Efficacy and safety of tebentafusp in patients with metastatic uveal melanoma: A systematic review and meta-analysis

ABSTRACT

Patients with metastatic uveal melanoma (mUM) have a poor prognosis, and few appropriate medications are available. Tebentafusp is approved by the Food and Drug Administration for mUM recently. However, the real efficacy and safety of tebentafusp are still unclear. We searched PubMed, Embase, and Cochrane Library from inception to March 20, 2024. The research was reported based on the preferred reporting items for systematic reviews and meta-analysis guidelines. We used random effects models to aggregate data on the response rates and adverse events of tebentafusp therapy. Six studies met the inclusion criteria with a total sample of 589 participants. The pooled objective response rate was 0.08 (95% CI: 0.05–0.12), and pooled disease control rate was 0.51 (95% CI: 0.44–0.57). The overall incidence was 0.99 (95% CI: 0.95–1.00) for any grade adverse events, 0.50 (95% CI: 0.41–0.59) for grade 3–4 adverse events, and 0.01 (95% CI: 0–0.03) for discontinuation due to adverse events. Tebentafusp exhibits promising treatment outcomes for mUM patients. Although accompanied with a common occurrence of adverse events, which can typically be managed and controlled. Future research is necessary for substantiating these findings and refining guidelines for management of mUM.

Introduction

Uveal melanoma originates from melanocytes in the stroma and is the most common intraocular malignancies in adults.¹ Although local treatment of primary uveal melanoma can achieve 5-y survival in up to 85% of cases, more than 50% patients will ultimately develop distant metastases during the progression of the disease, affecting principally the liver.^1–3 Consequently, there is an urgent medical need to improve the long-term survival of patients with metastatic uveal melanoma (mUM).

Current treatments for mUM include local surgery, radioembolization, chemoembolization, percutaneous hepatic infusion of melphalan, and systemic chemotherapy and immunotherapy.⁴ However, their clinical benefit has been disappointing, although monoclonal antibodies such as anti-PD(L)1 or anti-CTLA4 have achieved encouraging efficacy in cutaneous melanoma.^4–6 Notably, recent reports have shown that a combination of nivolumab and ipilimumab can improve median overall survival over up to 12 months in patients with uveal melanoma.^7,8 While its safety profile is concerning, as reported in two studies where over 90% patients experienced treatment-related adverse reactions, there is a medical need for more opportunities to explore novel drug with improved efficacy and higher safety profiles.

Tebentafusp was the first drug approved by the Food and Drug Administration (FDA) in 2022 for use in patients with mUM.⁹ It is an immune-mobilizing monoclonal T cell receptor. Nathan et al. reported encouraging efficacy in mUM.¹⁰ However, more clinical evidence is needed to demonstrate the accuracy of the conclusions. To shorten this gap, we performed a meta-analysis of existing clinical trials to explore the real efficacy and safety of tebentafusp.

Methods

We designed and performed this meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines¹¹ and have registered prospectively with Prospero (ID-number CRD42024526675).

Data sources and search strategy

We meticulously examined research that evaluates the efficacy and safety of tebentafusp in treating mUM. Our comprehensive search encompassed renowned databases such as PubMed, Embase, and the Cochrane Library, covering their entire archives up until March 20, 2024. We employed an efficient search strategy, integrating Medical Subject Headings (MeSH) keywords and text-based queries, specifically focusing on “uveal melanoma” and “tebentafusp.” The exhaustive methodology for database exploration is meticulously outlined in Table S1 for transparency and reproducibility.

Study selection

The inclusion criteria for this meta-analysis included the patient was diagnosed with mUM; patients were treated with tebentafusp therapy; and the studies reporting safety or efficacy of tebentafusp, including any of the following outcomes: objective response, partial response, stable disease, minor response, progressive disease, tumor shrinkage, adverse events (any grade, grade 3–4, severe), cytokine release syndromes (any grade, grade 3–4), discontinuation due to adverse events.

The exclusion criteria for this meta-analysis included patients receiving tebentafusp in combination with other immunotherapies or targeted therapies; the study types were commentaries, reviews, or case report; or the study was a secondary analysis of the original cohort. For studies that reported the same cohort cohorts, we chose the one with longer follow-up and more detailed information. Two reviewers (YD and YS) independently screened the titles and abstracts and reviewed the full article, with disagreements resolved by a third researcher (YH L)

Data extraction

Two researchers (YD and YS) independently extracted data from the eligible papers, including first author, year of publication, country or region, sample size, age, sex, follow-up time, route of administration, trial staging, and outcome variables of interest. By cross-checking our work, we ensured the accuracy and consistency of the extracted data, which forms the foundation for our subsequent analysis and interpretation.

