To enhance the representativeness and accessibility of the data, we systematically searched the Web of Science (WoS) Core Collection and PubMed. All data were retrieved from these databases on 12 December 2024. The search strategy for WoS was as follows: TS = (“retinitis pigmentosa” OR “pigmentary retinopathy” OR “retinal pigment degeneration”). For PubMed, the search strategy was as follows: (“retinitis pigmentosa” [MeSH Major Topic]). No restrictions were placed on publication date, language, or document type.

Studies were included if they focused on RP, covering one or more aspects of clinical diagnosis, treatment, prevention, basic experimental and biological studies. The following exclusion criteria were applied: (1) duplicate records; (2) publications unrelated to RP; and (3) records lacking essential information such as author, keywords, year of publication, which are critical for analysis.

All retrieved data were downloaded and imported into NoteExpress 4.1.0 for duplicate removal. Two independent medical researchers screened the studies based on the predefined eligibility criteria. Any discrepancies were resolved through discussion, and if necessary, a third-party ophthalmologist provided a final decision. After screening, 8,121 studies were included. The literature selection process is illustrated in Figure 1. The final dataset was exported to Excel to analyze publication volume and the characteristics of highly cited publications, including study type, journal distribution, and scientific area. To conduct a systematic analysis of the conceptual structure and emerging trends, CiteSpace 6.3.R1 and VOSviewer 1.6.20 were used. These tools were selected for their advanced capabilities in network visualization and citation analysis, which are essential for deconstructing complex bibliographic relationships beyond the sorting functions of standard search engines.

Specifically, cluster analysis was used to group related keywords, which helped identify major research hotspots and the evolutionary paths of different sub-fields. The specific parameters set in CiteSpace were as follows: Timespan 2014–2024 with Slice Length set to 1; Selection Criteria g-index (k = 10), LRF = 2.5, L/N = 10, LBY = 5, and e = 1.0; Network N = 264 and E = 1,429 with a Density of 0.0412; Largest 1 CCs 255 (96%); Nodes Labeled 1.0%; Pruning none; Modularity Q = 0.3474; Weighted Mean Silhouette S = 0.6463; and Harmonic Mean (Q, S) = 0.4519. At the same time, burst detection was applied to track keywords showing sharp increases in citation frequency. This measure is important for identifying emerging frontiers and capturing early signals of new scientific directions before they become widely recognized. In addition, co-occurrence analysis was performed to generate density maps that visualize the relationships between gene therapy and oxidative stress. Together, these methods move beyond simple descriptive statistics and help clarify the evolutionary trajectory and potential mechanistic links within the RP research landscape.

Publication volume is a fundamental indicator of research productivity (13). The development trends in RP research were analyzed by examining annual publication volumes (Figure 2). As shown in Figure 2, the number of publications on RP has steadily increased over time, reflecting growing interest in this field. According to publication volume, the evolution of RP research can be divided into four distinct phases: (1) Initial stage (1944–1964): This period marked the early exploration of RP, with limited publications available. (2) Slow exploration stage (1965–1995): Research activity increased gradually, with most publications consisting of case reports. Due to insufficient data, these publications were largely excluded from keyword analysis. (3) Steady growth stage (1996–2013): Research in this phase reached momentum, leading to a more structured development of the field. (4) Rapid development stage (2014–present): Since 2014, RP research has significantly expanded, driven by advancements in biomedical technology and therapeutic approaches.

Citation count is a key indicator of a study’s impact and contributions to the academic community (14). It functions as a quantitative reference for scientific significance and reflects the extent to which a study shapes later research. We confirmed that highly cited studies carry strong scientific value through direct content analysis. Such literature helps pinpoint milestones and shifts in research focus within RP studies. Figure 3 highlights publications with more than 500 citations, revealing a transition in research focus from pathogenesis in earlier studies to treatment approaches in recent years (15–29).

