The efficacy and safety of datopotamab deruxtecan (Dato-DXd) in advanced solid tumors: a systematic review and meta-analysis Authors: Jie Huang, Ting Huang, Junhua Guo, Keke Hu, Heran Zhou Published: 2025 ABSTRACT BackgroundThe cell surface protein TROP-2 is overexpressed in various solid tumors, making it an attractive therapeutic target. Datopotamab deruxtecan (Dato-DXd) is a novel antibody–drug conjugate (ADC) designed to selectively deliver cytotoxic agents to TROP-2-expressing cancer cells. It has recently been approved for treating unresectable or metastatic hormone receptor-positive, HER2-negative breast cancer. While preclinical and early phase clinical trials have shown promising efficacy, this meta-analysis aims to provide a comprehensive evaluation of the efficacy and safety of Dato-DXd in patients with advanced solid tumors by synthesizing the available clinical evidence.MethodsWe systematically searched PubMed, Web of Science, EMBASE, and the Cochrane Library for studies published up to February 18, 2025, that investigated Dato-DXd in advanced solid tumors. The primary outcomes were the overall response rate (ORR) and disease control rate (DCR). Secondary outcomes included progression-free survival (PFS), overall survival (OS), and treatment-related adverse events (TRAEs). Data were extracted and pooled for quantitative synthesis.ResultsSix studies involving 1,166 patients with non-small cell lung cancer (NSCLC) or breast cancer who received Dato-DXd monotherapy were included. The pooled ORR was 30.0%, and the pooled DCR was 76.9%. The median PFS and OS were 5.56 months and 12.46 months, respectively. The most common TRAEs were stomatitis (all grades: 55.1%, grade ≥ 3: 6.8%), nausea (all grades: 49.4%, grade ≥ 3: 1.3%), and alopecia (all grades: 38.5%, grade ≥ 3: 0.0%). Interstitial lung disease (ILD), a critical adverse event of interest, occurred in 5.2% of patients (all grades) and 1.7% (grade ≥ 3).ConclusionsDato-DXd demonstrates encouraging antitumor activity and a manageable safety profile in several advanced solid tumors, particularly breast cancer and NSCLC. This synthesis provides a valuable evidence base to guide clinical practice and future trial design. However, as the current evidence is predominantly from early phase and single-arm studies, confirmation through large-scale randomized controlled trials is necessary before Dato-DXd can be widely adopted in clinical practice.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40001-025-03538-8. Introduction Trophoblast cell surface antigen 2 (TROP-2), a surface glycoprotein of the EpCAM family [1], is overexpressed in a wide range of epithelial tumors—breast, lung, cervical, colorectal, esophageal, and gastric cancers—with reported frequencies ranging from 44 to 88.7% [2–4]. TROP-2 promotes tumor cell proliferation, invasion, and metastasis by mediating key intracellular signaling pathways, such as PTEN/PIK3CA/Akt, MAPK/ERK, JAK/STAT, ErbB, TGFβ, and WNT/β-catenin [5, 6]. Consequently, TROP-2 overexpression is frequently associated with a poor prognosis in various malignancies [7, 8], establishing it as a highly promising therapeutic target. This potential has spurred the development of diverse TROP-2-directed modalities, including monoclonal antibodies (mAbs), antibody–drug conjugates (ADCs), and cellular therapies, several of which have demonstrated considerable clinical promise. ADCs are designed to deliver potent cytotoxic agents directly to tumor cells via target-specific antibodies, thereby minimizing damage to healthy tissues [9]. Early generation ADCs, however, were hampered by suboptimal efficacy and significant toxicity, often due to unstable linkers and mismatched pharmacokinetics [10, 11]. Advances in ADC technology—including humanized antibodies, optimized cleavable linkers, and novel cytotoxic payloads [12]—have yielded agents with an improved therapeutic index, solidifying their role as cornerstones of modern oncology. Datopotamab deruxtecan (Dato-DXd) is a novel ADC consisting of a humanized anti-TROP-2 IgG1 monoclonal antibody covalently linked to a topoisomerase I inhibitor (an exatecan derivative) via a tetrapeptide-based cleavable linker, with a drug-to-antibody ratio (DAR) of 4 [13]. Upon binding to TROP-2 on the tumor cell surface, Dato-DXd is internalized, and the cytotoxic payload is released within lysosomes to induce DNA damage and apoptosis [14]. A key feature of Dato-DXd is its ability to exert a potent bystander effect on neighboring tumor cells, even those with low TROP-2 expression while largely sparing normal tissues. This selective targeting strategy enhances the therapeutic window and offers a valuable new option for patients