Efficacy and Biomarker-Based Insights of Immunotherapy in Ovarian Cancer: A Retrospective Multi-Dataset Analysis

This retrospective study evaluated the effectiveness of and biomarker-based responses to immunotherapy in ovarian cancer using data from SEER, TCGA, and ClinicalTrials.gov, including 3,750 patients divided into chemotherapy alone, immunotherapy alone, and combination therapy groups. Kaplan-Meier analysis showed that the combination of immunotherapy and chemotherapy resulted in the best overall and progression-free survival outcomes. Biomarker analysis identified PD-L1, CTAG1B, IGSF8, MSI-H, and TMB as potential predictors of an improved response to immunotherapy. Objective response rates were highest among patients with biomarker-positive tumors receiving immunotherapy, while chemotherapy alone had the lowest efficacy and highest toxicity. These findings underscore the clinical potential of immunotherapy, particularly when guided by biomarker profiles, and highlight the need for prospective studies to validate these results and to inform personalized treatment strategies for ovarian cancer.

Introduction

Ovarian cancer is one of the deadliest gynecological malignancies, accounting for a significant proportion of cancer-related mortality among women worldwide. It is estimated that over 300,000 new cases of ovarian cancer are diagnosed annually, with a five-year survival rate below 50% owing to its late-stage presentation and the lack of effective early detection methods [1], [2]. The disease is often asymptomatic in its initial stages, and symptoms that appear, such as abdominal bloating, pelvic discomfort, and changes in bowel habits, are often mistaken for benign conditions, leading to delayed diagnosis [3], [4]. Conventional treatment approaches for ovarian cancer consist of cytoreductive surgery combined with platinum-based chemotherapy, most often carboplatin and paclitaxel [5], [6]. While initial response rates to chemotherapy are high, with 70%–80% of patients showing tumor shrinkage, the majority eventually relapse and develop resistance, making long-term disease management challenging [7]–[10]. The development of poly (ADP-ribose) polymerase (PARP) inhibitors, including olaparib and niraparib, has expanded treatment options, especially for patients with BRCA1/2 mutations and homologous recombination deficiency (HRD) [11], [12]. However, these treatments are not curative and resistance to PARP inhibitors remains a significant issue, necessitating the development of alternative strategies. In recent years, immunotherapy has revolutionized cancer treatment by harnessing the immune system to identify and eliminate cancer cells. A key advancement in this field is the development of immune checkpoint inhibitors (ICIs), which block critical regulatory proteins, such as programmed cell death protein 1 (PD-1), its ligand PD-L1, and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4). These inhibitors have shown significant effectiveness in treating various cancers, including melanoma, non-small cell lung cancer (NSCLC), and renal cell carcinoma [13], [14]. However, the application of ICIs in ovarian cancer has yielded mixed results, with response rates varying significantly among patients [15]–[17]. The ovarian tumor microenvironment plays a crucial role in determining the efficacy of immunotherapy. Many ovarian cancers exhibit an immunosuppressive milieu characterized by low T-cell infiltration, high regulatory T-cell (Treg) activity, and presence of myeloid-derived suppressor cells (MDSCs) [18], [19]. These factors contribute to immune evasion, rendering ICIs less effective compared to their success in other malignancies. Ongoing research continues to focus on discovering predictive biomarkers and refining combination treatment strategies to improve the effectiveness of immunotherapy for ovarian cancer. Several biomarkers have been proposed to predict immunotherapy responses in ovarian cancer. PD-L1 expression is among the most extensively investigated biomarkers, with elevated levels associated with better responses to PD-1/PD-L1 inhibitors in certain studies [20], [21]. However, PD-L1 positivity alone is not a reliable predictor, as many PD-L1-negative tumors respond to ICIs, indicating the involvement of additional immune mechanisms [22]–[24]. Microsatellite instability-high (MSI-H) and tumor mutational burden (TMB) are potential biomarkers associated with increased immunogenicity and immune checkpoint blockade response. MSI-H tumors exhibit a high number of neoantigens, making them more susceptible to immune-mediated destruction [25], [26]. Although ovarian cancer generally has a low TMB compared with other malignancies, a subset of patients with high TMB may benefit from immunotherapy [27], [28]. Other emerging biomarkers include cancer-testis antigens such as CTAG1B, and immune-related genes such as IGSF8, which have been implicated in modulating immune responses in ovarian cancer [29]–[31]. Identifying a robust set of predictive biomarkers remains a key challenge for optimizing patient selection for immunotherapy. Owing to the limitations of immunotherapy as a standalone treatment, researchers are investigating combination strategies to improve its effectiveness in ovarian cancer. One promising strategy is the combination of ICIs with chemotherapy, which can induce immunogenic cell death and enhance tumor antigen presentation, thereby improving the ability of the immune system to recognize cancer cells [32], [33]. Clinical trials, such as JAVELIN Ovarian 100 and KEYNOTE-100, have assessed the effectiveness of combining pembrolizumab (an anti-PD-1 agent) and avelumab (an anti-PD-L1 agent) with chemotherapy, yielding mixed yet promising outcomes [34], [35]. Another approach involves combining ICIs with angiogenesis inhibitors such as bevacizumab, which has been shown to modulate tumor vasculature and improve immune cell infiltration [36]. The IMagyn050 trial assessed the combination of atezolizumab (anti-PD-L1) with bevacizumab and chemotherapy and demonstrated potential benefits in specific patient subgroups. Additionally, adoptive T-cell therapies, such as chimeric antigen receptor (CAR) T-cell therapy and cancer vaccines targeting tumor-specific antigens, are being explored in ovarian cancer immunotherapy [37], [38]. These approaches aim to overcome immune evasion mechanisms and generate durable anti-tumor immunity.

