Aim and Study design
The purpose of this study is to capitalize on the immunogenic properties of concurrent cisplatin/radiation via addition of pembrolizumab. Addition of pembrolizumab to cisplatin-sensitized radiation (Cis-RT) therapy is hypothesized to increase overall response rate, and recurrence free survival via increased cytotoxic T-cell engagement and activation. Based on this hypothesis, immunotherapy would need to be given concurrently with Cis-RT. Unlike CTLA-4 inhibitors, pembrolizumab has not been shown to have a dose-dependent relationship to toxicity with the currently utilized dose of 200 mg/m2 every 3 weeks. Due to this, a phase 1 trial with varying doses of pembrolizumab would not be instructive regarding toxicity and would likely impact potential efficacy. Also, due to the long half-life and frequency of administration, to test the hypothesis of synergy outlined in this study, pembrolizumab has to be present when Cis-RT is given. This makes a phase 2 trial design with pembrolizumab lead-in less ideal. In light of this, we decided to pursue a single-arm phase 2 clinical trial of concurrent Cis-RT and pembrolizumab. Finally, published and emerging toxicity data from immunotherapy plus chemoradiation combinations in other solid tumors have been reassuring, as described in the discussion section. The first three (3) patients with locally advanced or unresectable disease will receive cycle 1 of pembrolizumab followed by a break and resumption of pembrolizumab at cycle 4 and as part of a safety cohort. These patients will receive a total of 10 cycles of pembrolizumab (figure 1). All other patients, including the fourth patient with locally advanced/unresectable disease, will receive weekly cisplatin and pembrolizumab every 3 weeks, concurrently with daily radiation therapy (figure 2). Following completion of Cis-RT, patients will continue pembrolizumab maintenance for a total of 10 cycles in the safety cohort or 12 cycles for all other participants. Archived tissue will be used for MSI, PD-L1 and TIL stratification post-hoc. Imaging will be performed at baseline and every 3 cycles (21-day cycles) as per standard-of-care. Laboratory analysis will occur on the first day of each cycle. We will collect and evaluate HPV status, PD-L1, and MSI status prior to therapy, but this will not be used for eligibility. Radiographic imaging will be performed every 9 weeks and overall response rate will be determined by RECIST 1.1 criteria.
Patient characteristics
For this study, women ≥18 years with histologically or cytologically confirmed unresectable, incompletely resected, recurrent, or metastatic squamous cell carcinoma of the vulva will be eligible. Patients with unresectable disease defined as having T2 or T3 primary tumors (N0-3, M0) not amenable to surgical resection by standard radical vulvectomy will be eligible. Patients who have received prior chemotherapy, immunotherapy or radiation therapy will also be eligible. For patients who have received prior radiation therapy, re-irradiation to a previously treated site will not be permitted (table 1).
Study procedures
After review and signing of written informed consent and trial registration, participants will be screened for eligibility and participation in the clinical trial. Pembrolizumab may be administered up to 1 day before or after the scheduled Day 1 of each cycle due to administrative reasons. Cisplatin should be given on day 1 of every cycle. At some centers, cisplatin is administered on day 3 for administrative reasons and this will be allowable on study after notification of the PI during enrollment. The first 3 patients receiving definitive chemoradiation will receive pembrolizumab on cycle 1 only and then resume on cycle 4 after completion of chemoradiation (figure 1). This is for the first 3 patients with locally advanced, unresectable disease only. All other patients will be treated with concurrent pembrolizumab-cisplatin-radiation throughout (figure 2). For logistical reasons, patients are allowed to receive standard-of-care cisplatin-radiation locally. For these patients, radiation treatment plans are to be submitted to the radiation oncologist PI. These patients must also be available for AE assessments in person on day 1 of every cycle (pembrolizumab administration days) and weekly AE assessments by telephone. All trial treatments will be administered on an outpatient basis. Radiation therapy can be administered before or after immunotherapy or chemotherapy.