Statistical analysis

We used R statistical software (version R 4.2.0) to execute the meta-analysis calculations. Recognizing the inherent heterogeneity across different studies, we employed a random effects model when examining both efficacy and safety data. This approach is advantageous as it takes into account the variability among studies, ensuring a more cautious and robust estimation of the overall effect size compared to a fixed effects model.^12,13 We strive to provide a comprehensive and reliable assessment of tebentafusp’s performance in mUM patients. To synthesize the findings, we employed the Freeman-Tukey double arcsine transformation. This method allowed us to derive point estimates and 95% confidence intervals for both efficacy and safety outcomes. Heterogeneity among studies was rigorously evaluated using the I² index, where values below 50% or p-values greater than or equal to 0.1 indicated low heterogeneity. To understand the sources of any observed heterogeneity, we conducted a sensitivity analysis, examining how changes in individual studies might affect the overall results. Visual representations of these findings were presented in the form of forest plots, which clearly display the effect sizes and their associated confidence intervals for each study, facilitating easy interpretation. Additionally, funnel plots were utilized to assess publication bias. By employing these statistical methods, we aimed to provide a thorough and transparent analysis of the available data on tebentafusp’s impact in mUM treatment.

Results

Study retrieved and characterastics

We obtained a total of 265 citations after literature search. After removing duplicates and downloading the full text, we obtained 207 articles. A total of 201 articles were excluded through screening titles and abstracts and applying the exclusion criteria. A total of 6 articles were eligible for the final meta-analysis. Figure 1 shows the PRISMA flow chart of the meta-analysis. These studies included four clinical trials,^14–17 one retrospective cohort study,¹⁸ and one case-series study,¹⁹ all of which were published in full text. Five studies explored the efficacy and safety of tebentafusp, while one study investigated only the safety of tebentafusp. All included studies involved HLA-A *02:01-positive patients with mUM, and nearly all patients had Eastern Cooperative Tumor Group (ECOG) performance status score of 0 or 1. The median follow-up was 19.5 to 43.3 months. Median overall survival was 16.8 to 25.5 months, and disease-free survival was 3.0 to 4.6 months. Other characteristics are listed in Table 1.

Figure 1.

PRISMA flow diagram for search and selection of studies.

Table 1.

The characteristics of the included studies. RCT: Randomized controlled trial; CI: Confidence interval; (a) Phase I clinical trial; (b) Phase II clinical trial.

Source	Year of publication	Country	Age, average (range)	Sex (male/female)	Study design	Phase	Clinical trial number	Frequency of administration	Administration route	Follow-up	Median overall survival (months)	Progression-free survival (months)
Middleton	2020	Britain	59 (25–78)	54/30	clinical trial	phase I/II	NCT01211262	once weekly	intravenous	/	/	/
Carvajal (a)	2022	American	61 (45–79)	19/23	clinical trial	phase I	NCT02570308	once weekly	intravenous	21 months	25.5 (95% CI: 0.89–31.1)	4.6 (range: 0.7–25.9 months)
Carvajal (b)	2022	American	61 (25–88）	63/64	clinical trial	phase II	NCT02570308	once weekly	intravenous	19.5 months	16.8 (95% CI: 12.9–21.3)	/
Hassel	2023	Germany	64 (23–92)	128/124	RCT	phase III	NCT03070392	once weekly	intravenous	43.3 months	21.6 (95% CI: 19.0–24.3)	3.4 (95% CI: 3.0 to 5.4)
Rodriguez	2023	American	52 (36–76)	3/3	case series	/	/	/	/	/	/	/
Tomsitz	2023	Germany	63 (27–91)	39/39	retrospective Study	/	/	/	/	35 months	22.0 (95% CI: 10.6–33.4)	3.0 (95% CI: 2.7–3.3)

Efficacy

Five studies reported data on the efficacy following tebentafusp therapy in patients with mUM. Figure 2 shows the forest map of the pooled results. The pooled rate was 0.08 (95% CI: 0.05–0.12) for objective response and 0.51 (95% CI: 0.44–0.57) for disease control, and no significant heterogeneity was found. The pooled rates of partial response, stable disease, and progressive disease were 0.08 (95% CI: 0.06–0.11), 0.36 (95% CI: 0.28–0.44) and 0.45 (95% CI: 0.34–0.56), respectively. The significant heterogeneity was detected in stable disease (I²  = 70%, p = .01) and progressive disease (I²  = 85%, p < .01). Additionally, the pooled rate of minor response and tumor shrinkage following tebentafusp treatment was 0.07 (95% CI: 0.04–0.12) and 0.41 (95% CI: 0.33–0.50) (Supplementary material Figure S1-S2).

Figure 2.

Forest plot of pooled incidence of efficacy. CI: confidential interval. (a) Phase I clinical trial; (b) Phase II clinical trial.