Despite the promising potential of immunotherapy, several challenges remain in its application in ovarian cancer. The immunosuppressive tumor microenvironment, lack of universally predictive biomarkers, and relatively low response rates observed in clinical trials underscore the need for further research [39], [40]. Future research should prioritize improving patient selection, enhancing combination therapy approaches, and identifying new immune targets to boost treatment effectiveness. Innovative tools such as single-cell RNA sequencing and spatial transcriptomics are opening new possibilities for gaining deeper insights into tumor-immune system interactions [41]. These tools can help identify distinct immune phenotypes within ovarian tumors, paving the way for more personalized immunotherapeutic approaches.

The rest of this article is organized as follows. Section 2 outlines the methodology, details of the data sources, criteria for patient selection, and the analytical methods employed. Section 3 details key findings, including survival outcomes, biomarker expression analysis, and treatment-related toxicities. Section 4 discusses the implications of the results in the context of current research and future directions. Finally, Section 5 provides concluding remarks and summarizes the study’s contributions and potential avenues for further investigation.

Methods

This study drew upon data obtained from three publicly accessible databases: Surveillance, Epidemiology, and End Results (SEER) [42], TCGA (The Cancer Genome Atlas) [43], and  ClinicalTrials.gov [44]. Each database provided distinct datasets that were essential for comprehensive analysis of immunotherapy in ovarian cancer. SEER was used to obtain epidemiological and clinical data, including patient demographics, tumor staging, and treatment history. This information allowed the classification of patients into treatment groups based on whether they received chemotherapy alone, immunotherapy alone, or a combination of both. TCGA provided molecular and genomic data, particularly focusing on the biomarker expression levels of PD-L1, CTAG1B, IGSF8, MSI-H, and TMB, which were analyzed to identify potential predictive markers of immunotherapy response.  ClinicalTrials.gov was reviewed to extract data from ongoing and completed clinical trials related to immunotherapy for ovarian cancer, including trial design, patient eligibility criteria, treatment protocols, and reported outcomes. Together, these datasets provide a comprehensive framework for assessing the effectiveness and safety of immunotherapy for ovarian cancer.

This study included 3,750 patients with ovarian cancer, selected based on stringent inclusion and exclusion criteria, to ensure consistency and data integrity. Eligible patients had confirmed Stage III or IV ovarian cancer; documented records for chemotherapy, immunotherapy, or combination therapy; and complete follow-up data on survival and treatment response. Biomarker data (PD-L1, CTAG1B, IGSF8, MSI-H, and TMB) were obtained from TCGA for molecular analysis. Patients with early stage disease, missing data, or concurrent malignancies were excluded. To avoid redundancy, data from SEER, TCGA, and  ClinicalTrials.gov were cross-referenced and de-duplicated, yielding a final dataset of 2,800 SEER cases, 450 from TCGA, and 500 from  ClinicalTrials.gov. All datasets were preprocessed, including formatting, normalization, and imputation, where necessary, to ensure high-quality analysis of clinical and molecular parameters.

