Tracing Five Decades of Psoriasis Pharmacotherapy: A Large-Scale Bibliometric Investigation with AI-Guided Terminology Normalization

Abstract

Background/Objectives: Large-scale bibliometric assessments of psoriasis pharmacotherapy research remain limited despite significant research output in this rapidly evolving field. This study aimed to map the evolution of systemic psoriasis therapy research over five decades and demonstrate how systematic analysis of research trajectories can illuminate the transformation of specialized medical fields into central components of precision medicine. Methods: A comprehensive bibliometric analysis was conducted using Web of Science Core Collection as the single data source, examining 19,284 publications spanning 1975-2025. The methodology employed AI-enhanced terminology normalization for standardizing pharmaceutical nomenclature, VOSviewer version 1.6.20 for network visualization, and Bibliometrix package for temporal trend analysis and thematic evolution mapping. International collaboration networks, thematic evolution across three distinct periods (1975-2000, 2001-2010, 2011-2025), and citation impact patterns were systematically analyzed. Results: Four distinct developmental phases were identified, with publications growing from 9 articles in 1975 to 1638 in 2024. The United States dominated research output with 5959 documents, while Canada achieved the highest citation efficiency at 62.65 citations per document. Global collaboration encompassed 70 countries organized into four regional clusters, with a 28-nation Asia-Pacific-Africa-Middle East alliance representing the largest collaborative group. Citation impact peaked during 2001-2008, coinciding with revolutionary biological therapy introduction. Thematic evolution demonstrated systematic transformation from two foundational themes to nine specialized domains, ultimately consolidating into four core areas focused on targeted therapeutics and evidence-based methodologies. Keyword analysis demonstrated progression from basic immunological studies to sophisticated targeted interventions, evolving from tumor necrosis factor alpha inhibitors to contemporary interleukin-17/interleukin-23 pathway targeting and Janus kinase inhibitors. Conclusions: Over five decades, psoriasis therapeutics research has shifted from a niche dermatological discipline to a central model for innovation in immune-mediated diseases. This evolution illustrates how bibliometric approaches can capture the dynamics of scientific transformation, offering strategic insights for guiding pharmaceutical innovation, shaping research priorities, and informing precision medicine strategies across inflammatory conditions.

Keywords: Python; VOSviewer; bibliometric analysis; psoriasis; psoriasis therapy; web of science.

Figure 1

Temporal evolution of scientific publications (number) in systemic psoriasis therapy research (1975–2024).

Figure 2

Citation impact evolution in systemic psoriasis therapy research, mean citations per article (1975–2025).

Figure 3

Heat map illustrating global research leadership trends in the pharmacological management of psoriasis (1975–2025). Each cell represents the number of new publications per country per year, with color intensity (Blues colormap) indicating publication volume. X-axis shows years at 5-year intervals, Y-axis lists countries. Generated using Python 3.12.3 with Seaborn and Matplotlib libraries.

Figure 4

Annual cumulative publications in leading dermatology journals on systemic psoriasis therapies (1975–2025).

Figure 5

Heat map of institutional publication growth trajectories in systemic psoriasis research (1975–2025). Each cell represents publication counts per institution per year, with color intensity indicating publication volume. X-axis shows years at 5-year intervals, Y-axis lists top contributing academic institutions. Generated using Python with Seaborn and Matplotlib libraries.

Figure 6

International collaboration clusters in psoriasis therapeutics research. Colors represent clusters of countries with stronger co-authorship links, as identified by network analysis.

Figure 7

Sankey diagram illustrating thematic evolution in psoriasis therapeutics research across three dis-tinct periods (1975–2000, 2001–2010, 2011–2025). Flow thickness represents the strength of thematic connections between periods, with wider flows indicating stronger persistence or transformation of research themes. Generated using Bibliometrix 5.0.0package in R.

Figure 8

Chronological emergence of therapeutic terminology in psoriasis research. Each dot indicates the first occurrence year of a term in the literature, while the horizontal line represents its period of activity.

Figure 9

Thematic landscape of psoriasis therapeutics research revealed through keyword co-occurrence network analysis. Colors indicate clusters of keywords that frequently co-occur, reflecting distinct thematic areas within psoriasis pharmacotherapy research.

Figure 10

Methodological flowchart depicting the systematic literature search and bibliometric analysis workflow. The diagram illustrates the progression from research focus definition through dual-component search strategy construction, database querying (Web of Science Core Collection), inclusion criteria application (29,386 to 19,284 documents), and final AI-enhanced bibliometric analysis with network visualization generation. The asterisk * denotes the use of wildcard characters in the search query to capture multiple word variants.

Figure 11

The system architecture underlying the pipeline for medical term normalization and synonym resolution. Initially, a list of input terms is subjected to preprocessing and vector representation. A sequential matching framework is then applied, combining exact string comparison, semantic similarity, and fuzzy logic techniques. Identified synonym pairs are integrated into a graph-based structure, from which canonical terms are determined for each connected component. The resulting output is a structured dictionary that links each canonical term to its associated lexical variants.