Safety

Six studies reported data on the adverse events following tebentafusp therapy in patients with mUM. Figure 3 shows the pooled rates of adverse events. The overall incidence was 0.99 (95% CI: 0.95–1.00) for any grade adverse events, with significant heterogeneity (I² = 76%, p < .01), and 0.50 (95% CI: 0.41–0.59) for grade 3–4 adverse events with significant heterogeneity (I² = 72%, p = .01). Cytokine release syndrome, as one of the major pathologies induced by tebentafusp, had a pooled incidence of 0.80 (95% CI: 0.69–0.89) with significant heterogeneity (I² = 88%, p < .01), while grade 3–4 cytokine release syndrome had a pooled incidence of 0.02 (95% CI: 0.00–0.04) with significant heterogeneity (I² = 53%, p = .07). The pooled rate of discontinuation due to adverse events was 0.01 (95% CI: 0.00–0.03) with insignificant heterogeneity (I² = 0%, p = .94). The pooled rate of severe adverse events was 0.31 (95% CI: 0.26–0.37) with insignificant heterogeneity (I² = 0%, p = .67) (Supplementary material Figure S3).

Figure 3.

Forest plot of pooled incidence of adverse events. CI: confidential interval. (a) Phase I clinical trial; (b) Phase II clinical trial.

Publication bias and sensitivity analysis

In our analysis, funnel plots were employed as a visual tool to evaluate the potential for publication bias. The number of included studies was too small to detect any significant asymmetry in the funnel plot (supplementary materials figures S4). To further investigate the robustness of our findings, we conducted leave-one-out sensitivity analyses. We examined the changes in effect estimates and statistical significance when individual studies are excluded. The results of these sensitivity analyses are presented in separate figures (supplementary materials figures S5 to S17), providing insight into the impact of each study on the overall conclusions.

Discussion

Uveal melanoma, an aggressive form of eye cancer, poses a significant challenge due to its poor prognosis after metastasis.²⁰ Although primary uveal melanoma has made significant advancements in most cases, metastatic recurrence occurs in about 30–50% patients.^1–3,21,22 Metastatic uveal melanoma (mUM) has been linked to a dismal prognosis, lacking a recognized standard treatment approach. It is reported that mUM patients’ overall survival (OS) was approximately 3–12 months in unselected populations.^14,23–26 Traditional treatments targeting skin melanoma, such as PD-1 inhibitors, have proven to be less effective for mUM. However, tebentafusp, an innovative immunomodulatory agent, has gained approval by the FDA for treating HLA-A *02:01-positive patients with mUM. Generally, natural T-cell responses are triggered by the interaction between the T-cell receptor (TCR) and peptide antigens presented by HLA molecules on target cells.^27,28 Nevertheless, the TCR repertoire is limited by thymic selection to T cells with low affinity for self-antigens. To overcome this limitation, bispecific molecules combining antibodies and TCRs have been developed to redirect and activate T cells regardless of their TCR specificity.¹⁷ Tebentafusp is a bispecific fusion protein consisting of a TCR-targeting domain and a single-chain variable fragment (scFv) CD3 effector domain.²⁹ The TCR-targeting domain specifically binds to gp100 presented by HLA-A *02:01. Gp100 is a lineage antigen expressed in melanoma and melanocytes, with strong expression in melanoma and weak expression in normal melanocytes.^30,31 Meanwhile, the CD3 effector domain activates CD3+ T cells, enabling the redirection and activation of T cells to eliminate gp100-expressing tumor cells. This gives tebentafusp the potential to overcome the limitations of immune checkpoint inhibitors in mUM, as the latter necessitate the presence of cancer-specific T cells to initiate cancer cell destruction.

Our meta-analysis revealed that tebentafusp demonstrated a pooled objective response rate of 8% and a disease control rate of 51%. In terms of safety, the pooled incidence of any grade adverse events following tebentafusp treatment was 99%, and the incidence of grade 3–4 adverse events was 50%. It is worth noting that these results highlight the potential for significant adverse events, emphasizing the importance of careful monitoring during therapy. This study represents the first meta-analysis to systematically evaluate the safety and efficacy of tebentafusp for uveal melanoma.