Multiple statistical techniques were applied to evaluate treatment impact. Kaplan-Meier curves were constructed to assess differences in overall survival (OS) and progression-free survival (PFS) across treatment groups [45]. A Cox proportional hazards model was used to analyze the effects of variables such as age, tumor stage, BRCA1/2 status, PD-L1 expression, and treatment modality on survival. Biomarker expression patterns were visualized using heatmaps to compare responders and non-responders to immunotherapy, by applying clustering techniques for signature identification. Toxicity severity was analyzed using boxplots across treatment groups, and bar charts were used to compare objective response rates (ORR). Analyses and visualizations were conducted using Python tools—Matplotlib, Seaborn, NumPy, Pandas, Scikit-learn, SciPy, and Statsmodels—providing a comprehensive and structured evaluation of clinical outcomes, biomarker associations, and treatment safety.

This study drew upon data obtained from three publicly accessible databases: Surveillance, Epidemiology, and End Results (SEER) [42], TCGA (The Cancer Genome Atlas) [43], and  ClinicalTrials.gov [44]. Each database provided distinct datasets that were essential for comprehensive analysis of immunotherapy in ovarian cancer. SEER was used to obtain epidemiological and clinical data, including patient demographics, tumor staging, and treatment history. This information allowed the classification of patients into treatment groups based on whether they received chemotherapy alone, immunotherapy alone, or a combination of both. TCGA provided molecular and genomic data, particularly focusing on the biomarker expression levels of PD-L1, CTAG1B, IGSF8, MSI-H, and TMB, which were analyzed to identify potential predictive markers of immunotherapy response.  ClinicalTrials.gov was reviewed to extract data from ongoing and completed clinical trials related to immunotherapy for ovarian cancer, including trial design, patient eligibility criteria, treatment protocols, and reported outcomes. Together, these datasets provide a comprehensive framework for assessing the effectiveness and safety of immunotherapy for ovarian cancer.

This study included 3,750 patients with ovarian cancer, selected based on stringent inclusion and exclusion criteria, to ensure consistency and data integrity. Eligible patients had confirmed Stage III or IV ovarian cancer; documented records for chemotherapy, immunotherapy, or combination therapy; and complete follow-up data on survival and treatment response. Biomarker data (PD-L1, CTAG1B, IGSF8, MSI-H, and TMB) were obtained from TCGA for molecular analysis. Patients with early-stage disease, missing data, or concurrent malignancies were excluded. To avoid redundancy, data from SEER, TCGA, and  ClinicalTrials.gov were cross-referenced and de-duplicated, yielding a final dataset of 2,800 SEER cases, 450 from TCGA, and 500 from  ClinicalTrials.gov. All datasets were preprocessed, including formatting, normalization, and imputation, where necessary, to ensure high-quality analysis of clinical and molecular parameters.

Multiple statistical techniques were applied to evaluate treatment impact. Kaplan-Meier curves were constructed to assess differences in overall survival (OS) and progression-free survival (PFS) across treatment groups [45]. A Cox proportional hazards model was used to analyze the effects of variables such as age, tumor stage, BRCA1/2 status, PD-L1 expression, and treatment modality on survival. Biomarker expression patterns were visualized using heatmaps to compare responders and non-responders to immunotherapy, by applying clustering techniques for signature identification. Toxicity severity was analyzed using boxplots across treatment groups, and bar charts were used to compare objective response rates (ORR). Analyses and visualizations were conducted using Python tools—Matplotlib, Seaborn, NumPy, Pandas, Scikit-learn, SciPy, and Statsmodels—providing a comprehensive and structured evaluation of clinical outcomes, biomarker associations, and treatment safety.

Results

The study population was divided into three treatment groups: chemotherapy alone, immunotherapy alone, and a combination of both, allowing for comparative analysis of treatment effectiveness and patient outcomes. The dataset was structured to ensure that all the included patients had complete records for survival analysis, biomarker expression, treatment response, and adverse event severity. Stratification based on treatment modality enabled the evaluation of distinct survival trends and response patterns, providing insights into the potential benefits and limitations of immunotherapy for ovarian cancer.

Kaplan-Meier survival analysis was performed to evaluate the effects of different treatment strategies by comparing the overall survival (OS) and progression-free survival (PFS) among the three groups. Biomarker expression analysis was performed to explore potential correlations between molecular signatures and immunotherapy responses. Objective response rates (ORR) were compared to determine differences in tumor response among treatment modalities. Additionally, adverse event severity was evaluated to assess treatment tolerability and identify variations in toxicity profiles across groups. The results are presented in the following figures, which illustrate the key findings related to survival outcomes, biomarker expression, treatment response, and adverse event severity.

