Globo H

• CAS: 207384-96-1
• Molecular Formula: C38H65NO30
• Molecular Weight: 1015.9 g/mol
• InChI: InChI=1S/C38H65NO30/c1-8-16(46)20(50)25(55)35(59-8)69-32-21(51)17(47)10(3-40)63-38(32)67-30-15(39-9(2)45)34(61-11(4-41)18(30)48)68-31-19(49)12(5-42)62-37(27(31)57)66-29-14(7-44)64-36(26(56)23(29)53)65-28-13(6-43)60-33(58)24(54)22(28)52/h8,10-38,40-44,46-58H,3-7H2,1-2H3,(H,39,45)/t8-,10+,11+,12+,13+,14+,15+,16+,17-,18-,19-,20+,21-,22+,23+,24+,25-,26+,27+,28+,29-,30+,31-,32+,33+,34-,35-,36-,37+,38-/m0/s1
• InChIKey: GTDZRNBIPFWIHJ-RWQPMTCSSA-N

Globo H: Core Characterization and In-Class Context

Globo H (CAS 207384-96-1, Fucα2Galβ3GalNAcβ3Galα4Galβ4Glc-Cer) is a hexasaccharide glycosphingolipid of the globo-series classified as a tumor-associated carbohydrate antigen (TACA). It is biosynthetically derived from the stage-specific embryonic antigen-3 (SSEA-3, Gb5) pentasaccharide scaffold via α1,2-fucosylation catalyzed by fucosyltransferases FUT1 or FUT2 . The resulting molecule differs from its closest structural analog, SSEA-4, solely at the terminal saccharide residue—L-fucose in Globo H versus N-acetylneuraminic acid (sialic acid) in SSEA-4—yet this single substitution confers mechanistically distinct binding specificity, tissue distribution, and immunological relevance . Owing to its overexpression on at least 16 epithelial cancer types and predominantly luminally-restricted expression in normal tissues, Globo H has been extensively investigated as both a diagnostic biomarker and a therapeutic target in vaccine- and antibody-based immunotherapies .

1

Fucose-terminated globo-series TACA

2

Terminal sugar distinguishes from SSEA-4

3

Restricted normal tissue expression profile

• [1] KEGG PATHWAY: Glycosphingolipid biosynthesis – globo series. hsa00603. Kyoto Encyclopedia of Genes and Genomes. View Source
• [2] Eller, S. et al. (2014). Affinity of monoclonal antibodies for Globo-series glycans. Carbohydrate Research, 397: 1–6. DOI: 10.1016/j.carres.2014.07.003. View Source
• [3] Sigal, D.S. et al. (2022). Abstract P5-08-01: Globo H: A Globo series glycosphingolipid target antigen in breast cancer. Cancer Research, 82(4_Supplement): P5-08-01. DOI: 10.1158/1538-7445.SABCS21-P5-08-01. View Source
• [4] Pochechueva, T. et al. (2017). Naturally occurring anti-glycan antibodies binding to Globo H-expressing cells identify ovarian cancer patients. Journal of Ovarian Research, 10: 8. DOI: 10.1186/s13048-017-0305-8. View Source

Globo H: Risks of Analog Substitution

Substituting Globo H with its nearest structural relatives—SSEA-3, SSEA-4, or fucosyl-GM1—introduces verifiable errors in binding studies, immunological assays, and in vivo targeting experiments. SSEA-3 (Gb5) lacks the terminal fucose that defines the Globo H epitope; consequently, anti-Globo H monoclonal antibody VK-9 shows negligible cross-reactivity with SSEA-3 as demonstrated by glycan microarray dissociation constant profiling . SSEA-4 replaces the α2-linked fucose of Globo H with a β3-linked sialic acid; monoclonal antibodies against these two hexasaccharides exhibit specificity ratios where cognate binding is >100-fold tighter than non-cognate binding in fluorescence polarization assays . In hepatocellular carcinoma specimens, Globo H expression demonstrates differential prognostic significance from SSEA-4 in multivariate analysis, with high Globo H independently predicting shorter relapse-free survival (p < 0.001) while SSEA-4 showed weaker association (p = 0.005), indicating that the two antigens are not biologically interchangeable even within the same tumor type . For procurement decisions, generic glycan standards lacking the terminal fucose modification will produce false-negative results or erroneous quantification in any Globo H-targeted analytical workflow.

SSEA-3 (Gb5)

Lacks the terminal fucose defining the Globo H epitope; anti-Globo H antibodies show negligible cross-reactivity, which may lead to false-negative detection in binding studies.