A detailed analysis of the top 100 most-cited articles was conducted based on research types, journals, and scientific fields (Figure 4). This analysis aimed to identify dominant study types, influential journals, and trends in interdisciplinary collaboration. The primary research areas in this field include preclinical studies, such as cellular and animal experiments, and clinical trials, focusing on therapeutic interventions. Among academic journals, Nature Genetics (NAT GENET) has published a substantial number of high-impact publications on RP. Research in this field spans multiple scientific disciplines, reflecting its broad and interdisciplinary nature, with emerging synergies between genetics, bioengineering (e.g., retinal prosthetics), nanomedicine (e.g., drug delivery systems), and artificial intelligence-based diagnostics driving recent advancements. This interdisciplinary convergence has accelerated the development of innovative therapeutic strategies, such as CRISPR-based gene editing and stem cell therapies, while enabling high-throughput screening of potential drug candidates through computational approaches.

To identify emerging research hotspots and trends in RP, we conducted an in-depth keyword analysis of literature published over the past decade. This included clustering analysis, burst detection, and tracking high-frequency keyword trends. The clustering analysis results, displayed in a timeline view (Figure 5), illustrate the major research clusters and their evolution over time. RP research is broadly categorized into two primary areas: disease understanding (shown as clusters #0 retinal degeneration, #1 usher syndrome, #3 stargardt disease, #4 congenital amaurosis) and treatment approaches (shown as clusters #2 cell-based therapy, #5 retinal remodeling, #6 drug discovery strategies). The emergence of these clusters is closely tied to the dominant technologies of each research stage. Clusters #0 retinal degeneration and #1 usher syndrome reflect the Genomic Era, driven by the introduction of Next-Generation Sequencing, which required more precise disease classification. In contrast, cluster #2 cell-based therapy and cluster #5 retinal remodeling reflect the Regenerative Era, supported by advances in stem cell research and bioengineering that address the limits of gene-focused strategies in late-stage disease. This pattern shows that research clusters tend to form around the tools available at each stage of field development.

Burst detection identifies emerging topics with rapidly increasing research activity (11, 30). Figure 6 highlights keywords with strong bursts continuing through 2024, including “natural history,” “optical coherence tomography angiography (OCTA),” “case report,” “retinal organoid” and “cell death.” The rise of these keywords is driven by clear practical needs. The strong burst of OCTA reflects demand from clinical trials for sensitive and non-invasive biomarkers that can measure retinal capillary loss in real time, which goes beyond what traditional fundus photography can provide. Likewise, the growing focus on “retinal organoid” research stems from the need to move past the limits of animal models. Induced pluripotent stem cell technology now supports human-specific drug screening and aligns with the broader shift toward personalized treatment and functional recovery.

Figure 7 displays the top 10 most frequently occurring keywords, along with trends in their usage over time. The analysis reveals that “gene therapy” and “oxidative stress” are not only hot topics in RP research but also exhibit an upward trend, highlighting their growing significance in the field. To explore how these themes are connected, we conducted a keyword co-occurrence density analysis centered on “gene therapy” and “oxidative stress” in Figures 8a, b. The density map for “gene therapy” in Figure 8a shows strong co-occurrence with terms such as “animals”, “humans”, “clinical trials”, as well as “antioxidants” and “oxidative stress”. This pattern suggests that research extends beyond basic concepts and covers the full path from experimental models to clinical use. The strong association between “gene therapy” and “antioxidants” points to a growing interest in supportive treatment strategies. It suggests a growing recognition in the field that the long-term efficacy of therapy relies not only on restoring wild-type gene function but also on modulating the disease microenvironment, specifically by targeting oxidative stress, to achieve disease modification (31). At the same time, the density map for “oxidative stress” in Figure 8b shows close links with “antioxidants”, “ferroptosis”, “inflammation”, and “gene deletion”. In this map, “ferroptosis” and “inflammation” appear alongside “oxidative stress” and “antioxidants”, forming a clear disease and treatment-related subnetwork. From a mechanistic perspective, genetic defects lead to metabolic imbalance and accumulation of reactive oxygen species. This triggers oxidative stress, ferroptosis, and subsequent inflammation through innate immune responses. This interconnected process highlights the need to target oxidative stress, ferroptosis, and inflammation together when designing combination therapies for RP. The frequent appearance of “gene deletion” also reflects the continued reliance on genetic models to confirm these specific molecular targets.

