The Target Atlas for Antibody-Drug Conjugates across Solid Cancers

doi:10.21203/rs.3.rs-2884661/v1

Download PDF

Article

The Target Atlas for Antibody-Drug Conjugates across Solid Cancers

https://doi.org/10.21203/rs.3.rs-2884661/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Antibody-Drug Conjugates (ADCs) is a rapidly developing type of oncology therapeutic, spanning the targeted therapy for hematologic malignancies and solid cancers. A major requirement in ADC research is the identification of ideal surface antigens that can distinguish target cells from most mammalian cell types. Herein, we develop an algorithm and comply a comprehensive cell membrane protein annotation dataset integrated from the large transcriptome, proteome, and genome data from 19 types of solid cancer and normal tissues, to discover potentially therapeutic surface antigens for ADC targeting. The resulting target landscape includes 165 target-indication combinations and 75 cell surface protein candidates, 35 of which with features suitable for ADC targeting are never reported in ADC research and development. In addition, we identify a total of 159 ADCs from 760 clinical trials, 72 ADCs among them targeting 36 unique antigens are currently under interventional evaluation for various types of solid cancers. We analyze their normal tissue expression using the comprehensive annotation dataset and reveal a broad range of profiles for the current ADC targets. In addition, we emphasize that the biological effects of target antigens could improve their clinical actionability and put forward to comprehensively assess the drugability of target antigens from multiple aspects. This is the first attempt at pan-cancer ADC target exploration over the past two decades, and our findings indicate that the target atlas across solid cancers can provide great opportunities to expand the broader prospects of ADC therapies.

Health sciences/Biomarkers

Biological sciences/Cancer/Tumour biomarkers

ADC target discovery

transcriptomics

proteomics

genomics

TCGA

GTEx

Cancer is the first or second leading cause of premature death (i.e., at ages 30–69 years) in 134 of 183 countries; in another 45 countries, it ranks third or fourth [1]. The section of Cancer Surveillance (CSU) of International Agency for Research on Cancer (IARC) estimates that there will be 19.3 million new cases and 10.0 million cancer deaths worldwide in 2020 in both sexes and all ages combined (https://gco.iarc.fr/today/home). Of the 19.3 million new cases and 10.0 million premature deaths, 18.0 million (93.37%) and 9.3 million (92.85%) will come from solid tumours, respectively. A systematic evaluation of oncology shows that most cancer drugs authorised by European Medicines Agency in 2009-13 came onto the market without achieving significant improvement on overall quality or quantity of patients’ life [2]. Thus, there is an urgent need to increase the therapeutic efficacy of drugs for solid tumours.

ADCs are immunoconjugates that consist of recombinant monoclonal antibodies (mAbs) covalently bound to cytotoxic chemicals (known as the payloads or warheads). ADCs are designed to leverage the exquisite specificity of mAb to deliver and release highly cytotoxic chemotherapeutic agents to tumour cells and their surrounding stromal cells, and cancer stem cells [3, 4]. To date, eleven ADCs have received marketing authorization and 760 clinical studies have been registered in ClinicalTrials.gov (Supplementary Table 1–2). Over 400 of the 760 clinical trials are being actively conducted in various stages including 227 trials targeting different types of solid tumours (Supplementary Table 2). In 2000, the first ADC, gemtuzumab ozogamicin, gained regulatory approval from the US Food and Drug Administration (FDA), while in the next ten years, no ADC had been approved worldwide. In the past two years from 2019 to 2022, however, a total of ten ADCs were approved, seven of which were authorized to the treatment of solid cancers (Supplementary Table 1). Trastuzumab deruxtecan (DS-8201) monotherapy targeting HER2 has been particularly effective in a pretreated patient population with HER2-positive metastatic breast cancer, resulting in the confirmed overall response rate and disease-control rate as high as 60.9% and 97.3%, respectively [5]. In short, ADC has become a popular ‘training ground’ for drug innovation.