with advanced solid tumors. Following the U.S. FDA approval of sacituzumab govitecan (SG) for metastatic triple-negative breast cancer (mTNBC) [15], Dato-DXd has been approved for unresectable or metastatic hormone receptor-positive, HER2-negative breast cancer (HR +/HER2-BC) [16], marking it as the second TROP-2-directed ADC to enter clinical practice. This milestone has accelerated clinical investigation, with a growing number of trials evaluating Dato-DXd as both monotherapy and in combination regimens across various advanced solid tumors, including those of the lung, urothelial tract, ovary, and endometrium. To date, however, only two randomized controlled trials (RCTs) in non-small cell lung cancer (NSCLC) and HR +/HER2-BC have reported results, which indicate promising efficacy and a manageable safety profile. The current evidence base remains largely composed of single-arm or cohort studies, limiting the robustness and generalizability of the findings. Presently, no systematic review or meta-analysis has comprehensively synthesized the evidence for Dato-DXd. Therefore, this study was conducted to systematically evaluate the efficacy and safety of Dato-DXd in advanced solid tumors, thereby providing a consolidated evidence base to inform clinical practice and guide future research. Methods This systematic review and meta-analysis was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [17]. The study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number CRD420251032671). Search strategy A systematic literature search was conducted in PubMed, Web of Science, EMBASE, and the Cochrane Library for relevant studies published up to February 18, 2025. Search terms included "Dato-DXd", "Datopotamab deruxtecan", "DS-1062a", "DS-1062", and "Datroway". No language restrictions were applied, and conference abstracts were eligible for inclusion. In addition, the reference lists of all included studies were manually screened to identify other potentially relevant publications. Selection criteria The inclusion criteria were: (1) population: patients diagnosed with advanced solid tumors of any subtype; (2) intervention: treatment with Dato-DXd, either as monotherapy or in combination with other agents; (3) study design: prospective clinical trials, including randomized controlled trials (RCTs), cohort studies, and single-arm studies; and (4) outcomes: availability of data on tumor response (objective response rate [ORR] or disease control rate [DCR]), survival (median progression-free survival [mPFS] or median overall survival [mOS]), and/or treatment-related adverse events (TRAEs). Only cohorts receiving Dato-DXd monotherapy were included in the quantitative meta-analysis; studies involving combination therapies were narratively summarized in the discussion due to inconsistent outcome reporting. Studies were excluded if they were: (1) animal studies, in vitro experiments, reviews, meta-analyses, case reports, or letters; (2) lacking outcome data necessary for meta-analysis; (3) duplicate publications; or 4) Unpublished or with inaccessible full text or data. Data extraction Two investigators (J.H. and H.-R.Z.) independently screened the search results and selected studies for inclusion. Any disagreements were resolved through discussion with a third author (T.H.) until a consensus was reached. The following data were extracted from each included study: first author, year of publication, trial name, intervention details, sample size, cancer type, key tumor characteristics, median age, sex distribution, median treatment duration, median follow-up time, and primary (ORR, DCR) and secondary (PFS, OS, TRAEs) outcomes. Risk-of-bias assessment The methodological quality of single-arm and cohort studies was evaluated using the MINORS tool [18]. This instrument assesses eight key items: (i) a clearly stated aim; (ii) inclusion of consecutive patients; (iii) prospective collection of data; (iv) endpoints appropriate to the study aim; (v) unbiased assessment of study endpoints; (vi) a follow-up period sufficiently long for outcomes to occur; (vii) loss to follow-up of less than 5%; and (viii) a prospective calculation of the study sample size. Each item is scored from 0 to 2 (0 = not reported; 1 = reported but inadequate; 2 = reported and adequate). For RCTs, the Cochrane Collaboration's Risk-of-Bias tool was employed [19], evaluating domains, such as random sequence generation, allocation concealment, blinding of participants and personnel, incomplete outcome data, and selective reporting. Two reviewers (J.H. and H.