SSEA-4

Replaces fucose with sialic acid, resulting in >100-fold weaker antibody binding; substitution may severely compromise assay signal and specificity in immunological workflows.

Fucosyl-GM1

Displays a narrower cancer type expression profile and distinct antibody recognition; analytical performance and prognostic interpretation may not transfer directly.

• [1] Wang, C.-C. et al. (2008). Glycan microarray of Globo H and related structures for quantitative analysis of breast cancer. Proceedings of the National Academy of Sciences, 105(33): 11661–11666. DOI: 10.1073/pnas.0804923105. View Source
• [2] Eller, S. et al. (2014). Affinity of monoclonal antibodies for Globo-series glycans. Carbohydrate Research, 397: 1–6. DOI: 10.1016/j.carres.2014.07.003. View Source
• [3] Ho, M.-Y. et al. (2017). High expression of embryonic stem cell marker SSEA3 confers poor prognosis and promotes epithelial mesenchymal transition in hepatocellular carcinoma. DOAJ: 5377dd55055e4bf2991993ed61ddafe4. View Source

Globo H: Quantitative Differentiation Evidence

Antibody Specificity: Globo H vs. SSEA-4

Anti-Globo H monoclonal antibody (α-GH) binds its cognate Globo H glycan with low-nanomolar affinity (Kd = ~2-10 nM) while showing negligible binding to the structurally analogous SSEA-4 glycan. In both fluorescence polarization (solution-phase) and surface plasmon resonance (surface-immobilized) assays, non-cognate binding was at least 100-fold lower in affinity, with the SPR format revealing an apparent ~20-fold enhancement in apparent affinity for the cognate glycan when arrayed on a surface compared to free solution . The glycan microarray study further resolved this: monoclonal antibody VK-9 showed a KD,surf of approximately 1.2 nM for immobilized Globo H, while binding to SSEA-3 (the truncated pentasaccharide precursor) was not detectable at the highest concentration tested, and binding to SSEA-4 was absent .

Antibody specificity

Head-to-head

Globo H Kd ~1.2–10 nM vs. SSEA-4 >500 nM (>100-fold margin)

Supports Globo H-selective detection without cross-reactive false positives

SPR and glycan microarray data; VK-9 no binding to SSEA-3

Evidence Dimension	Binding affinity and specificity of monoclonal antibodies
Target Compound Data	α-GH monoclonal antibody Kd ~1.2–10 nM for Globo H; VK-9 KD,surf ≈ 1.2 nM (glycan microarray)
Comparator Or Baseline	SSEA-4 glycan: Kd >500 nM (>100-fold weaker); SSEA-3 (Gb5): no detectable binding for VK-9
Quantified Difference	>100-fold specificity margin; >20-fold surface vs. solution affinity enhancement
Conditions	Fluorescence polarization (solution-phase), surface plasmon resonance (SPR), and glycan microarray with immobilized glycans

Why This Matters

α-GH antibody selectivity exceeding 100-fold against the closest structural analog (differing only in terminal saccharide) ensures that Globo H-targeted detection reagents do not produce cross-reactive false positives, a critical criterion for quality-controlled assay procurement.

• [1] Eller, S. et al. (2014). Affinity of monoclonal antibodies for Globo-series glycans. Carbohydrate Research, 397: 1–6. DOI: 10.1016/j.carres.2014.07.003. View Source
• [2] Wang, C.-C. et al. (2008). Glycan microarray of Globo H and related structures for quantitative analysis of breast cancer. Proceedings of the National Academy of Sciences, 105(33): 11661–11666. DOI: 10.1073/pnas.0804923105. View Source

Tumor-Specific Expression Profiling

Immunohistochemistry screening across multiple clinical trials has established that Globo H is detectable (H-score ≥ 15) in approximately 41% (173/420) of triple-negative breast cancer patients screened, with a median H-score of 70 among positive cases and 40% (69/173) of those positive cases scoring H ≥ 100 . In contrast to many glycan antigens that show broad normal tissue distribution, Globo H expression in normal tissues is predominantly restricted to the luminal borders of secretory epithelia—an anatomical site generally sequestered from systemic immune surveillance—resulting in a tumor-to-normal selectivity window that is quantifiable by histochemical scoring . Tissue microarray analysis comparing Globo H with SSEA-3 across multiple organ sites confirmed that normal brain, for instance, showed 0/5 positive for both antigens, while breast cancer tissue showed positive staining predominantly in epithelial compartments .

