to thank BMS for their industry support.
This study was supported by National Cancer Institute grants to NRG Oncology SDMC (U10CA180822) and NRG Oncology Operations (U10CA180868), and U01 CA176067 Grant to AS from the National Institutes of Health.
The following Gynecologic Oncology Group member institutions participated in the primary treatment studies: Yale University, Memorial Sloan Kettering Cancer Center, MD Anderson Cancer Center, California Pacific Medical Center-Pacific Campus, Sutter Medical Center Sacramento, Palo Alto Medical Foundation-Gynecologic Oncology, Christiana Care Health System-Christiana Hospital, Lewis Cancer and Research Pavilion at Saint Joseph’s/Candler, University of Iowa/Holden Comprehensive Cancer Center, Lahey Hospital and Medical Center, Dana-Farber/Harvard Cancer Center, Southwest gynecologic Oncology Associates Inc., Women’s Cancer Center of Nevada, Riverside Methodist Hospital, Lehigh Valley Hospital-Cedar Crest and University of Pennsylvania/Abramson Cancer Center.
Dr. Santin received research grants from Puma, Immunomedics, Gilead, Synthon, Merck, Boehringer-Ingelheim, Genentech and Tesaro and consulting fees from Merck and Tesaro.
Dr. Michael Frumovitz received monies for consultancy positions at Stryker, Biom’Up and Genentech. He received grants/grants pending from Astra Zeneca and Stryker, and also received payment for lectures, including service on speakers bureaus from Stryker.
Dr. Warner Huh reports personal fees from Inovo outside the submitted work.
Dr. Samir Naif Khleif received money for board membership through Advaxis. He also served as consultancy for BioLine Therapeutics, Cancer Panels, IO Biotechnologies, NewLink Genetics, Northwest Biotherapeutics, PDS, Syndax Pharmaceuticals, UbiVac, Incyte, AratingoBio, CanlmGuide, Kahr Medical and McKinsey Medical. He has stock/stock options from Advaxis. He received research support from AstraZeneca, BioLine Therapeutics, Bristol Myers Squibb, Merck, IO Biotechnologies, Lycera, MediImmune, Syndax Pharmaceuticals and KAHR Medical.
Dr. Elena Ratner received monies from Tesaro and Genentech for advisory board membership. She also served as a consultant for Zai Labs, Covidient. She provided expert testimony.
Dr. Roisin O’Cearbhaill reports personal fees from Clovis, Tesaro and GlaxoSmithKline outside the submitted work.
Dr. Amir Jazaeri served as a consultancy for Gerson and Lehrman Group, Guidepoint, and Iowance Advisory Board Meeting. Dr. Jazaeri also has grants/grants pending from AstraZeneca, BMS, Iovance, Aravive, Pfizer and Immatics USA for clinical trial support. Dr. Jazaeri also received money from AstraZeneca for travel to Investigator’s meeting.
Footnotes
CONFLICT OF INTEREST
All other co-authors have no conflicts of interest to declare.