-R.Z.) independently performed these assessments, with any discrepancies adjudicated by the third reviewer (T.H.). Statistical analysis All statistical analyses were performed using the 'meta' package in R software (version 4.3.2). Single-arm meta-analyses were conducted to pool estimates for ORR, DCR, PFS, OS, and the incidence of TRAEs (any grade and grade ≥ 3). The metaprop function was used for proportional data (ORR, DCR, TRAEs). For time-to-event outcomes (PFS and OS), individual patient data (IPD) were reconstructed from published Kaplan–Meier (K–M) curves using WebPlotDigitizer (v4.6) and the iterative algorithm of the IPD from KM R package [20, 21]. The reconstructed IPD from all studies were then pooled, and a combined survival curve was generated using the Kaplan–Meier method (survfit function in R). Between-study heterogeneity was quantified using Cochran's Q test and the I2 statistic. A random-effects model was applied if substantial heterogeneity was present (I2 > 50% or p value for Cochran's Q < 0.10); otherwise, a fixed-effects model was used [22, 23]. Pre-specified subgroup analyses, based on cancer type, dosage, and study design, were performed using χ2 tests, with a two-sided p < 0.05 considered statistically significant. Potential publication bias was assessed using Egger's test, with a p value < 0.05 indicating significant asymmetry. Appraisal of the quality of evidence The overall certainty of the evidence for the primary outcomes was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach [24]. A summary-of-findings table was created using the GRADEpro GDT online tool (https://gradepro.org). Results Search results The initial literature search conducted on February 18, 2025, identified 593 articles. After removing duplicates, 356 records underwent title and abstract screening, which excluded 270 irrelevant studies. A full-text review of the remaining 86 articles identified 14 potentially eligible studies. Of these, 8 studies were excluded due to insufficient data for meta-analysis (see Supplementary Table 1 for reasons). Ultimately, 6 studies—comprising four single-arm studies [25–28] and 2 RCTs [29, 30]—met all inclusion criteria. All included studies evaluated Dato-DXd monotherapy. The study selection process is detailed in the PRISMA flow diagram (Fig. 1), and the characteristics of the included studies are summarized in Table 1.Fig. 1Flow diagram for study inclusionTable 1Characteristics of the studies included in the meta-analysisStudyTrialTrial phaseNOMean age, years, (range)Sex, male (%)HistologyPrevious therapies, median (range)Toshio Shimizu [25]TROPION-PanTumor01Phase I1804 mg/kg: 61 (35–82)6 mg/kg: 62.5 (38–76)8 mg/kg: 64 (31–84)53.3Unresectable advanced/metastatic NSCLC4 mg/kg: 3 (1–9)6 mg/kg: 3 (1–9)8 mg/kg: 3 (1–13)Aditya Bardia [26, 46]TROPION-PanTumor01Phase I85HR +/HER2- BC: 57(33–75)TNBC: 52.5 (32–82)1.2Advanced HR +/HER2– BC and TNBCHR +/HER2- BC: 6 (4–11)TNBC: 4 (1–12)Myung-Ju Ahn [29]TROPION-Lung01Phase III29963.0(26.0–84.0)61.2Stage IIIB/C or IV NSCLCN/AJacob Sands [28]TROPION-Lung05Phase II13761 (29–79)39.4Advanced or metastatic NSCLC with the actionable genomic alterations3 (1–9)David Planchard [27]ICARUS-LUNG01Phase II10060 (26–83)62.0Advanced and/or unresectable NSCLC2 (1–5)Aditya Bardia [30]TROPION- Breast01Phase Ⅲ36556 (29–86)1.4Inoperable or metastatic HR +/HER2– BC3 (1–7)StudyInterventionMedian follow-up, months, (range)Median treatment, months, (range)EndpointsToshio Shimizu [25]Dato-DXd, 4/6/8 mg/kg, q3w4 mg/kg: 15.96 mg/kg: 13.38 mg/kg: 20.64 mg/kg: 4.16 mg/kg: 3.58 mg/kg: 3.3ORR, DCR, PFS, OS, and TRAEsAditya Bardia [26, 46]Dato-DXd, 6 mg/kg, q3wHR +/HER2- BC: 13.7(9–16)TNBC: 19.3(15–25)HR +/HER2- BC: 4.8 (0.7–14.9)TNBC: 4.3 (0.7–21.8)ORR, DCR, PFS, OS, and TRAEsMyung-Ju Ahn [29]Dato-DXd, 6 mg/kg, q3w23.16.0(1–39)ORR, DCR, PFS, OS, and TRAEsJacob Sands [28]Dato-DXd, 6 mg/kg, q3w15.2 (9.1–20.5)4.4 (0.7–20.6)ORR, DCR, PFS, OS, and TRAEsDavid Planchard [27]Dato-DXd, 6 mg/kg, q3w19.42.8ORR, DCR, PFS, and TRAEsAditya Bardia [30]Dato-DXd, 6 mg/kg, q3w10.86.7 (0.7–15.6)ORR, DCR, PFS, and TRAEsNSCLC non-small cell lung cancer, Dato-DXd datopotamab deruxtecan, HR hormone receptor, HER human epidermal growth factor receptor, BC breast cancer, TNBC triple-negative breast cancer, ORR objective response rate, DCR disease control rate, PFS progression-free survival, OS overall survival, TRAEs treatment-related adverse events, N/A not applicable Flow diagram for study inclusion Table 1Characteristics of the studies included in the meta-analysisStudyTrialTrial phaseNOMean age, years, (range)Sex, male (%)HistologyPrevious therapies, median (range)Toshio Shimizu [25]TROPION-PanTumor01Phase I1804 mg/kg: 61 (35–82)6 mg/kg: 62.5 (38–76)8 mg/kg: 64 (31–84)53.3Unresectable advanced/metastatic NSCLC4 mg/kg: 3 (1–9)6 mg/kg: 3 (1–9)8 