PD-1 inhibitor plus anlotinib for metastatic castration-resistant prostate cancer: a real-world study

Abstract

Management and treatment of terminal metastatic castration-resistant prostate cancer (mCRPC) remains heavily debated. We sought to investigate the efficacy of programmed cell death 1 (PD-1) inhibitor plus anlotinib as a potential solution for terminal mCRPC and further evaluate the association of genomic characteristics with efficacy outcomes. We conducted a retrospective real-world study of 25 mCRPC patients who received PD-1 inhibitor plus anlotinib after the progression to standard treatments. The clinical information was extracted from the electronic medical records and 22 patients had targeted circulating tumor DNA (ctDNA) next-generation sequencing. Statistical analysis showed that 6 (24.0%) patients experienced prostate-specific antigen (PSA) response and 11 (44.0%) patients experienced PSA reduction. The relationship between ctDNA findings and outcomes was also analyzed. DNA-damage repair (DDR) pathways and homologous recombination repair (HRR) pathway defects indicated a comparatively longer PSA-progression-free survival (PSA-PFS; 2.5 months vs 1.2 months, P = 0.027; 3.3 months vs 1.2 months, P = 0.017; respectively). This study introduces the PD-1 inhibitor plus anlotinib as a late-line therapeutic strategy for terminal mCRPC. PD-1 inhibitor plus anlotinib may be a new treatment choice for terminal mCRPC patients with DDR or HRR pathway defects and requires further investigation.

INTRODUCTION

The management and treatment of terminal metastatic castration-resistant prostate cancer (mCRPC) remains elusive and heavily debated.^1,2 After resistance to standard treatment (next-generation hormonal agents plus docetaxel-based chemotherapy), only modest benefits can be achieved by limited treatment options including poly(ADP-ribose) polymerase inhibitor (PARPi) and radium-223.^1,3–7 Novel late-line treatment is necessary for patients who experience treatment failure.

Medications targeting programmed cell death 1 (PD-1) have exhibited promising antitumor activity and provide better objective responses and disease control rates for mCRPC. However, the benefits remain unclear among unselected patients.^2,8,9 This lack of clarity might result from the heterogeneous nature of prostate cancer (PCa) and the immunologically “cold” tumor microenvironment. PCa has a low infiltration level of cytotoxic CD8⁺ T-cells and an increased presence of immunosuppressive Th17 and Treg cells.^10–12 Thus, PD-1 inhibitors and other systemic medications could be combined to promote responsiveness. Anlotinib, a tyrosine kinase inhibitor (TKI) and anti-vascular endothelial growth factor (VEGF) agent, is recommended as a salvage treatment for various solid tumors.^13,14 This multitarget drug can act as a booster for PD-1 inhibitors and overcome their insufficiency. Anti-VEGF agents have a proven inhibitory effect on the signaling process that promotes Treg population and myeloid-derived suppressor cells (MDSCs).¹⁵ Observational and randomized studies simultaneously applying PD-1 inhibitors and anlotinib have demonstrated promising antitumor efficacy against solid tumors.^16,17

The development of predictive biomarkers is crucial for every emerging therapy and precise medicine. Circulating tumor DNA (ctDNA) sequencing has high accessibility and tumor specificity.¹⁸ Exploratory biomarker analysis has shown that DNA damage repair (DDR) and/or homologous recombination repair (HRR) gene defects can result in better outcomes for therapies targeting immune checkpoints.^9,19,20 Moreover, Chinese patients tend to possess a comparatively higher mutation rate for such genes.²¹ This research suggests that further exploration is needed to identify how individual gene mutation indicates treatment responsiveness.²¹ Further verification of the predictive efficiency of ctDNA sequencing in patients receiving a combination of PD-1 inhibitors and boosters is required.

Currently, few prior studies have explored the efficacy of PD-1 inhibitors plus anlotinib in terminal mCRPC patients. Therefore, it is necessary to elucidate the clinical value of this strategy and to explore the possible predictive biomarkers. Herein, we reviewed 25 patients who had received a combination of a PD-1 inhibitor and anlotinib to determine the clinical benefits and explore the potential predictive biomarkers of this combination therapy.