Pembrolizumab 200 mg will be administered as a 30-minute IV infusion every 3 weeks. Pembrolizumab will be administered first, prior to administration of cisplatin. Pembrolizumab will be continued for a total of 12 cycles, except for patients in the safety cohort who will receive a total of 10 cycles. After premedication, cisplatin 40 mg/m2 will be administered as a 30-60 minute IV infusion (depending on institutional guidelines) weekly during radiation therapy. Cisplatin should be administered at the beginning of the week during radiation therapy. To allow for variation of administration of cisplatin at various clinical sites, cisplatin administration up to day +3 will be allowed. This request will be noted prior to patient enrollment. Cisplatin will only be administered during radiation therapy, minimum of 2 weeks, maximum of 8 weeks.
For patients with locally advanced or unresectable disease, external beam radiotherapy will be directed to the vulva and regional lymph nodes as determined by primary tumor location, including but not limited to inguinofemoral, external and internal iliac nodal regions. Treatment will adhere as closely as possible to standard of care. Any deviations will be submitted for approval to the radiation oncology PI prior to initiation of therapy. Patients with localized recurrent and/or metastatic disease with vulvar or inguinal and/or pelvic involvement and no prior history of radiation therapy are also candidates for tumor-directed radiotherapy to the site of recurrence. Patients with recurrent metastatic disease and a history of prior radiation can be considered if the target lesions are outside the prior radiation field, symptomatic, or if in the opinion of the investigator the patient can safely receive and benefit from additional radiation. CT simulation is required to define the gross tumor volume (GTV), clinical target volume (CTV), and planning target volume (PTV). The CT scan must be acquired in the same position and immobilization device as for treatment. The vulvar GTV and the groin GTV when inguinal nodes are unresectable will receive 68-70 Gy in 32-38 fractions of 1.8 - 2.2 Gy per fraction, provided organs at risk (OAR) metrics are met. Dose painting is permitted. Treatment will be delivered once daily, 5 days per week for the treatment duration. Breaks from treatment should be minimized and reasons for breaks must be documented. The groin PTV when high-risk positive nodal features are present after lymph-node dissection will receive 60 Gy in 30 fractions. Uninvolved pelvic or inguinal nodes should receive 45-50.4 Gy. Patients with unresectable / unresected pelvic nodes will receive radiation to a dose of 64 Gy, as OARs allow. For patients with a prior history of radiation, no overlap of previously treated sites will be permitted. Thoracic metastasis will not to be irradiated. Treatment of bone metastases are allowed. Patients who in the opinion of the investigator cannot safely receive a minimum of 30 Gy in 10 fractions for metastatic disease are not eligible for the trial.
Expected toxicities and Management
Toxicity will be graded using CTCAE version 5.0, and treatment criteria will be evaluated prior to each cycle (table 2). Participants will be followed for up to 3 years or until death, whichever occurs first. Participants removed from protocol therapy for unacceptable adverse event(s) will be followed until resolution or stabilization of the adverse event. Survival status will be checked every 6 months during that time. Participants removed from study for unacceptable adverse events will be followed until resolution or stabilization of the adverse event. Expected toxicities from cisplatin include nausea, diarrhea, nephrotoxicity, tinnitus, ototoxicity, neutropenia, and thrombocytopenia. In the cases of grade 3 or 4 nausea, cisplatin would be held, and dose reduced (table 3). Commonly seen adverse events seen with pembrolizumab include immune-mediated pneumonitis, colitis, hepatitis, hypophysitis, hypo/hyperthyroidism, thyroiditis, type 1 diabetes mellitus, nephritis and rashes including; Stephens-Jonson syndrome (SJS) and Toxic epidermal necrolysis (TEN). Similarly, management outlines for neutropenia and thrombocytopenia are also shown in table 3. In the event of suspected or confirmed immune-related adverse events (irAEs) attributed to pembrolizumab, guidelines for management are detailed in table 4. Adverse events from radiation therapy include rashes, wounds, and other cutaneous manifestations, dysuria, and diarrhea.