mg/kg: 3 (1–13)Aditya Bardia [26, 46]TROPION-PanTumor01Phase I85HR +/HER2- BC: 57(33–75)TNBC: 52.5 (32–82)1.2Advanced HR +/HER2– BC and TNBCHR +/HER2- BC: 6 (4–11)TNBC: 4 (1–12)Myung-Ju Ahn [29]TROPION-Lung01Phase III29963.0(26.0–84.0)61.2Stage IIIB/C or IV NSCLCN/AJacob Sands [28]TROPION-Lung05Phase II13761 (29–79)39.4Advanced or metastatic NSCLC with the actionable genomic alterations3 (1–9)David Planchard [27]ICARUS-LUNG01Phase II10060 (26–83)62.0Advanced and/or unresectable NSCLC2 (1–5)Aditya Bardia [30]TROPION- Breast01Phase Ⅲ36556 (29–86)1.4Inoperable or metastatic HR +/HER2– BC3 (1–7)StudyInterventionMedian follow-up, months, (range)Median treatment, months, (range)EndpointsToshio Shimizu [25]Dato-DXd, 4/6/8 mg/kg, q3w4 mg/kg: 15.96 mg/kg: 13.38 mg/kg: 20.64 mg/kg: 4.16 mg/kg: 3.58 mg/kg: 3.3ORR, DCR, PFS, OS, and TRAEsAditya Bardia [26, 46]Dato-DXd, 6 mg/kg, q3wHR +/HER2- BC: 13.7(9–16)TNBC: 19.3(15–25)HR +/HER2- BC: 4.8 (0.7–14.9)TNBC: 4.3 (0.7–21.8)ORR, DCR, PFS, OS, and TRAEsMyung-Ju Ahn [29]Dato-DXd, 6 mg/kg, q3w23.16.0(1–39)ORR, DCR, PFS, OS, and TRAEsJacob Sands [28]Dato-DXd, 6 mg/kg, q3w15.2 (9.1–20.5)4.4 (0.7–20.6)ORR, DCR, PFS, OS, and TRAEsDavid Planchard [27]Dato-DXd, 6 mg/kg, q3w19.42.8ORR, DCR, PFS, and TRAEsAditya Bardia [30]Dato-DXd, 6 mg/kg, q3w10.86.7 (0.7–15.6)ORR, DCR, PFS, and TRAEsNSCLC non-small cell lung cancer, Dato-DXd datopotamab deruxtecan, HR hormone receptor, HER human epidermal growth factor receptor, BC breast cancer, TNBC triple-negative breast cancer, ORR objective response rate, DCR disease control rate, PFS progression-free survival, OS overall survival, TRAEs treatment-related adverse events, N/A not applicable Characteristics of the studies included in the meta-analysis 4 mg/kg: 61 (35–82) 6 mg/kg: 62.5 (38–76) 8 mg/kg: 64 (31–84) 4 mg/kg: 3 (1–9) 6 mg/kg: 3 (1–9) 8 mg/kg: 3 (1–13) HR +/HER2- BC: 57(33–75) TNBC: 52.5 (32–82) HR +/HER2- BC: 6 (4–11) TNBC: 4 (1–12) 4 mg/kg: 15.9 6 mg/kg: 13.3 8 mg/kg: 20.6 4 mg/kg: 4.1 6 mg/kg: 3.5 8 mg/kg: 3.3 HR +/HER2- BC: 13.7(9–16) TNBC: 19.3(15–25) HR +/HER2- BC: 4.8 (0.7–14.9) TNBC: 4.3 (0.7–21.8) NSCLC non-small cell lung cancer, Dato-DXd datopotamab deruxtecan, HR hormone receptor, HER human epidermal growth factor receptor, BC breast cancer, TNBC triple-negative breast cancer, ORR objective response rate, DCR disease control rate, PFS progression-free survival, OS overall survival, TRAEs treatment-related adverse events, N/A not applicable Risk-of-bias assessment Among the four single-arm studies, all clearly stated their objectives, employed appropriate endpoints, and reported adequate follow-up periods exceeding 12 months. However, one study [27] had unclear consecutive patient inclusion, did not specify prospective data collection, provided unclear evidence of blinded endpoint assessment, and reported a loss to follow-up exceeding 5%. Another study [26, 46] failed to report the loss to follow-up rate. In addition, three studies [25, 26, 28, 46] did not explicitly confirm prospective data collection. Overall, three studies were rated as high quality and one [27] as intermediate quality (Supplementary Table 2). For the two RCTs [29, 30], the open-label design introduced a high risk of performance bias due to the lack of blinding for participants and personnel. However, both trials utilized blinded independent central review (BICR) for endpoint assessment, mitigating detection bias. One trial [29] did not explicitly describe allocation concealment. All other risk-of-bias domains were judged as low risk (Supplementary Table 3). Objective response rate All six studies reported ORR. Substantial heterogeneity was observed (I2 = 59.0%, p = 0.0232), prompting the use of a random-effects model, which yielded a pooled ORR of 30.0% (95% CI 25.6–34.3%). In subgroup analyses, the ORR varied by cancer type, with estimates of 27.5% (95% CI 24.2–30.7%) for lung cancer, 35.3% (95% CI 30.7–40.0%) for HR +/HER2-BC, and 31.8% (95% CI 18.6–47.6%) for TNBC (Fig. 2A). A significant difference was found among these subgroups (χ22 = 7.39, p = 0.0249) (Fig. 2A), suggesting that cancer type is a key source of the observed heterogeneity. This is supported by the low heterogeneity within the lung cancer (I2 = 46%, p = 0.1353) and HR +/HER2-BC (I2 = 41.3%, p = 0.1919) subgroups. When stratified by dose, the ORR was 30.9% (95% CI 26.6–35.2%) at 6 mg/kg, 22.0% (95% CI 11.5–36.0%) at 4 mg/kg, and 23.7% (95% CI 14.95–34.6%) at 8 mg/kg (Fig. 2B); no significant difference was detected between these dose subgroups (χ22 = 3.37, p = 0.1855), indicating that dosage was not a major contributor to heterogeneity. Analysis by study design revealed an ORR of 28.5% (95% CI 24.5–32.4%) for