Bibliometric evidence shows that the main driving force of RP research has shifted from early phenomenological observation to intervention-driven innovation. This transition is supported by annual publication trends and changes in the focus of highly cited studies. The shift has been shaped by a combined technology and policy environment. On the technology side, major advances such as CRISPR/Cas9 and adeno-associated virus vectors have played a central role. On the policy side, regulatory milestones, including the approval of Luxturna, have further accelerated progress. The increasing maturity and precision of CRISPR/Cas9 have made it routine to build high-fidelity disease models and perform targeted gene editing, which has greatly shortened the path from basic discovery to preclinical research (32, 33). At the same time, ongoing improvements in the safety, delivery efficiency, and immunogenicity of adeno-associated virus vectors have supported the clinical translation of gene-based therapies (34). In parallel, regulatory events such as the FDA approval of voretigene neparvovec in 2017 opened dedicated review pathways for orphan drugs and gene therapies. This reduced barriers to market entry and encouraged closer collaboration among academic researchers, industry partners, and clinicians (35). As a result, the rapid growth of RP research reflects not just rising academic interest, but a convergence of technical readiness, supportive regulation, and focused investment.

Consistent with this trend, our analysis of journals, research types, and disciplinary backgrounds within highly cited publications suggests that RP research is moving toward a more integrated scientific structure. Interdisciplinary collaboration in this field is no longer a simple layering of different disciplines. Instead, it has developed into a convergent research model that is strongly oriented toward clinical problems. For example, the development of retinal prostheses requires the combined input of microelectronic engineering, neuroscience, materials science, and surgical practice. Overall, the RP research landscape has evolved into a diverse, collaborative, and solution-focused innovation environment. The time-based patterns seen in highly cited publications and keyword clusters also point to a deeper paradigm change. Earlier work followed a reductionist logic centered on single genes or pathways. More recent studies emphasize integrative strategies aimed at systemic disease repair. Early milestone discoveries, such as RPE65 (15) and MERTK (19)] established the genetic foundation of RP but largely remained descriptive in nature. In contrast, recent years have seen a clear shift toward intervention, reflected by clustered advances in cell transplantation and retinal prosthetics. This change is closely linked to the highly heterogeneous nature of RP, where no single target or therapy can address all patient groups. As a result, there is growing agreement that effective management requires a multi-dimensional intervention strategy that combines genetic repair, cellular stress modulation, and functional replacement. This also explains the increasing influence of bioengineering, materials science, and data science in the field.

This paradigm shift has also reshaped research methods and ways of thinking. Earlier studies mainly focused on individual genes and their roles in RP pathogenesis. Current research places more weight on interactions among multiple genes and on the complex regulatory networks involved in disease progression. For instance, omics techniques—such as proteomics and metabolomics—provide insights into alterations at specific molecular levels in RP (36, 37). These approaches collectively contribute to a more comprehensive understanding of the molecular mechanisms underlying the disease. In addition, progress in systems biology has introduced new perspectives by treating the retina as a complex biological system. From this viewpoint, disease onset and progression are explored through interactions between the system and its environment. These integrative strategies help identify promising therapeutic targets and support the development of more effective treatment options.