The Kaplan-Meier survival curves, in Fig. 1, reveal a distinct benefit for patients treated with a combination of immunotherapy and chemotherapy, as shown by the green solid line maintaining the highest overall survival (OS) and progression-free survival (PFS) over time. Immunotherapy alone, represented by the blue dash-dot line, provides a moderate improvement compared to chemotherapy alone, while the red dashed line for chemotherapy shows the steepest decline in survival probability. These findings suggest that combining immunotherapy with chemotherapy may produce a synergistic effect, leading to more durable clinical benefits. This supports the rationale for further clinical research into combination treatment strategies to enhance survival outcomes in ovarian cancer.

Fig. 1. Kaplan-Meier survival curves–ovarian cancer treatment comparison.

Fig. 2 compares objective response rates (ORR) among ovarian cancer patients receiving chemotherapy alone, immunotherapy alone, or a combination of both. The highest ORR (65%) was observed in the combination group, followed by immunotherapy alone (50%), and chemotherapy alone (35%). These results suggest a superior tumor response with combination therapy, likely due to a synergistic effect between the two treatments. Immunotherapy alone also outperforms chemotherapy, supporting its role as a valuable option. The low ORR for chemotherapy highlights its limitations in achieving sustained tumor control. These findings emphasize the need for biomarker-driven strategies to guide optimal treatment selection.

Fig. 2. Objective response rate (ORR) by treatment type.

The heatmap, in Fig. 3, illustrates average biomarker expression levels (PD-L1, CTAG1B, IGSF8, MSI-H, and TMB) across 3,750 ovarian cancer patients, grouped into 30 percentile-based cohorts for clarity. Color gradients range from blue (low expression) to red (high expression), providing a visual summary of molecular patterns within each cohort. The data reveal notable variability in biomarker expression across groups, with some cohorts showing elevated levels of PD-L1 and MSI-H, markers associated with increased immunotherapy sensitivity. Conversely, lower expression in other groups suggests reduced immuno-responsiveness. Variations in TMB also indicate differing mutation loads among patients, a factor linked to immune checkpoint inhibitor efficacy. These heterogeneous patterns underscore the need for biomarker-guided treatment strategies and suggest the existence of distinct molecular subtypes with varying immunotherapy potential. This analysis supports a personalized medicine approach and warrants further correlation with clinical outcomes.

Fig. 3. Grouped biomarker expression patterns in ovarian cancer: immunotherapy response across 3,750 patients.

Fig. 4 compares the severity of treatment-related side effects among ovarian cancer patients across three groups: chemotherapy alone (red), immunotherapy alone (blue), and combination therapy (green). Chemotherapy alone shows the highest median toxicity score (~6), with some cases nearing the maximum severity (~10), reflecting its well-known adverse profile. Immunotherapy alone demonstrates the lowest toxicity burden, with a median score around 3, indicating better tolerability. Combination therapy presents intermediate toxicity (~4.5), suggesting it is more tolerable than chemotherapy but more toxic than immunotherapy alone. These findings confirm that chemotherapy is the most toxic regimen, while immunotherapy offers a safer profile. Combination therapy balances higher efficacy with manageable side effects. This highlights the need for personalized treatment strategies that consider both clinical benefit and patient tolerability, and encourages further optimization of combination regimens to reduce toxicity without compromising effectiveness.

Fig. 4. Side effects severity by treatment type.

Table I contrasts key baseline characteristics across treatment groups. Patients on chemotherapy alone are oldest on average (62 years), while those receiving immunotherapy are slightly younger (58 years), suggesting younger, fitter patients are steered toward novel therapies. Advanced-stage disease (III/IV) is common in all cohorts, peaking at 80 % in the chemotherapy group and reflecting ovarian cancer’s late diagnosis. BRCA1/2 mutations occur most often in the combination group (35 %), followed by immunotherapy (30 %) and chemotherapy (25 %), implying that DNA-repair–deficient tumors may derive extra benefit from combined regimens. These patterns underscore the value of genomic profiling and patient fitness in guiding therapy selection and optimizing outcomes.