single-arm trials and 31.4% (95% CI 21.6–32.4%) for RCTs (Fig. 2C), a non-significant difference (χ21 = 0.20, p = 0.6584). Notably, substantial heterogeneity was confined to the RCT subgroup (I2 = 87.2%, p = 0.0052). Sensitivity analysis confirmed that the overall heterogeneity was largely driven by one RCT (Aditya Bardia, 2025), as excluding this study led to a marked reduction in heterogeneity (Supplementary Fig. 1 A).Fig. 2Forest plot about the pooled results of ORR by cancer type, dosage, and study design. ORR objective response rate Forest plot about the pooled results of ORR by cancer type, dosage, and study design. ORR objective response rate Disease control rate All six studies reported DCR. A fixed-effect model was employed due to the absence of heterogeneity (I2 = 0.0%, p = 0.7134), yielding a pooled DCR of 76.9% (95% CI 74.5–79.4%). Subgroup analysis by cancer type showed DCRs of 76.9% (95% CI 73.8–80.0%) for lung cancer, 79.1% (95% CI 69.6–88.6%) for HR +/HER2-BC, and 79.5% (95% CI 64.7–90.2%) for TNBC (Fig. 3A), with no significant differences observed between these subgroups (χ22 = 0.34, p = 0.8417). Similarly, analyses of subgroups stratified by either dose or study design revealed no significant differences and were not characterized by substantial heterogeneity (Fig. 3B, C). Finally, sensitivity analysis confirmed that no single study exerted a disproportionate influence on the pooled DCR estimate (Supplementary Fig. 1B), underscoring the robustness of this finding.Fig. 3Forest plot about the pooled results of DCR by cancer type, dosage, and study design. DCR disease control rate Forest plot about the pooled results of DCR by cancer type, dosage, and study design. DCR disease control rate Progression-free survival All studies reported PFS data, with a pooled median PFS of 5.56 months (95% CI 5.39–5.91). Subgroup analysis by cancer type revealed a median PFS of 6.99 months (95% CI 5.66–7.99) for HR +/HER2-BC, 4.92 months (95% CI 4.30–5.50) for lung cancer, and 4.48 months (95% CI 3.01–10.52) for TNBC (Fig. 4A). In addition, a landmark analysis was performed to extract PFS probabilities at 4, 6, and 8 months (Fig. 4B). The overall PFS rate was 63% (95% CI 57–69%) at 4 months, which declined to 46% (95% CI 42–51%) at 6 months and further to 38% (95% CI 34–42%) at 8 months. At each timepoint, the PFS rate was highest in the HR +/HER2-BC subgroup, followed by the lung cancer and TNBC subgroups; detailed data for these timepoints are provided in Supplementary Fig. 2.Fig. 4Pooled results of PFS. A Pooled Kaplan–Meier curves for PFS by cancer type. B Pooled results of PFS rate at 4, 6, and 8 months by cancer type. PFS progression-free survival Pooled results of PFS. A Pooled Kaplan–Meier curves for PFS by cancer type. B Pooled results of PFS rate at 4, 6, and 8 months by cancer type. PFS progression-free survival Overall survival OS data were available from only four of the six eligible trials, as two studies (David Planchard 2024 and Aditya Bardia 2025) were excluded from the pooled analysis due to the reporting of only median OS with 95% CIs without accessible Kaplan–Meier plots or risk tables. Among the four included studies, the median OS for the HR +/HER2-BC cohort in Aditya Bardia 2024 had not been reached. The pooled median OS was 12.46 months (95% CI 11.16–13.37), with tumor-specific estimates of 12.39 months (95% CI 11.05–13.23) for lung cancer and 13.55 months (95% CI 10.22–16.28) for TNBC (Fig. 5A). A landmark analysis of OS probabilities showed that the overall OS rate was 75% (95% CI 71–78%) at 6 months, 52% (95% CI 48–56%) at 12 months, and 35% (95% CI 29–42%) at 18 months (Fig. 5B). The corresponding rates for each cancer subtype at these timepoints are available in Supplementary Fig. 3.Fig. 5Pooled results of OS. A Pooled Kaplan–Meier curves for OS by cancer type. B Pooled results of OS rate at 6, 12, and 18 months by cancer type. OS overall survival Pooled results of OS. A Pooled Kaplan–Meier curves for OS by cancer type. B Pooled results of OS rate at 6, 12, and 18 months by cancer type. OS overall survival Treatment-related adverse events We analyzed the incidence of common TRAEs, graded by CTCAE (Table 2). The incidence of any-grade TRAEs was 94.8% (95% CI 90.8–98.8%), while grade ≥ 3 TRAEs occurred in 28.7% (95% CI 20.4–37.1%) of patients. The most frequent any-grade TRAEs were stomatitis (55.1%; 95% CI 46.7–63.4%), nausea (49.4%; 95% CI 41.9–56.9%), alopecia (38.5%; 95% CI 32.8–44.2%), fatigue (28.1%; 95% CI 19.6–36.5%), and decreased appetite (19.1%; 95% CI 14.6–23.5%). The most common grade ≥ 3 TRAEs were stomatitis (6.8%; 95% CI 4.4–9.3%), anemia (2.9%; 95% CI 1.1–4.8%), interstitial lung disease (ILD) (1.7%; 95% CI 0.5–2.9%), fatigue (1.5%; 95% CI 0.7–2.4%), and asthenia (1.3%; 95% CI 0.5–2.0%) (Supplementary Fig. 4). Table 2TRAE of the studies included in the meta-analysisEndpointsTotalIR, % (95% CI, %)I2,%PTRAEAny grade105994.8 (90.8–98.8)90.3 < 0.0001Grade ≥ 3115928.7 (20.4–37.1)89.1 < 0.0001StomatitisAny grade115955.1 (46.7–63.4)84.5 < 0.0001Grade ≥ 311596.8 (4.4–9.3)61.10.0247NauseaAny grade115949.4 (41.9–56.9)86.7 < 0.0001Grade ≥ 311591.3 (0.5–2.0)7.40.3691AlopeciaAny grade105938.5 (32.8–44.2)69.00.0118Grade ≥ 310590.0 (0.0–0.3)0.00.9553FatigueAny grade76228.1 (19.6–36.5)81.00.0013Grade ≥ 37621.5 (0.7–2.4)0.00.7218Decreased appetiteAny grade105919.1 (14.6–23.5)72.80.0054Grade ≥ 310590.5 (0.0–1.1)0.00.6405AstheniaAny grade79414.1 (9.5–18.7)69.80.0364Grade ≥ 37941.3 (0.5–2.0)36.70.2060AnemiaAny grade92213.8 (9.4–18.2)68.70.0223Grade ≥ 39222.9 (1.1–4.8)66.30.0307ILDAny grade10595.2 (1.9–8.5)82.60.0001Grade ≥ 310591.7 (0.5–2.9)54.00.0692TRAE treatment-related adverse event, ILD Interstitial Lung Disease Table 2TRAE of the studies included in the meta-analysisEndpointsTotalIR, % (95% CI, %)I2,%PTRAEAny grade105994.8 (90.8–98.8)90.3 < 0.0001Grade ≥ 3115928.7 (20.4–37.1)89.1 < 0.0001StomatitisAny grade115955.1 (46.7–63.4)84.5 < 0.0001Grade ≥ 311596.8 (4.4–9.3)61.10.0247NauseaAny grade115949.4 (41.9–56.9)86.7 < 0.0001Grade ≥ 311591.3 (0.5–2.0)7.40.3691AlopeciaAny grade105938.5 (32.8–44.2)69.00.0118Grade ≥ 310590.0 (0.0–0.3)0.00.9553FatigueAny grade76228.1 (19.6–36.5)81.00.0013Grade ≥ 37621.5 (0.7–2.4)0.00.7218Decreased appetiteAny grade105919.1 (14.6–23.5)72.80.0054Grade ≥ 310590.5 (0.0–1.1)0.00.6405AstheniaAny grade79414.1 (9.5–18.7)69.80.0364Grade ≥ 37941.3 (0.5–2.0)36.70.2060AnemiaAny grade92213.8 (9.4–18.2)68.70.0223Grade ≥ 39222.9 (1.1–4.8)66.30.0307ILDAny grade10595.2 (1.9–8.5)82.60.0001Grade ≥ 310591.7 (0.5–2.9)54.00.0692TRAE treatment-related adverse event, ILD Interstitial Lung Disease TRAE of the studies included in the meta-analysis TRAE treatment-related adverse event, ILD Interstitial Lung Disease Publication bias and assessment of the quality of evidence To ensure the validity of the meta-analysis results, we performed Egger's test for all outcome indicators, which confirmed no significant publication bias for any of the evaluated metrics (Supplementary Table 4). Furthermore, the overall quality of evidence for each outcome was evaluated using the GRADE approach, as summarized in a summary-of-findings (SoF) table generated with the GRADE profiler software (Supplementary Fig. 5). According to the GRADE assessment, the certainty of evidence was rated as low for the DCR and very low for the ORR. This rating primarily reflects the inherent limitations of the included body of evidence, which is predominantly composed of single-arm trials that carry a high risk of bias, combined with the significant heterogeneity observed for certain outcomes, which undermines the consistency of the results. Discussion Early studies established that TROP-2 can be recognized by human cytotoxic T lymphocytes (CTLs), which exhibit highly specific cytotoxicity against TROP-2-expressing tumor cells [31]. This rationale spurred the development of anti-TROP-2 monoclonal antibodies (mAbs). However, as naked antibodies, their clinical utility has been limited by modest antitumor activity. The transformation of this therapeutic class has been enabled by ADC technology, which capitalizes on the strong specificity and rapid internalization of anti-TROP-2 mAbs to deliver potent cytotoxic payloads directly to tumor cells [32]. Dato-DXd exemplifies this advance, and its clinical investigation, both as monotherapy and in combination regimens, is now underway across multiple malignancies, including NSCLC, urothelial carcinoma, and gynecologic cancers [33, 34]. This meta-analysis synthesizes data from 1,166 patients with advanced NSCLC and breast cancer across six clinical studies. By focusing exclusively on Dato-DXd monotherapy, we ensured that the efficacy and safety outcomes were not confounded by concomitant treatments. Dato-DXd demonstrated a pooled ORR of 30.0% and a DCR of 76.9%. The observed moderate heterogeneity for ORR was largely explained by tumor type, with significant differences in ORR between HR +/HER2-BC (35.0%), TNBC (31.8%), and NSCLC (27.5%). The notably high ORR in one RCT (Aditya Bardia 2025) was a key contributor to this heterogeneity. In contrast, the DCR was consistently high across tumor types with minimal heterogeneity. Neither dosage nor study design significantly influenced heterogeneity for either outcome. The pooled median PFS of 5.56 months and OS of 12.46 months indicate clinically meaningful antitumor activity in these heavily