An ideal ADC should be a combination of the following three parts [6]: 1) an mAb that is highly specific for homogeneous or heterogeneous antigen overexpressed in tumours; 2) a linker that is stable in the blood circulation yet can easily cleavage at target sites; 3) a warhead highly sensitive for certain indications. The antibody, linker and payload can be optimized through various strategies. For example, site-specific conjugation technology can be used to improve the safety and efficacy of ADCs with or without antibody engineering [7, 8, 9, 10, 11]; off-target payload exposure can be mitigated by promoting the stability of conjugation and linker [12, 13, 14]; the efficiency of ADC uptake and processing can be improved by the employment of bispecific antibodies or biparatopic antibodies [15, 16]; explore payloads with novel anti-tumor mechanisms to deal with the drug resistance challenge brought by the warhead part [3, 17]. However, the intrinsic characteristics of ADC target proteins other than the three are basically constant, and there is no way to optimize their gene copy number variation, endocytosis rate and trafficking route, absolute protein levels, and differential expression patterns, etc. in addition, considering that most antigens are tumour-associated rather than tumour-specific, the on-target off-tumour toxicity thus becomes inevitable. For example, BR96-doxorubicin treatment led to hemorrhagic gastritis in patients due to the unrecognized level of Lewis(y) antigen on gastric mucosal cells [18]. Treatment with CD44v6-directed Bivatuzumab mertansine and BAY794620 targeting CA9 antigen led to fatal epidermal necrolysis and gastrointestinal toxicity, respectively, due to the expression of targets in corresponding tissues [17, 19]. Therefore, selecting a promising ADC target must meet numerous criteria to enable efficacious tumour eradication while preventing intolerable damage to normal tissues.

Importantly, an ideal ADC target antigen should be expressed at absolutely high density in target cells in a large percentage of patients regardless of pathologic stage, and should show restricted expression in normal tissues, especially the vital tissues and hematopoietic stem cells (HSCs) and multipotential progenitors (MPPs). It also should be noted that high levels of a target antigen on various types of blood cells could impact drug exposure and thereby limit the aggregation of ADCs to the target site. After fulfilling the above criteria and considerations, each candidate should have a substantial ratio of differential expression in tumour relative to its adjacent normal tissue as well as other normal tissues. In addition to the protein expression profiles, the biological functions of target antigen in tumourigenesis are proven to be additional factors that can greatly influence the potency of ADCs [20]. It is therefore imperative to carefully evaluate the druggability of candidates based on comprehensive antigen annotation sources, as well as analytical tools specifically designed to identify potential ADC target antigens.

The goal of this study is to define an ADC target discovery strategy by integrating transcriptome, proteome, and genome datasets from solid tumour cell and normal cell populations, and compiling a comprehensive dataset of surface antigen annotations to identify potential therapeutic targets. Meanwhile, we also provide a survey of the ADC interventional studies annotated from ClinicalTrials.gov, so as to provide the academia and the pharmaceutical industry with the most comprehensive ADC target atlas across solid cancers.

2.1 Assembling a Comprehensive Dataset and Designing an Algorithm to Identify ADC Targets

We assembled a comprehensive annotation dataset and designed an algorithm to discover surface molecules differentially expressed between tumour tissues and normal tissues (Fig. 1a-b). We used RNA-sequencing data to discover differentially expressed therapeutic targets, while the best-in-class RNA-seq data processing pipelines were shown to produce consistent expression estimates for just 88% of protein-coding gene [21]. A pipeline was successfully developed to unify RNA-sequencing data of TCGA and GTEx by minimizing differences between matching tissues from TCGA and normal tissues from GTEx, and by processing data sets uniformly without the inter-project normalization step [22]. We performed differential analysis for 20,242 HUGO genes across 19 solid cancer types based on the sequencing data processed from this pipeline. By setting a threshold of Benjamini-Hochberg adjusted p-values of 0.01 and log₂FoldChange of 1.0, we kept those HUGO genes significantly overexpressed in tumour cell populations. To search for potential extracellular membrane proteins from the overexpression lists, we compiled a membrane protein annotation dataset (6176 entries) from three databases (i.e., Uniprots, HPA, and Compartments) that contain protein subcellular location information, and then merged it with the above overexpression lists. To exclude molecules highly expressed in normal tissues, we used the mRNA expression dataset integrated from the GTEx, HPA, and FANTOM5, and supplemented the dataset with proteome data from the HPA and HPM. We excluded molecules exhibiting high expression in the vital tissues (i.e., heart, liver, spleen, lung, kidney), as well as molecules exhibiting high expression in more than three normal tissues except for the blood, bone marrow, brain, and placenta. Thereafter, we used the publicly available resource BloodSpot [23] and found those surface molecules that are restrictedly expressed in the bone marrow Lin⁻ CD34⁺ CD38⁻ CD90⁺ CD45RA⁻ HSCs and Lin⁻ CD34⁺ CD38⁻ CD90⁻ 45RA⁻ MPPs. The next quality control step further leaved us with 90 potentially therapeutic surface membrane proteins and their 243 target-indication combinations (Supplementary Table 3).