Over the past few decades, treating mUM has been a persistent challenge. Some treatment modalities have been investigated for the treatment of mUM, such as systemic chemotherapy, combined immunotherapy, targeted therapy, and liver-directed therapy.³² The benefits of conventional chemotherapy like dacarbazine, treosulfan, temozolomide, and others in mUM are very limited. Studies have shown that the survival period of mUM patients receiving systemic chemotherapy remains between 2 and 7 months.^32–34 Immune checkpoint inhibitors, initially successful in cutaneous melanoma, have not shown significant improvements in uveal melanoma, with objective response rates as low as 0%⁵ and 3.6%⁶ in two monotherapy trials. The median survival was similar to the survival described in a previous meta-analysis.²³ In the case of combination immunotherapy, prospective⁸ and retrospective studies^35–37 and meta-analyses^23,38 have demonstrated that the combination of nivolumab and ipilimumab did not yield any survival advantage compared to monotherapy or chemotherapy. Targeted therapy is also used in mUM, such as selumetinib, sunitinib, etc. However, multiple studies have indicated that the efficacy of selumetinib is limited and it has shown small impact on overall survival.^39–41 Similarly, no notable improvement in overall survival has been observed with the use of sunitinib.³² The liver is the most frequent site of mUM metastasis. Consequently, local liver-directed treatment approaches have been explored to delay tumor progression and improve survival outcomes. One such promising liver-targeted therapy is chemosaturation by percutaneous hepatic perfusion (CS-PHP). Studies have reported that CS-PHP treatment can extend progression-free survival and overall survival in mUM patients.^42–44 Furthermore, when combined with immunotherapy, CS-PHP has demonstrated enhanced and sustained treatment responses, both within the liver and systemically.⁴⁵ Therefore, local therapy can be appropriately combined with systemic therapy to treat mUM patients. Our meta-analysis demonstrates that tebentafusp achieved a pooled objective response rate of 8% and disease control rate of 51%. The median overall survival range of 16.8 to 25.5 months suggests a meaningful improvement over historical benchmarks, though the absolute numbers may seem low. Hassel et al.'s findings showed hazard ratios of 0.51 and 0.73 for overall and progression-free survival, respectively, indicating a favorable risk-benefit profile compared to standard treatments.¹⁵ Petzold et al.'s population-adjusted analysis further strengthens the case for tebentafusp, revealing that it outperforms chemotherapy, monoclonal antibody therapies, and targeted therapies in terms of both overall and disease-free survival.⁴⁶ This comparative advantage underscores the potential of tebentafusp to significantly extend the lives of patients with mUM.

While tebentafusp can unleash a powerful immune response, it also increases the likelihood of adverse events, primarily due to the release of cytokines, activation of T cells, and its targeting for healthy melanocytes.¹⁵ The high rate of any-grade adverse events with tebentafusp, at 99%, reflects the inherent immunogenic nature of the therapy. By closely monitoring patients during and after the first few treatments, clinicians can identify early signs of cytokine release or other severe side effects, enabling them to administer appropriate interventions like antipyretics, intravenous fluids, or corticosteroids.⁴⁷ It is worth noting that the discontinuation rate due to adverse reactions was only 1%, which is a testament to the effectiveness of these supportive care strategies in mitigating the impact of these events. Hamid et al. found that tebentafusp in combination with durvalumab and/or tremelimumab in patients with metastatic disease in cutaneous melanoma, there was little synergy or exacerbation of adverse events.⁴⁸

Our study has positioned tebentafusp as a promising treatment option for mUM, by presenting a comprehensive assessment of its safety and efficacy. We have compiled essential data, offering a foundation for a deeper comprehension of tebentafusp’s utilization in clinical settings.

There are some limitations in our work. Despite its novelty, our analysis lacks randomized controlled trials, which hinders the availability of robust evidence to fully substantiate the drug’s performance. The absence of raw survival data hindered us from calculating pooled survival figures, leaving us to report median survival rates from various studies, thus preventing a comprehensive understanding of tebentafusp’s duration of response. Furthermore, the limited sample size in our investigation underscores the need for future research with expanded cohorts, enabling more definitive conclusions about the treatment’s efficacy and safety profile. As the field evolves, larger-scale studies will undoubtedly shed more light on tebentafusp’s potential as a life-changing therapy for mUM patients.

Conclusion

Tebentafusp exhibits promising treatment outcomes for mUM patients. Although accompanied with a common occurrence of adverse events, which can typically be managed and controlled. Extensive multi-institutional, longitudinal clinical trials with extended follow-ups are necessary for substantiating these findings and refining guidelines for management of mUM.

Funding Statement

This work was supported by grants from the National Natural Science Foundation of China [82102803 and 82272849 to GT D]; Graduate students independent exploration and innovation project of the Central South University [2024ZZTS0999 to YT D].

Biographical note

Guangtong Deng is a dermatologist and doctoral supervisor in Xiangya Hospital, Central South University. His research expertise is investigating the mechanisms of cell death, as well as the pathogenesis of skin tumors and inflammatory diseases. Deng also holds the esteemed positions of guest editor for Frontier in Medicine and young editorial board member for the Journal of Internal Medicine.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions section

Conception and design: Guangtong Deng and Yuming Sun. Acquisition of data: Yating Dian, Yuming Sun. Interpretation of data, statistical analysis and manuscript writing: Yating Dian, Yuming Sun. Revision of manuscript and administrative, technical, or material support: Furong Zeng, Guangtong Deng, Yuming Sun, Yating Dian, Yihuang Liu.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2374647