Table I. Demographic and Clinical Characteristics

Characteristic	Chemotherapy alone	Immunotherapy alone	Combination Immuno + Chemo
Average age (years)	62	58	60
Advanced Tumor stage (%)	80	70	75
BRCA1/2 Mutation (%)	25	30	35

Table II compares clinical outcomes across three treatment groups in ovarian cancer: chemotherapy alone, immunotherapy alone, and combination therapy. Combination therapy achieves the best results, with the highest ORR (65%), longest median PFS (24 months), and OS (36 months). Immunotherapy alone follows with a 50% ORR, PFS of 18 months, and OS of 30 months, while chemotherapy alone shows the lowest efficacy (ORR 35%, PFS 12 months, OS 24 months). Regarding toxicity, chemotherapy has the highest rate of severe side effects (50%), immunotherapy the lowest (20%), and combination therapy falls in between (35%). These findings suggest that combining chemotherapy with immunotherapy enhances effectiveness but adds moderate toxicity. Immunotherapy alone also outperforms chemotherapy in both efficacy and safety. The results support a tailored approach, balancing benefit and risk through biomarker-guided treatment selection.

Table II. Clinical Outcomes

Outcome	Chemotherapy alone	Immunotherapy alone	Combination Immuno + Chemo
Objective response rate (ORR, %)	35	50	65
Median Progression-Free Survival (PFS, months)	12	18	24
Median overall survival (OS, months)	24	30	36
Severe side effects (%)	50	20	35

Discussion

The results of this study highlight the growing importance of immunotherapy for ovarian cancer, particularly when combined with chemotherapy. Patients receiving combination therapy showed the highest ORR, PFS, and OS, suggesting a synergistic effect supported by previous studies showing that chemotherapy enhances the efficacy of immune checkpoint inhibitors by boosting tumor antigen presentation and immune activation [35], [46]. While immunotherapy alone offers some benefits, its limited effectiveness in ovarian cancer, likely due to the immunosuppressive tumor microenvironment, makes combination strategies more compelling [47]. Chemotherapy-induced immunogenic cell death appears to enhance this response [48], though variable outcomes with immunotherapy alone emphasize the need for precise patient selection [49]. A key finding of this study is the predictive value of biomarkers in immunotherapy response. Patients with high PD-L1 expression, MSI-H status, and elevated tumor mutational burden (TMB) showed better outcomes with immunotherapy, consistent with the literature linking high neoantigen load to improved immune checkpoint blockade response [50], [51]. These results support the need for biomarker-guided patient selection to enhance the efficacy and reduce toxicity in non-responders. However, challenges persist in clinical implementation, such as variability in PD-L1 testing platforms and cut-off thresholds, which can affect patient stratification [52]. Further research is needed to standardize and validate the predictive biomarkers. Although immunotherapy alone had a favorable safety profile, combination therapy led to moderate toxicity, consistent with previous studies suggesting a manageable increase in side effects when immune checkpoint inhibitors were added [46]. Chemotherapy alone remains the most toxic, with common adverse effects, such as myelosuppression and neuropathy [35]. Balancing efficacy and safety will require proactive toxicity management, including dose adjustments and biomarker-informed treatment decisions, to optimize the clinical utility of combination regimens. Despite the encouraging results, challenges remain. Immunotherapy alone showed limited response rates, underscoring the need for alternative combinations, such as PARP inhibitors or anti-angiogenic agents [53], [54]. Genomic tools may improve patient selection, but the long-term efficacy of combination therapy requires further validation through prospective trials. In conclusion, this study supports immunotherapy, particularly in combination with chemotherapy, as a promising approach for the treatment of ovarian cancer. Biomarker-driven strategies and further research are essential to refine treatment, manage toxicity, and fully integrate immunotherapy into standard care.

Conclusion

This study provides a comprehensive retrospective analysis of immunotherapy in ovarian cancer, revealing that its combination with chemotherapy significantly improves overall survival, progression-free survival, and objective response rates compared with chemotherapy alone. While this synergistic approach shows notable clinical benefits, it also introduces moderate toxicity, underscoring the need for careful management of side effects. Biomarker analysis identified key predictors of response: PD-L1 expression, MSI-H status, and high tumor mutational burden, highlighting the importance of personalized, biomarker-driven treatment strategies. Despite the promise of immunotherapy, challenges remain, including tumor microenvironment resistance and variability in patient responses. The findings advocate for further prospective validation and continued development of precision oncology approaches to enhance the outcomes and quality of life of patients with ovarian cancer.