Clinical responses are more commonly observed in patients whose tumours exhibit high target antigen levels. To remove surface molecules whose absolute expression levels in tumour tissues are only modest, we utilized publicly available gene expression dataset and analyzed the differential expression pattern of target antigen expression in the paired indication and normal tissues (not limited to the counterpart). This step further leaved us with 87 candidates and their 225 target-indication combinations. To eliminate any molecule highly heterogeneously expressed in the patient population, we utilized large oncology repositories that have paired genomic and transcriptomic data. This step identified several target antigens whose gene expression levels significantly decrease in a large proportion of patients bear a gene mutation, thus leaving us with 84 candidates and their 213 target-indication combinations. To further annotate the tumour heterogeneity of the target antigens, we then predicted their protein overexpression rates in each of the 22 different tumour (sub)types and prioritized which target antigens could be considered for subsequent immunohistochemistry validation. With this algorithm, we finally identified 75 molecules with features suitable for ADC targeting and their 165 target-indication combinations (Supplementary Table 4). In the last stage, we comprehensively scored them from twelve aspects of four dimensions, including expression profiling (differential expression, absolute levels in tumour, expression homogeneity, normal tissue expression, expression in HSC and MPP), internalization (endocytosis and trafficking route), difficulty of target antibody discovery (single/multiple transmembrane, extracellular domain size, extracellular domain homology), and biological functions (genetic basis and cancer stem cell features).

2.2 Identified ADCs and Their Target Antigens in Clinical Trials

A total of 159 ADCs were identified from 760 clinical trials in ClinicalTrials.gov, 406 of which are being actively evaluated in various stages across hematologic malignancies and solid cancers (Supplementary Table 2). For the various solid cancer types, 72 ADCs targeting 36 antigens identified from 227 interventional studies are currently under evaluation (Fig. 2a). Meanwhile, 118 trials for 51 ADCs are completed, 46 trials regarding 31 ADCs are terminated, and statuses of 6 trials on 5 ADCs are unknown (Supplementary Table 2). The overwhelming majority of ADC trials for solid cancers was initiated in 2008 or later, and BRCA is the only indication that has had new clinical trials initiated every year since 2006 (Fig. 2b). The largest numbers of ADCs and trials on solid cancers (n_ADC = 28, n_study = 139) were evaluated for BRCA, followed by pan-cancer (n_ADC = 73, n_study = 113), NSCLC and SCLC (n_ADC = 21, n_study = 43), BLCA (n_ADC = 7, n_study = 18), OV (n_ADC = 12, n_study = 18), STAD (n_ADC = 11, n_study = 18), and PRAD (n_ADC = 9, n_study = 18) (Fig. 2b and Supplementary Table 2). However, in the case of LIHC, the only trial to evaluate the safety and tolerability of GPC3-directed ADC in patients was initiated in 2016 and then terminated without publicly disclosed results. Similarly, the investigation of ADC for CESC only involves TF target (also known as F3). In addition, there is no clinical trial regarding ESCA and THCA, which are commonly diagnosed cancers that account for 3.1% and 3.0% of the total cancer cases, respectively (https://gco.iarc.fr/today/home).