pretreated populations. The safety profile of Dato-DXd was generally manageable. The overall incidence of TRAEs was high (94.8% for any grade), yet grade ≥ 3 events were less frequent (28.7%). The most common TRAEs were stomatitis (55.1%), nausea (49.4%), and alopecia (38.5%), which were predominantly mild to moderate in severity. With the exception of stomatitis (6.8% grade ≥ 3) and anemia (2.9% grade ≥ 3), the incidence of severe (grade ≥ 3) TRAEs was below 2.0%. Proactive management strategies, including strict oral care and early prophylaxis, can effectively mitigate the incidence and severity of stomatitis [35, 36]. ILD, a known class effect of deruxtecan-based ADCs [37, 38], occurred in 5.2% of patients (any grade) and 1.7% (grade ≥ 3), a rate notably lower than that associated with trastuzumab deruxtecan [39]. Importantly, the toxicity profile of Dato-DXd is distinct from that of sacituzumab govitecan, another TROP-2-directed ADC, which is characterized predominantly by hematologic toxicities, such as neutropenia and anemia [40]. This differentiation may indicate a more favorable therapeutic window for Dato-DXd in clinical practice. This meta-analysis provides evidence supporting the efficacy and safety of Dato-DXd in advanced NSCLC and breast cancer. The two included RCTs provided robust evidence: Aditya Bardia 2025 demonstrated that Dato-DXd significantly improved PFS and showed a trend toward improved OS compared to single-agent chemotherapy. Based on these results, Dato-DXd was approved for treating patients with unresectable or metastatic HR +/HER2- breast cancer. In the only RCT on Dato-DXd for NSCLC to date (Myung-Ju Ahn 2025), Dato-DXd showed significantly higher ORR and DCR than docetaxel, along with a clear advantage in PFS. These findings have prompted a growing portfolio of clinical trials evaluating Dato-DXd in NSCLC, and further evidence is anticipated to support its future clinical application in this setting. Ongoing clinical trials are exploring Dato-DXd both as a monotherapy and in combination regimens across various solid tumors, with some yielding preliminary positive results. However, due to limited follow-up durations, these data remain immature, necessitating further investigation to fully elucidate the benefits of Dato-DXd. As a monotherapy, Dato-DXd has also shown therapeutic potential in other epithelial solid tumors. It demonstrated promising clinical activity in heavily pretreated patients with locally advanced or metastatic urothelial carcinoma [41], achieving 1 complete response (CR) and 4 partial responses (PR) among 18 patients. In addition, in a study involving 75 patients with recurrent, irresectable advanced or metastatic ovarian and endometrial cancers [42], treatment with Dato-DXd resulted in 2 CRs and 24 PRs. In addition to studies on Dato-DXd monotherapy, combination therapies with immunotherapies and chemotherapies are currently under investigation to explore their potential for enhancing therapeutic efficacy. Such studies were excluded from this analysis due to incomplete outcome data. In preclinical studies using syngeneic mouse tumor models, DXd-based ADCs have demonstrated the ability to sensitize tumors to immune-checkpoint inhibitors (ICIs) by eliciting DXd-mediated enhancement of antitumor immunity [43, 44]. Preliminary clinical data suggest that the combination of Dato-DXd with durvalumab, with or without carboplatin, yields objective responses in advanced or metastatic NSCLC [45]. Ongoing trials are evaluating the efficacy and safety of Dato-DXd monotherapy and its combination with durvalumab as neoadjuvant therapy for TNBC, in patients with residual TNBC following neoadjuvant treatment and surgery, as well as in patients with advanced PD-L1-positive TNBC [46–48]. The combination of Dato-DXd with immunotherapy is being explored in first-line settings. Preliminary data indicate the Dato-DXd and durvalumab combination is efficacious in the first-line treatment of advanced or metastatic TNBC [49], reporting 6 CRs and 43 PRs among 62 patients. The TROPION–Lung07 trial is currently evaluating the combination of Dato-DXd with pembrolizumab, with or without platinum-based chemotherapy, as first-line therapy for advanced or metastatic non-squamous NSCLC patients without actionable genomic alterations and with PD-L1 expression below 50% [50]. In contrast, TROPION–Lung08 aims to provide high-level evidence for Dato-DXd combined with pembrolizumab as first-line immunotherapy for advanced or metastatic NSCLC patients without driver mutations and with PD-L1 expression above 50% [51]. While neither trial has reported results yet, their outcomes are anticipated to offer more personalized