NCT#: NCT02257528
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018; 68(1):7–30 [DOI] [PubMed] [Google Scholar]
- 2.Monk BJ, Sill MW, McMeekin DS, et al. Phase III trial of four cisplatin-containing doublet combinations in stage IVB, recurrent, or persistent cervical carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2009; 27:4649–4655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tewari KS, Sill MW, Long HJ 3rd, Penson RT, Huang H, Ramondetta LM, et al. Improved survival with bevacizumab in advanced cervical cancer. N Engl J Med. 2014; 370(8):734–743 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Santin AD, Hermonat PL, Ravaggi A, Chiriva-Internati M, Zhan D, Pecorelli S, et al. Induction of human papillomavirus-specific CD4(+) and CD8(+) lymphocytes by E7-pulsed autologous dendritic cells in patients with human papillomavirus type 16- and 18-positive cervical cancer. J Virol. 1999; 73(7):5402–5410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Santin AD, Bellone S, Gokden M, Cannon MJ, Parham GP. Vaccination with HPV18 E7-pulsed dendritic cells in a patient with metastatic cervical cancer. N Engl J Med. 2002; 346:1752–1753. [DOI] [PubMed] [Google Scholar]
- 6.Santin AD, Bellone S, Palmieri M, Ravaggi A, Romani C, Tassi R, et al. HPV16/18 E7-pulsed dendritic cell vaccination in cervical cancer patients with recurrent disease refractory to standard treatment modalities. Gynecol Oncol. 2006; 100(3):469–478. [DOI] [PubMed] [Google Scholar]
- 7.Santin AD, Bellone S, Palmieri M, Zanolini A, Ravaggi A, Siegel ER, et al. Human papillomavirus type 16 and 18 E7-pulsed dendritic cell vaccination of stage IB or IIA cervical cancer patients: a phase I escalating-dose trial. J Virol. 2008; 82(4):1968–1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yao S, Zhu Y, Chen L. Advances in targeting cell surface signaling molecules for immune modulation. Nat Rev Drug Discov. 2013; 12:130–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chen L, Flies DB. Molecular mechanisms of T cell co-stimulation and co-inhibition. Nat Rev Immunol. 2013; 13:227–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Frenel JS, Le Tourneau C, O’Neil B, Ott PA, Piha-Paul SA, Gomez-Roca C, et al. Safety and Efficacy of Pembrolizumab in Advanced, Programmed Death Ligand 1-Positive Cervical Cancer: Results from the Phase Ib KEYNOTE-028 Trial. J Clin Oncol. 2017; 35(36):4035–4041. [DOI] [PubMed] [Google Scholar]
- 11.Chung HC, Schellens JHM, Delord JP, Perets R, Italiano A, Shapira-Frommer R, et al. Pembrolizumab treatment of advanced cervical cancer: Updated results from the phase 2 KEYNOTE-158 study. J Clin Oncol. 2018. ABS.5522 ASCO. [DOI] [PubMed] [Google Scholar]
- 12.Yang W, Song Y, Lu YL, Sun JZ, Wang HW. Increased expression of programmed death (PD)-1 and its ligand PD-L1 correlates with impaired cell-mediated immunity in high-risk human papillomavirus-related cervical intraepithelial neoplasia. Immunology. 2013; 139(4):513–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Karim R, Jordanova ES, Piersma SJ Kenter GC, Chen L, Boer JM, et al. Tumor-expressed B7-H1 and B7-DC in relation to PD-1+ T-cell infiltration and survival of patients with cervical carcinoma. Clin Cancer Res. 2009; 15:6341–6347. [DOI] [PubMed] [Google Scholar]
- 14.Meng Y, Liang H, Hu J, Liu S, Hao X, Wong MSK, et al. PD-L1 Expression Correlates with Tumor Infiltrating Lymphocytes and Response to Neoadjuvant Chemotherapy in Cervical Cancer. J Cancer. 2018; 9(16):2938–2945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hollebecque A, Meyer T, Moore KN, Machiels JH, De Greve J, López-Picazo JM. An open-label, multicohort, phase I/II study of nivolumab in patients with virus-associated tumors (CheckMate 358): Efficacy and safety in recurrent or metastatic (R/M) cervical, vaginal, and vulvar cancers. J Clin Oncol. 2018. ABS.5504 ASCO. [Google Scholar]
- 16.Wolchok JD, Hoos A, O’Day S, Weber JS, Hamid O, Lebbé C, et al. Guidelines for the Evaluation of Immune Therapy Activity in Solid Tumors: Immune-Related Response Criteria. Clin Cancer Res. 2009; 15(23):7412–7420. [DOI] [PubMed] [Google Scholar]
- 17.Chen TT, Ng TH. Optimal flexible designs in phase II clinical trials. Stat Med. 1998; 17, 2301–2312. [DOI] [PubMed] [Google Scholar]