In the past two decades, there are a total of 397 clinical trials for solid cancers among various stages, including 163 in Phase I, followed by 127 in Phase II, 65 in Phase I-II, 37 in Phase III, 4 in Phase II-III, and 1 in Phase IV. Publicly disclosed targets for the 397 clinical trials are 67 unique antigens, 32 of which involve two or more studies (Fig. 2c). Twenty-one ADCs are directed against HER2, followed by TROP2 (n = 5), MSLN (n = 4), and FOLR1 (n = 3), each attracts at least three different ADCs (Fig. 2c). HER2 is the only target that has at least 20 clinical studies in each Phase I, II, and III. HER2, TROP2, MSLN, FOLR1, and DLL3 are the five antigens targeted by at least one ADC under more than ten trials evaluation (Fig. 2c). To date, a total of ten ADCs are currently under Phase III trials, including the registered trastuzumab emtansine (T-DM1, HER2-directed), enfortumab vedotin-ejfv (ASG-22CE, NECTIN4-directed), trastuzumab deruxtecan (DS-8201a, HER2-directed), sacituzumab govitecan (IMMU-132, TROP2-directed), and six other ADCs, two of which target CEACAM5 and FOLR1 for NSCLC and OV, respectively. In addition, twenty ADCs are evaluated in Phase II trials and two ADCs are in Phase II/III trials. Nine ADCs that have been evaluated in Phase II did not proceed to Phase III and 27 ADCs that have been assessed in Phase I did not proceed to Phase II. For these ADCs currently in Phase II or Phase II/III trials, their target antigens -F3 (TF), ROR1, FOLR1, CD56 (NCAM1), ENPP3, MET, CD166 (ALCAM), HER3 (ERBB3), MSLN, and SLC39A6 (LIV1)- have not yet been in the ADC market (Fig. 2a). According to the expression dataset, the antigens targeted by drugs in the market (i.e., HER2, EGFR, NECTIN4, TROP2) to treat solid cancers are restrictedly expressed in normal tissues. Nevertheless, target antigens currently active in clinical trials show relatively high expression in normal tissues. CD166 (ALCAM) and CD56 (NCAM1), for examples, are highly expressed in the kidney, liver, lung, and heart (Fig. 2a). Overall, target antigens in drugs approved or active trials are highly expressed in multiple solid cancers, especially LUAD, but significantly decreased their expression in LIHC (Fig. 2d). This result indicates that most ADCs are developed to target LUAD, while LIHC lacks suitable target antigens. This may be due to the fact that LIHC shows relatively unique genome-wide global expression profiles with almost no overlap with the other cancer types [24].

2.3 Differential Gene Expression Patterns Between Tumour and Normal Tissues

Of the 90 target antigen candidates discovered via our algorithm, it was found that CELSR1, GPR87, OR51E1, SLC2A12, VANGL1 have no medium-high expression in the 32 rearranged tissues of interest, and there are 26, 18, and 14 targets showing medium-high expression in one, two, and three tissues, respectively (Fig. 3). CD276, ERBB2, ERBB3, CMTM4, and GPNMB exhibit medium or medium-high expression in more than nine normal tissues, but CD276, ERBB2, and ERBB3 are not highly expressed in any tissues except for the brain. Among the 243 target-indication combinations, MUC17-STAD is the combination that has the highest differential ratio (log₂ transformed) of MUC17 expression level in gastric tumour tissue to that of normal gastric tissue, reaching 9.8, followed by VTCN1-UCEC, DSC3-LUSC, GPR87-LUSC, GJB6-LUSC, and GPC3-LIHC, with ratios that reach 9.1, 8.1, 8.1, 8.1, and 7.8, respectively. Similarly, the differential expression ratio of MUC17-ESCA is also as high as 12.3, but the absolute gene expression level of MUC17 in ESCA is too low for this pair to be a target-indication combination. It is noteworthy that although the differential ratios of ERBB2 expression level across 19 solid cancer tissues to that of adjacent normal tissues are not prominent, it is the target accumulated with the largest number of ADC, an issue we will further discuss later. TACSTD2 (TROP2) accounts for the second largest number of ADC, while the first approved indication for TROP2-directed sacituzumab govitecan is BRCA, the pair with the lowest differential expression ratio. For several other indications such as BLCA, NSCLC, PRAD, and UCEC with relatively high differential expression ratios, five Phase II trials and two Phase III trials were launched in the past three years 2018-20 (Supplementary Table 2).