Severe diarrhea could be from cisplatin, radiation, or pembrolizumab, and precise attribution might be challenging in some cases. To prevent patients receiving definitive therapy from unnecessary treatment breaks and delays due to suspected superimposed immune-related colitis, the first three patients with locally advanced/ unresectable disease will get pembrolizumab on cycle 1 day 1 and omit further pembrolizumab until cycle 4 day 1. If no unexpected toxicities are seen, all remaining patients in this category will receive pembrolizumab uninterrupted every 3 weeks.
Translational studies
Multiplex IHC will be performed on all archival samples for determination of; i) CD3+, CD8+ TILs, CD8+/CD4+FOXP3+ TIL ratio, CD137+CD8+ TILs, CD137+CD8+/CD4+FOXP3+ TIL ratio, peritumoral lymphocytes and correlation with response; ii) expression of immune checkpoints including TIM-3, LAG-3, CTLA-4, PD-L2, PD-L1, PD-1, and correlation with response; iii) targeted next generation T-cell receptor sequencing (Adaptive biotechnologies) to determine T-cell clonality and correlate with response. Multiparameter flow cytometry will be used to determine the effect of treatment on circulating CD8+, CD4+FOXP3+ T-cells. Cytokine analysis will also be performed to measure IFN-a, IFN-b, TNF-a, IL-6, IL-10, IL-12, and IL-2 levels and correlated with response. This will allow for comprehensive baseline profiling that will be correlated with treatment response. This approach has been validated as a means to determine the potential predictive value of immune cell phenotype and spatial distribution relative to the tumor 28. This might allow for a better understanding of patients who might benefit from this form of combined therapy. We will collect HPV status, and MSI status at baseline. Identification of microsatellite instability resulting from defects in the mismatch repair pathway and specifically within MLH1, MSH2, MSH6 and PMS2 have been suggested to portend a significantly improved response to checkpoint blockade 29. This information will be correlated with treatment response. One of the key mechanisms via which cisplatin-sensitized radiation therapy mediates its immunogenic effect is via calreticulin exposure, release of ATP, induction of programmed death receptor 1-ligand (PD-L1) and high-mobility protein box-1 (HMGB-1) 20,21. As such, we will measure baseline serum HMGB1 and at cycle 1, day 1 (D1), D5, D8, D12, D15, D19, D22 and D26 while on combination therapy via commercially available ELISA assays.
Statistical considerations
The target enrolment is 24 patients. The primary endpoint for this study is overall response rate (ORR). This sample size calculation is based on an ORR of ³ 60%. For recurrent metastatic disease, there are is no standard of care. The ORR estimation is informed by data showing an ORR of 40% in patients with advanced or recurrent metastatic vulvar cancer treated with platinum-based combination therapy 4. Single-agent chemotherapy has an ORR of about 12% 5. Patients with primary disease who are not candidates for upfront surgery have a reported response rates from 55% 30 to 64% 11. Only patients who undergo treatment on protocol will be eligible for analysis. Patients who sign consent and do not undergo any treatment will be ineligible for evaluation. With n=24, the power to reject Ho: ORR ≤ 30% in favor of H1: ORR ≥ 60% will be 89% at a target significance level of 0.05 for a one-sided exact Binomial test (12 or more responses are required to reject Ho in favor of H1). Six-month Recurrence Free Survival (RFS-6) is the secondary endpoint. Exploratory biomarkers include; Predictive values of baseline dMMR/MSI-status and PD-L1; Anti-tumor inflammatory responses of cGas-STING- pathway, systemic inflammatory cytokines, and circulating T-cell receptor repertoire.
Analysis Strategy
Exact Binomial test and 95% CI for ORR (primary endpoint). For the secondary objectives of RFS-6, a point estimate with exact 95% CI and Kaplan-Meier estimate will be used for the RFS distribution. Operating characteristics of prediction (sensitivity, specificity, positive- and negative predictive values with 95% CI) for the baseline dMMR/MSI-status and PD-L1, and descriptive analysis will be used for anti-tumor inflammatory cytokines and HMGB-1 levels.