Expression profiling

Cross-study comparable

~41% TNBC H≥15; median H-score 70; normal expression limited to luminal epithelia

Supports tumor-to-normal selectivity assessment for IHC assay design

IHC on FFPE tissue cohorts; broader normal expression reported for SSEA-3/4

Evidence Dimension	Tumor expression prevalence and normal tissue distribution pattern
Target Compound Data	Globo H: 60–65% breast cancer positivity; 41% TNBC with H ≥ 15; normal expression restricted to apical/luminal epithelia
Comparator Or Baseline	SSEA-3 and SSEA-4: also expressed in some cancers but with broader normal tissue distribution; fucosyl-GM1: minimal normal tissue expression but narrower cancer type coverage
Quantified Difference	Globo H covers 60–65% of breast cancers with restricted normal expression vs. broader normal expression for SSEA-3/SSEA-4; fucosyl-GM1 absent from most normal tissues but not expressed on multiple Globo H-positive cancer types
Conditions	Immunohistochemistry (IHC) on formalin-fixed paraffin-embedded tissue; H-score quantification (intensity × % cells) across multiple clinical trial cohorts

Why This Matters

For industrial and clinical assay development, the high tumor prevalence combined with restricted normal distribution provides a favorable signal-to-background ratio, directly reducing false-positive rates in diagnostic IHC assay procurement decisions.

• [1] Sigal, D.S. et al. (2022). Abstract P5-08-01: Globo H: A Globo series glycosphingolipid target antigen in breast cancer. Cancer Research, 82(4_Supplement): P5-08-01. DOI: 10.1158/1538-7445.SABCS21-P5-08-01. View Source
• [2] Pochechueva, T. et al. (2017). Naturally occurring anti-glycan antibodies binding to Globo H-expressing cells identify ovarian cancer patients. Journal of Ovarian Research, 10: 8. DOI: 10.1186/s13048-017-0305-8. View Source
• [3] PMC Table 3. Globo H and SSEA3 expression by immunohistochemical staining of tissue array (2008). Proceedings of the National Academy of Sciences. View Source

Anti-Globo H IgG + CA125 Ovarian Cancer Detection

Naturally occurring anti-Globo H IgG antibodies in plasma significantly discriminated high-grade serous ovarian cancer patients from healthy controls (p < 0.05). When combined with the standard clinical biomarker CA125 in a logistic regression model, the diagnostic performance improved substantially compared to CA125 alone, yielding an area under the receiver operating characteristic curve (AUC) of 0.871 . The suspension glycan array technology used showed an extremely high correlation (r = 0.992) with standard flow cytometry for anti-Globo H antibody quantification, validating the analytical robustness of the platform . The glycan microarray-based detection method further demonstrated 5 orders of magnitude greater sensitivity than traditional ELISA for anti-Globo H detection in breast cancer patient sera, with statistically significant discrimination between breast cancer patients and normal blood donors (p < 0.05) .

Ovarian cancer detection

Reported

AUC 0.87 (anti-Globo H IgG + CA125)

Supports ovarian cancer biomarker panel context

Suspension glycan array, r = 0.992 vs. flow cytometry

Evidence Dimension	Diagnostic accuracy for cancer detection
Target Compound Data	Anti-Globo H IgG + CA125: AUC = 0.87; glycan array sensitivity = 5 orders of magnitude better than ELISA
Comparator Or Baseline	CA125 alone (standard-of-care ovarian cancer biomarker); traditional ELISA detection method
Quantified Difference	AUC improvement to 0.87 from CA125-alone baseline; 10^5-fold sensitivity enhancement over ELISA
Conditions	Suspension glycan array with chemically synthesized Globo H glycans; plasma from high-grade serous ovarian cancer patients vs. controls; glycan microarray for breast cancer sera

Why This Matters

The quantified improvement in diagnostic AUC when adding anti-Globo H to existing biomarker panels justifies incorporation of Globo H reagents into multiplex diagnostic assay procurement specifications.