In addition to the analysis of differential expression relative to adjacent tissue, we also calculated the differential expression ratios of target gene expression level in the paired indication to that of other normal tissues (Fig. 4). We found that the esophagus is most unfriendly, because in 14.5% of the target-indication combinations, the absolute gene expression levels of targets in their paired indications are lower than their expression levels in the normal esophagus. Liver, gallbladder, and pancreas are on the contrary the three most friendly tissues as in no target-indication combination is the target gene expression level lower than that in the three normal tissues. Although the differential expression ratios for ERBB2 are not impressive, its expression level in each paired indication is at least double that of each normal tissue. For TACSTD2, it is differentially expressed between numerous tumour tissues and their paired normal tissues more than five times. Yet its performance is relatively poor in the esophagus and salivary gland. CD276 has many target-indication combinations, while its expression level in each paired indication is less than double and sometimes even lower than its expression level in the adrenal gland, breast, cervix, prostate, and uterus. In addition, TNFRSF21 and TPBG are more expressed in the urinary bladder than their paired indications, and DLK1 expression level in LIHC is lower than its expression level in the normal adrenal gland, bone marrow, ovary, and testis tissues.

2.4 Heterogeneous Gene Expression Patterns in Solid Cancers

Homogeneous gene expression in clinical indications is an important characteristic for antigens to become potential therapeutic targets. We comprehensively analyzed the paired genome and transcriptome data and found that in 146 target-indication combinations, 65 mutated genes are significantly related to the gene expression level of 51 targets among 13 types of solid cancer (Fig. 5a and Supplementary Table 5). Among them, TP53 mutation has the greatest correlation with target gene expression, as in 21 combinations out of the 146, TP53 that takes the forms of coding and disruptive mutations significantly affect the expression of target genes (Fig. 5b). For example, the expression level of BMPR1B in 34.3% of BRCA samples bear mutated TP53 is 4.9 times higher than that in samples without TP53 mutation, while the expression levels of KCNE4 and SLC39A6 decrease by 76.1% and 72.5%, respectively. Similarly, the expression level of TACSTD2 in 58.8% of THCA samples bear BRAF mutation increases by 3.9 times, and FGFR3 expression level is upregulated by 3.4 times in 13.4% of BLCA patients harbouring self-gene mutation. Several other examples with a relatively large degree of correlation come from ERBB2_PTEN (57.4% of BLCA samples, FC = 3.2, direction = down), MSLN_TP53 (47.4% of LUAD samples, FC = 2.6, direction = down), MET_BRAF (58.8% of THCA samples, FC = 2.5, direction = up), and GPNMB_VHL (49.1% of KIRC samples, FC = 2.3, direction = down) (the former is the target gene, the latter is the mutated gene). In addition, the correlation between gene mutation and some target genes expression levels differs across pathological stages (Fig. 5c). For example, TP53 mutation has the greatest correlation with the downregulated BMPR1B expression in BRCA pathological stage II, and BRAF mutation is most related to the upregulation of TACSTD2 expression in the pathological stage II of THCA. It is noteworthy that IGF1R expression is upregulated in the pathological stage I of BRCA patients harbouring mutated TP53, while it is downregulated in stage II and stage III.

Regardless of the gene mutation, gene expression level also varies among different pathological stage and tumour size, and are different with the tumour invasion and metastasis to peripheral lymph node and distant metastasis (Supplementary Fig. 1–5). As the cancer progresses from pathological stage I to stage IV, the expressions of APOLD1 and ST14 are gradually down-regulated in KIRC and KIRP, respectively (Supplementary Fig. 2–3), while that of MET and TACSTD2 in THCA are first down-regulated and then up-regulated (Supplementary Fig. 4–5). The expression level of CDH3 in the ESCA pathology stage II is about twice that in other stages (Supplementary Fig. 1). As the tumour grows and spreads into nearby tissue, the expressions of APOLD1 and ST14 are gradually down-regulated in KIRC and KIRP, respectively, while that of MET and TACSTD2 in THCA are gradually up-regulated. As the tumour spreads to nearby lymph nodes and metastases distant, the expressions of APOLD1 and ST14 show a downward trend in KIRC and KIRP, respectively. Yet, the expressions of MET and TACSTD2 are significantly upregulated in THCA as the tumour spreads to nearby lymph nodes.