Burst detection analysis effectively identifies rapidly expanding research frontiers that address key bottlenecks in the field. The results indicate that “optical coherence tomography angiography” and “retinal organoids” have emerged as the fastest-growing research tools in recent years. Their increasing use directly reflects two major challenges in current RP research and clinical practice. One is the lack of methods for dynamic disease monitoring, and the other is the shortage of physiologically relevant experimental models. OCTA enables non-invasive and quantitative evaluation of the retinal microvascular network, allowing researchers to visually track vascular degeneration and hemodynamic changes during disease progression in vivo (38, 39). This capability provides valuable intermediate biomarkers for use in clinical trials. Retinal organoid technology, on the other hand, generates three-dimensional and multicellular retina-like tissues from induced pluripotent stem cells. This approach largely overcomes the limitations of traditional animal models related to species differences and genetic background variability (40, 41). As a result, retinal organoids have become a practical platform for personalized drug screening and studies of disease mechanisms. The growing adoption of these two tools suggests that RP research is entering a new stage characterized by high-resolution dynamic observation and human-relevant precision modeling.

In addition to OCTA and retinal organoids, several other emerging technologies are also beginning to shape RP research. Artificial intelligence shows strong potential in improving RP diagnosis (42). By training artificial intelligence models on large collections of clinical images and patient data, researchers can achieve automated identification and grading of RP phenotypes, which improves both accuracy and efficiency. Artificial intelligence can also support predictions of disease progression and treatment response, laying the groundwork for more personalized therapeutic planning. Nanomedicine represents another important development. Nanoparticles can act as drug carriers that deliver therapeutic agents to the retina with improved precision and efficiency, increasing bioavailability while reducing unwanted side effects (43). At the same time, nanosensor technology is being actively developed for the detection of ocular and systemic biomarkers. For instance, it has already been employed to visualize metabolic dynamics in retinal cells (44). This capability demonstrates the potential for adapting similar approaches to target RP-specific biomarkers in the future.

Despite strong research activity, RP studies still face several interconnected translational challenges. These include marked genetic and phenotypic heterogeneity, insufficient systems for monitoring disease progression and treatment efficacy, narrow therapeutic time windows, and concerns related to the efficiency and safety of delivery systems (45–49). Importantly, the bibliometric results of this study provide an integrated perspective that may help address these complex issues. Our analysis shows that “gene therapy” and “oxidative stress” are not only frequently occurring keywords but are also closely linked in both the timeline and density analyses. This pattern points to a synergistic therapeutic framework that goes beyond single-modality approaches. Genetic interventions aimed at correcting underlying defects can be viewed as hardware repair, while antioxidant or anti-inflammatory strategies help remodel the pathological microenvironment and act as software optimization. From a biological standpoint, even when a genetic defect is corrected, the retinal microenvironment often remains under chronic oxidative stress and inflammation, which can still impair photoreceptor survival and function (50, 51). Taken together, a framework that integrates genetic correction with microenvironment stabilization holds promise for improving therapeutic outcomes in RP.

Further research into the mechanisms linking gene therapy and cellular stress modulation is essential for developing more effective treatments. It is particularly important to clarify how oxidative stress influences the success of gene therapy and to determine the optimal choice, timing, and dosage of antioxidant interventions to achieve the best combined effect. Understanding these synergistic interactions offers a practical way to address several major translational challenges. First, oxidative stress represents a shared downstream pathway across many genetic subtypes of RP. Targeting this pathway with adjunctive therapies such as antioxidants or anti-inflammatory agents may help overcome genotype-specific limitations and provide broad protective benefits for patients with different mutations. Second, for patients in late stages of disease who have already experienced substantial photoreceptor loss, genetic correction alone often has a limited impact. Combining it with neuroprotective or microenvironment-modulating approaches may help preserve remaining cells and slow disease progression, thereby extending therapeutic benefit to a wider patient population. Third, adding antioxidant strategies may reduce local inflammatory responses triggered by gene therapy vectors and improve the cellular environment. This could enhance the survival and functional stability of treated cells and improve overall safety. This synergistic framework represents not only a novel treatment concept but also a structured way to manage the complexity of RP. Future studies should focus on defining the molecular basis of this synergy and identifying optimal combination strategies, including drug selection, timing, and dosage balance. Well-designed clinical trials will be needed to confirm whether this approach can overcome the limits of current single treatments and support a shift from isolated advances toward broader and more sustained benefits in RP care.