• [1] Pochechueva, T. et al. (2017). Naturally occurring anti-glycan antibodies binding to Globo H-expressing cells identify ovarian cancer patients. Journal of Ovarian Research, 10: 8. DOI: 10.1186/s13048-017-0305-8. View Source
• [2] Wang, C.-C. et al. (2008). Glycan microarray of Globo H and related structures for quantitative analysis of breast cancer. Proceedings of the National Academy of Sciences, 105(33): 11661–11666. DOI: 10.1073/pnas.0804923105. View Source

Early Anti-Globo H IgM Predicts IgG and PFS

In the international randomized Phase II trial of adagloxad simolenin (OBI-822)/OBI-821 in 349 metastatic breast cancer patients, early anti-Globo H IgM response ≥1:320 within the first 4 weeks of vaccination was associated with significantly higher maximum anti-Globo H IgG titers (p < 0.0001) compared to patients with lower early IgM . The mean anti-Globo H IgG titer reached its highest level at week 41 post-vaccination. Each log increase in peak anti-Globo H IgM was independently associated with a 10.6% reduction in the hazard ratio for disease progression (HR: 0.894, 95% CI: 0.833–0.960, p = 0.0019) . Notably, the polyclonal antibody response elicited by the Globo H-KLH vaccine was quantified to produce anti-Globo H IgG levels often exceeding those induced by licensed bacterial polysaccharide vaccines, with complete class-switch from IgM to IgG confirmed .

Vaccine IgM & PFS

Reported

HR 0.894 per log IgM increase (p=0.0019)

Supports immune-monitoring endpoint context

Phase II trial; IgM ≥1:320 linked to later IgG peak

Evidence Dimension	Vaccine-elicited antibody response kinetics and clinical outcome correlation
Target Compound Data	Anti-Globo H IgM ≥1:320 at week 4 → IgG peak at week 41; HR reduction = 10.6% per log increase in IgM; PFS: 11.1 vs. 7.3 months (IgM ≥1:320 vs. <1:320)
Comparator Or Baseline	Placebo group and vaccinated patients with anti-Globo H IgM <1:320; bacterial polysaccharide vaccine benchmarks
Quantified Difference	p < 0.0001 for IgG enhancement; p = 0.0019 for HR reduction per log IgM; 3.8-month PFS difference (not statistically significant in this subgroup analysis); antibody levels exceeding bacterial polysaccharide vaccines
Conditions	Phase II randomized controlled trial (N = 349) in metastatic breast cancer; ELISA-based anti-Globo H antibody quantification

Why This Matters

Quantified early IgM titer as a predictive biomarker for subsequent IgG magnitude and PFS provides a procurement-relevant rationale for selecting Globo H over non-fucosylated analogs that would fail to elicit or measure this titer-based surrogate endpoint.

• [1] Hung, J.-T. et al. (2022). The clinical relevance of humoral immune responses to Globo H-KLH vaccine adagloxad simolenin (OBI-822)/OBI-821 and expression of Globo H in metastatic breast cancer. Journal for ImmunoTherapy of Cancer, 10(6): e004358. DOI: 10.1136/jitc-2021-004358. View Source
• [2] Danishefsky, S.J. et al. (2000). Polyclonal antibodies from patients immunized with a globo H-keyhole limpet hemocyanin vaccine: Isolation, quantification, and characterization. Proceedings of the National Academy of Sciences, 97(7): 3098–3103. View Source

Globo H vs. SSEA-4 Prognostic Significance in HCC

In a cohort of 328 resectable hepatocellular carcinoma patients, immunohistochemical analysis of Globo H, SSEA-3, and SSEA-4 expression revealed that all three antigens were associated with shorter relapse-free survival (RFS) in Kaplan-Meier analysis: Globo H (p < 0.001), SSEA-3 (p < 0.001), and SSEA-4 (p = 0.005) . However, multivariate Cox regression identified SSEA-3, but not Globo H or SSEA-4, as an independent predictor for RFS (HR: 2.68, 95% CI: 1.93–3.72, p < 0.001) and overall survival (HR: 2.99, 95% CI: 1.81–4.96, p < 0.001), indicating that while all three globo-series antigens carry prognostic information, their biological contributions in HCC are mechanistically distinct . Globo H thus provides additive risk stratification value not redundant with SSEA-3 or SSEA-4 expression data.

HCC prognosis

Head-to-head

Globo H high vs. SSEA-4: stronger univariate RFS association (p

Supports non-redundant prognostic endpoint context in HCC

Multivariate analysis; SSEA-3 independent predictor, Globo H additive

Evidence Dimension	Prognostic significance in hepatocellular carcinoma
Target Compound Data	Globo H high expression: significant association with worse RFS (p < 0.001, univariate); not an independent predictor in multivariate model (HR not significant after SSEA-3 adjustment)
Comparator Or Baseline	SSEA-3: p < 0.001 (univariate RFS); independent predictor HR = 2.68 (multivariate). SSEA-4: p = 0.005 (univariate RFS); weaker OS association (p = 0.01) than SSEA-3
Quantified Difference	Globo H univariate p < 0.001 vs. SSEA-4 p = 0.005; distinct multivariate prognostic profiles confirming non-redundancy
Conditions	Immunohistochemistry on tumor tissue from 328 resectable HCC patients; Kaplan-Meier and multivariate Cox regression

Why This Matters

Globo H measurement provides prognostic information in HCC that is partially independent of SSEA-3 and SSEA-4, meaning that substituting these analogs in a prognostic IHC panel would yield different risk stratification results—directly impacting clinical trial stratification assay procurement.