PATIENTS AND METHODS

Patients and samples

This study was approved by the Committee for Ethics of Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China (Approval No. KY2021-176-B). All patients signed informed consent forms. We included 25 mCRPC patients from February 2018 to August 2021 who received combination therapy with PD-1 inhibitor and anlotinib in this study. All patients included had received standard-of-care therapies according to the NCCN Prostate Cancer Guidelines.⁶ They include next-generation hormonal agents or docetaxel-based chemotherapy as the first- or second-line therapies. All patients gained resistance to prior therapies. Castration resistance was defined according to the European Association of Urology (EAU)-European Association of Nuclear Medicine (EANM)-European Society for Radiotherapy and Oncology (ESTRO)-European Society of Urogenital Radiology (ESUR)-International Society of Geriatric Oncology (SIOG) guidelines for prostate cancer.²² In total, 22 ctDNA samples were collected and we conducted genomic profiling of multiple genes with a hybridization capture-based next-generation sequencing assay. The clinical information was extracted from the electronic medical records.

Measurements

Data from patients who received combination therapy with a PD-1 inhibitor and anlotinib were analyzed in this study. The primary endpoint was prostate-specific antigen (PSA) response. PSA response was defined as a decline of more than 50% in serum PSA from baseline, confirmed with an additional measurement after 3 weeks. The secondary endpoint was PSA progression. Progression-free survival (PFS) was defined as the time from the date of initiation of combination therapy to a confirmed PSA increase, clinical or radiographic progression, or death.

Targeted gene sequencing and bioinformatics

Twenty-two mCRPC patients underwent targeted gene sequencing-based ctDNA analysis. Targeted gene sequencing of all blood samples was performed at GloriousMed Clinical Laboratory Co., Ltd. (Shanghai, China). Sequence data analysis, including identification of germline mutations, somatic mutations, copy number alterations, and quality control; and the list of DDR and HRR genes are described in Supplementary Materials and Methods. Alternations were classified as deleterious when they were nonsense/stop-gains, frameshift insertions and deletions, and splice-site variants, or were previously reported as pathogenic or likely pathogenic in the ClinVar database.

SUPPLEMENTARY MATERIAL AND METHODS

Statistical analyses

All statistical analyses were completed using R software version 3.6.0 (www.R-project.org). The clinical characteristics of the different cohorts were summarized using the descriptive statistics. The Kaplan−Meier method was used to estimate the time to disease progression in different groups of patients, and differences between the groups were analyzed using the log-rank test in the survival package (version 2.44.1.1). Univariate and multivariate Cox regression analyses were used to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) when necessary. Only the factors that were significant according to the univariate analysis were included in the subsequent multivariate analysis. Differences were considered statistically significant at P < 0.05.

RESULTS

Patients’ clinicopathological features

A total of 25 mCRPC patients were included in the present study. All patients received combination therapy with PD-1 inhibitor and anlotinib. Twenty-two patients underwent targeted gene sequencing. The clinical characteristics of the study participants, including median age, Gleason score, and median baseline PSA level at the initiation of combination therapy, are summarized in Table 1.

Table 1: Patient characteristics

PD-1 inhibitor plus anlotinib showed remarkable clinical outcomes in terminal mCRPC patients

PSA response and PSA-PFS to combination therapy with a PD-1 inhibitor and anlotinib were evaluated in all patients. In total, 24.0% (6/25) of the patients with available data achieved a PSA response and 44.0% (11/25) of the patients had PSA reduction after treatment. The median PFS was 1.3 months. The examples of PSA reduction and metastatic lesion changes after treatment are shown in Figure 1. Overall, the median PSA-PFS in patients with a PSA response was 3.9 (range: 2.8−5.2) months, and the median PSA-PFS was 3.4 (range: 2.2−5.2) months in patients with a PSA reduction after treatment.

Figure 1: Example serological and radiological response to combination treatment. (a) Line chart of serum PSA levels from initial administered combined medication. (b) Pelvic MRI showing soft tissue metastatic lesions before and after four courses of combined medication. Left: PSA 236.0 ng ml⁻¹, soft-tissue metastasis, 4.2 cm (maximum diameter); right: PSA 21.5 ng ml⁻¹, soft-tissue metastasis, 1.3 cm (maximum diameter). PSA: prostate-specific antigen; PD-1: programmed cell death 1; MRI: magnetic resonance imaging.