treatment options for NSCLC patients with varying levels of PD-L1 expression. TROP-2 overexpression is associated with poor prognosis in multiple solid tumors and may predict response to TROP-2-targeted therapy [52, 53]. Preclinical models, including in vitro breast cancer systems, have further demonstrated that high TROP-2 expression enhances the antitumor activity of Dato-DXd. However, the absence of a standardized TROP-2 detection method and a validated "TROP-2-positive" cutoff has complicated patient stratification in both clinical trials and practice [54]. The TROPION–Lung01 study introduced a Quantitative Continuous Scoring (QCS) pathology platform that quantifies TROP-2 spatial distribution by calculating a normalized membrane ratio (NMR) in individual tumor cells. Exploratory analyses defined TROP-2–QCS-positive disease as cases in which ≥ 75% of tumor cells exhibited an NMR ≤ 0.56. This computational methodology helps overcome limitations of conventional immunohistochemistry and manual scoring in predicting Dato-DXd benefit, though related clinical findings have not yet been reported. Subsequently, the TROPION–PanTumor02 study applied the same QCS algorithm to examine the relationship between TROP-2 expression and efficacy in Chinese patients with NSCLC treated with Dato-DXd [55]. Future research is expected to provide a more robust foundation for developing precise, biomarker-driven treatment strategies involving Dato-DXd. Limitations This meta-analysis is subject to several limitations. First, the predominance of non-randomized, single-arm studies, driven by ethical considerations and disease complexity, restricted the analysis to characterizing the efficacy and safety of Dato-DXd without a direct comparator, thereby substantially lowering the certainty of the evidence. Second, considerable heterogeneity was observed across the included studies. Although we performed subgroup analyses by tumor type, dose, and study design, residual heterogeneity may be attributable to other factors, such as prior lines of therapy and differing patient selection criteria. Furthermore, limited data availability confined robust subtype analyses to breast cancer (HR +/HER2- and TNBC), precluding a similar exploration in lung cancer subtypes. Consequently, the sources of heterogeneity can only be hypothesized rather than conclusively determined. Finally, the small number of eligible studies may limit the robustness of the pooled estimates. Conclusion This meta-analysis concludes that Dato-DXd exhibits clinically meaningful antitumor activity in advanced solid tumors. Its safety profile, though characterized by events such as stomatitis that require careful management, remains manageable. Acknowledging the limitations inherent in the current evidence base, this synthesis provides a critical reference point for the design of future definitive trials. The considerable number of ongoing clinical investigations, particularly in NSCLC and breast cancer, underscores the potential of Dato-DXd to establish new therapeutic paradigms and address unmet needs in oncology. Supplementary Information Supplementary material 1: Figure 1. Forest plots of sensitivity analyses for ORR and DCR.Supplementary material 2: Figure 2. Forest plot about the pooled results of PFS rate at 4, 6, and 8 months.Supplementary material 3: Figure 3. Forest plot about the pooled results of OS rate at 6, 8, and 12 months.Supplementary material 4: Figure 4. Forest plot of treatment-related adverse events.Supplementary material 5: Figure 5. GRADE assessment of the primary outcomes.Supplementary material 6: Table 1. Exemplar full-text exclusions. Table 2. Assessment of the quality of included studies according to MINORS. Table 3. Assessment of the quality of randomized controlled trials according to Cochrane Risk of Bias Tool for Randomized Controlled Trials. Table 4. Publication of bias analysis results. Supplementary material 1: Figure 1. Forest plots of sensitivity analyses for ORR and DCR. Supplementary material 2: Figure 2. Forest plot about the pooled results of PFS rate at 4, 6, and 8 months. Supplementary material 3: Figure 3. Forest plot about the pooled results of OS rate at 6, 8, and 12 months. Supplementary material 4: Figure 4. Forest plot of treatment-related adverse events. Supplementary material 5: Figure 5. GRADE assessment of the primary outcomes. Supplementary material 6: Table 1. Exemplar full-text exclusions. Table 2. Assessment of the quality of included studies according to MINORS. Table 3. Assessment of the quality of randomized controlled trials according to Cochrane Risk of Bias Tool for Randomized Controlled Trials. Table 4. Publication of bias analysis results.