• [1] Ho, M.-Y. et al. (2017). High expression of embryonic stem cell marker SSEA3 confers poor prognosis and promotes epithelial mesenchymal transition in hepatocellular carcinoma. DOAJ: 5377dd55055e4bf2991993ed61ddafe4. View Source

Globo H: Verified Application Scenarios

IHC Assay for TNBC Patient Stratification

Procurement of Globo H antigen and validated anti-Globo H monoclonal antibodies (e.g., VK-9 or α-GH) enables development of quantitative IHC assays with H-score readout for identifying the ~41% of TNBC patients who may benefit from Globo H-targeted therapies. Clinical trials NCT03562637 (Phase III GLORIA) and NCT05442060 (Phase II EGFR-mutant NSCLC) both use centrally validated Globo H IHC assays with an H-score ≥ 100 for eligibility, establishing a procurement benchmark directly linked to active interventional studies .

Antibody Response Monitoring in Cancer Vaccine Trials

Quantification of vaccine-elicited anti-Globo H IgG and IgM by ELISA or glycan microarray is essential for immune monitoring in Phase II/III trials of adagloxad simolenin (OBI-822)/OBI-821 and OBI-833/OBI-821. The demonstrated ability to resolve IgM titers ≥1:320 as an early predictive biomarker directly depends on using Globo H (not SSEA-3 or SSEA-4 analogs) as the coating antigen, since cross-reactivity is negligible . The glycan microarray format offers 10^5-fold sensitivity advantage over traditional ELISA for this application .

CAR T Cell Potency and Target Validation Assays

Preclinical development of Globo H-directed CAR T cells (e.g., obi-R007) requires Globo H-positive target cell lines for in vitro cytotoxicity and activation assays. Specific lysis of Globo H-positive tumor cells at effector-to-target ratios of 0.5:1 to 2:1, coupled with CD69/CD25 upregulation and IFN-γ release, provides a quantitative potency readout that is completely dependent on Globo H expression. Substitution with SSEA-3 or SSEA-4-expressing lines would yield false-negative results due to the absence of the fucose epitope recognized by the CAR construct .

Multiplex Ovarian Cancer Autoantibody Detection Panel

The validated combination of anti-Globo H IgG detection with CA125 measurement (AUC 0.87) supports development of multiplex immunoassay panels for ovarian cancer screening. The suspension glycan array format demonstrated near-perfect correlation (r = 0.992) with flow cytometry-based detection, enabling scalable clinical laboratory implementation. Procurement of chemically synthesized, quality-controlled Globo H glycans ensures reproducibility across diagnostic lots .

Application

Selection Property

Validation Focus

TNBC research cohort enrichment

Globo H-specific IHC antibody binding

H-score ≥ 100 assay standardization

Cancer vaccine immune monitoring

Fucose-specific glycan coating antigen

IgM/IgG titer quantification by ELISA

CAR T target engagement assays

Globo H-positive target cell lines

Cytotoxicity and activation endpoint quantification

Ovarian cancer biomarker research panel

Suspension glycan array with Globo H

Multiplex panel performance review

• [1] ANZCTR Registration: The GLORIA Study – Phase 3, Randomized, Open-Label Study of Adagloxad Simolenin (OBI-822)/OBI-821. NCT03562637. View Source
• [2] Synapse Pharma Intelligence: Globo H – Drugs, Indications, Patents. Updated 2025-05-08. View Source
• [3] Lai, J.-S. et al. (2023). Abstract 4084: Globo H-targeted CAR T cell cancer immunotherapy. Cancer Research, 83(7_Supplement): 4084. DOI: 10.1158/1538-7445.AM2023-4084. View Source
• [4] Pochechueva, T. et al. (2017). Naturally occurring anti-glycan antibodies binding to Globo H-expressing cells identify ovarian cancer patients. Journal of Ovarian Research, 10: 8. DOI: 10.1186/s13048-017-0305-8. View Source