Genomic profiles derived from ctDNA sequencings before treatment

Among the 22 patients who received combination therapy and underwent ctDNA sequencing, 19 (86.4%) had at least one deleterious genomic alteration. The most frequently altered genes were cyclin-dependent kinase 12 (CDK12), tumor protein p53 (TP53), and androgen receptor (AR); and 31 deleterious genomic alterations were detected before treatment, as shown in Figure 2. Overall, 12 patients had defects in the DDR pathway, eight had defects in the HRR pathway, and five had defects in the cell cycle pathway.

Figure 2: The genomic landscape of pathogenic alterations identified before combination treatment. Each column represents alteration detected in individual sample. Upper track shows total number of mutations. Frequencies of specific gene alterations are displayed on the right side.

Clinical benefits of PD-1 inhibitor plus anlotinib observed in patients with DDR/HRR pathway defects

Considering the patients’ genomic profiles, among the five PSA-responsive patients, four (80.0%) were DDR pathway defective; for 11 patients with PSA reduction, six (54.5%) had HRR pathway defects (Figure 3). The observation of PSA-PFS is conducted in all patients. The median time to progression was 1.3 (95% CI: 0.2−2.4) months. The median PSA-PFS was longer in patients with DDR pathway defects (2.5 months, 95% CI: 0.6−4.5 months) than that in those with the wild-type DDR pathway (1.2 months, 95% CI: 1.1−1.3 months, P = 0.027), as shown in Figure 4a. The median PSA-PFS was significantly longer in patients with HRR pathway defects (3.3 months, 95% CI: 1.6−5.0 months) than that in patients with wild-type HRR pathway (1.2 months, 95% CI: 1.1−1.3 months, P=0.017), as shown in Figure 4b. There was no significant difference between the cell cycle pathway defects and the wild-type group (P = 0.952; Figure 4c). Multivariate Cox regression analyses revealed that HRR-defective patients had a significantly reduced risk for progression compared to HRR wild-type patients (HR = 0.25, 95% CI: 0.08−0.79, P = 0.018; Table 2). This reduced risk is reflected in the Kaplan−Meier analyses (Figure 4b), as HRR-defective patients had a significantly prolonged PFS compared to HRR wild-type patients.

Figure 3: The percentage of PSA decline among the patients with mCRPC harboring alterations in genes involved in DDR pathway and HRR pathway. Waterfall plots of best percentage change from baseline in PSA level in the 22 patients with baseline ctDNA information. DDR and HRR pathway defects are labelled as orange and diamonds, respectively. Each column represents the change of PSA after treatment in individual patient. DDR: DNA damage repair; HRR: homologous recombination repair; PSA: prostate-specific antigen; mCRPC: metastatic castration-resistant prostate cancer; ctDNA: circulating tumor DNA.

Figure 4: The survival outcomes of combination therapy in patients with different gene pathway defects. Kaplan−Meier curves and life tables for PSA-PFS of patients with and without (a) DDR pathway defects, (b) HRR pathway defects, and (c) cell cycle pathway defects. WT: wild type; DDR: DNA damage repair; HRR: homologous recombination repair; PSA: prostate-specific antigen; PFS: progression-free survival.

Table 2: Univariate analysis and multivariate analysis of various prognostic parameters in prostate-specific antigen-progression-free survival after anti-programmed cell death 1 plus anlotinib

DISCUSSION

This study included 25 patients who had received standard treatments before this study, but experienced resistance afterward. This study demonstrated the potential benefits of combining a PD-1 inhibitor and anlotinib as a novel approach for terminal mCRPC. Such a combination could result in improved PSA reduction and response treatment outcomes. Statistically significant concordance was observed between the baseline ctDNA sequencing findings and treatment outcomes. Patients with DDR and HRR pathway defects had a comparatively higher PSA response rate and a longer PSA-PFS.

Given the limited activity of single-agent PD-1/programmed death-ligand 1 (PD-L1)-directed therapy in mCRPC, combination studies of immune checkpoint inhibitors with other agents are of growing interest. VEGF acts as a stimulator that induces angiogenesis to ensure sufficient oxygen and nutrients for cancer cells.^16,23 However, emerging evidence has shown that VEGF induces a microenvironment transition to an immunosuppressive state. This altered microenvironment can promote immunosuppressive cytokines, including interleukin-10 (IL-10) and transforming growth factor-β (TGF-β), enhance the expression of inhibitory checkpoints in cytotoxic CD8⁺ T-cells and increase Treg cells.^16,24 Hence, anti-VEGF agents can inhibit the above-mentioned pathway to promote a more immunoresponsive microenvironment.²⁵ Clinical practices that take advantage of this crosstalk have provided promising data. The U.S. Food and Drug Administration (FDA) has approved the application of such a combination in treatment-naïve patients with metastatic renal cell carcinoma²⁶ and unresectable locally advanced or metastatic hepatocellular carcinoma.²⁷ In real-world clinical practice, the above regimen could achieve promising durable antitumor efficacy with acceptable toxicity in various advanced solid tumors.^16,23 PCa is a relatively “cold” cancer featuring low filtration of cytotoxic CD8⁺ T-cells and immunosuppressive Th17 and Treg skewing.²⁸ This characteristic may result in the differential effects of PD-1 inhibitors in unselected mCRPC patients.¹² The synergistic reaction between anti-VEGF agents and PD-1 inhibitors may activate such an unfavorable microenvironment.²⁹ Our study validated the potential of anti-VEGF agents to modify the antitumor immune response in the context of mCRPC. Multiagent approaches backboned by PD-1 inhibitors are efficacious, as shown by a promising PSA deduction, PSA-PFS, and a considerable reduction of lesions, as confirmed by subsequent magnetic resonance imaging (MRI). These findings suggest that the combination of PD-1 inhibitors and anti-angiogenic agents is worthy of consideration for terminal mCRPC.

For every newly introduced treatment, identifying a feasible and reproducible biomarker of response is of importance.¹⁵ Liquid biopsies based on ctDNA has a high concordance with tissue biopsies, and can reflect the ever-changing gene landscape across different stage groups. Thus, it has been introduced by previous studies as a biomarker candidate. This method enables the timely assessment of gene characteristics and frees patients from duplicate invasive biopsies, especially after multiple systemic treatments. DDR is one of the most important subtypes of somatic tumor mutations. This mutation generates non-self-antigen, which confers tumor immunogenicity and induces an antitumor immune response,^30,31 and identifies a high proportion of potential responders.^32,33 Trials applying PD-1 alone have shown that patients with HRR gene-deficient mCRPC can achieve more satisfactory overall response rate (ORR).^9,19 In particular, patients with CDK12 biallelic mutations have a unique immune profile and gene fusion, resulting in sensitivity to PD-1 inhibitors.^20,21 Our exploratory biomarker analysis showed that participants with defects in DDR and/or HRR gene pathways tended to benefit from a longer PSA-PFS compared to the wild type. Patients with these mutations have a higher probability of PSA reduction and response. Moreover, the HHR pathway defects were the only independent prognostic factor for treatment response. Our results further implemented the application of ctDNA sequencing in patients receiving a PD-1 inhibitor plus anlotinib. DDR pathway defects, particularly HRR pathway defects, may act as reliable predictive biomarkers with high efficacy and feasibility, which requires further investigation and validation.

This study had certain limitations. First, this was a single-center real-world study with a small patient population. Some random factors may have been left uneliminated and information bias may sometimes be inevitable. Second, this was a single-arm open-label study. There is a lack of comparisons between combined agents and PD-1 inhibitors alone; therefore, the assessment of the additional benefits of anlotinib remained indirect. In the future, more precisely designed large-scale clinical trials may provide more evidence for this combination therapy. Serological and histological information can be collected to characterize how such therapy modifies immunological microenvironment of PCa. More attention should be paid to how each mutation subtype affects the clinical prognosis and intrinsic mechanism.

In conclusion, this study introduces a novel application of a PD-1 inhibitor plus anlotinib in terminal mCRPC patients. Patients may be granted a decreased serum PSA level and an improved PSA-PFS. Among them, those with DDR and HRR pathway defects had more satisfactory clinical responses. These findings demonstrate a new strategy for terminal mCRPC patients and further implicate the prognostic value of ctDNA assessment.

AUTHOR CONTRIBUTIONS

XXD carried out the methodology, formal analysis, investigation, and original draft preparation. YHD carried out the methodology, formal analysis, and original draft preparation. HJZ carried out the validation. XCF carried out the software management, review, and editing. YMG carried out the software management. BBX carried out the validation. FW carried out the investigation. LCF carried out the project administration. JYW, JZL, YQW, LD, YJZ, and JHP carried out data curation. BJD carried out the conceptualization, resource management, data curation, review and editing, supervision, and project administration. WX carried out the conceptualization and visualization. All the work reported in the paper has been performed by the authors, unless clearly specified in the text. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declare no competing interests.

Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.