2.5 Predicted Protein Overexpression Rates of Target Antigens

To further annotate the tumour heterogeneity of the target antigen, we predicted protein overexpression rates for the 75 unique target antigens in each of the 22 different tumour (sub)types (Supplementary Table 6) based on their expression profiles. We observed that over 75% ADC targets in their 243 target-indication combinations show a predicted overexpression rate of > 10% of samples. A predicted protein overexpression rate > 75% of samples was observed for 24 ADC targets in 30 target-indication combinations, and > 50% for 41 in 54 (Fig. 6).

In BRCA (sub)types, a predicted overexpression rate of > 77% of samples for HER2 was observed in ER-negative/HER2-positive or ER-positive/HER2-positive breast cancer. In ER-positive/HER2-negative and triple-negative breast cancer (TNBC), the overexpression rate dropped to 23% and 5%, respectively. For BMPR1B, CELSR1, ERBB3, IGF1R, KCNE4, SLC39A6, and TPBG, their overexpression rates in ER-positive breast cancer samples are significantly higher than that in ER-negative samples,

while for CD276, LRRC15, PRLR, and VTCN1, each protein was observed to show overexpression independent of breast cancer intrinsic subtypes. In TNBC, the highest predicted overexpression rate with 52% was observed for VTCN1 (B7-H4), followed by 43% for CD276 (B7-H3). VTCN1 is a target antigen that has not been clinically explored in breast cancer. In kidney cancer (sub)types, it was observed that VCAM1 and HAVCR1 are two targets highly overexpressed in both KIRC and KIRP, while MET was observed to show overexpression independent of kidney cancer intrinsic subtypes. In KIRC, the highest predicted overexpression rate with 93% was observed for CA9, while in KIRP and KICH, 100% and 78% were observed for HAVCR1 and MET, respectively. In addition, several other unreported targets screened from our algorithm, including APOLD1, BACE2, CFTR, COL23A1, FLT1, FZD1, SLC4A1, SLC6A3, were observed to be overexpressed in > 50% kidney cancer (sub)type samples. In BLCA we observed the highest predicted overexpression rate of 67% for UPK1B, followed by 56% for TNFRSF21. In COREAD, RNF43 ranked the highest with a predicted overexpression rate of 78%, followed by NOX1 (53%). In LIHC, it was observed that, three target antigens -TM4SF4, GPC3, FGFR4- show the top three highest overexpression rates. PRAD was observed to be the cancer type with the largest number of target antigens overexpressed in more than 75% samples, TRPM8, SLC2A12, SLC30A4, and OR51E2 among them have not been reported for ADC application.

2.6 Genetic Basis of Target Antigen Candidates

We systematically analyzed the genetic basis of the 90 targets, and found that ERBB2, EGFR, MET, and FGFR3 as oncogenic driver are frequently altered in a wide type of solid cancers. The most common alterations in oncogene EGFR are mutation (5.5%), amplification (2.6%), exon 19 mutation (1.8%), exon 19 deletion (1.6%), and exon 21 mutation (1.3%). When it comes to LUAD, these proportions rise to 26.1%, 5.3%, 11.5%, 10.7%, and 8.6%, respectively [25] (Fig. 7a). The annotation about EGFR alterations curated by OncoKB shows that exon 19 deletion, L858R, and T790M, as well as amplification are all to be oncogenic and have gain-of-function effects. ERBB2 as another oncogene alters in 13.9% of BRCA patients with amplification and mutation present in 12.6% and 3.4% of all patients, respectively (Fig. 7a). FGFR3 mainly alters in BLCA patients, and mutations account for the overwhelming proportion of its alteration. S249C is the main source the mutations, followed by Y373C, G370C, and G380R (Fig. 7a). MET amplification and mutation occur in many solid tumours, but the alteration frequency in each cancer type does not exceed 5% (Fig. 7a). Copy number variations (CNVs) are usually propagated to the proteome level, while post-transcriptional mechanisms attenuate this impact. For ERBB2, the greatest agreements between its transcript and CNVs and transcript and protein, are observed in BRCA (Fig. 7b). Renal cell carcinoma, however, are with poor correlations. Similarly, EGFR shows good agreement in transcript/CNV and transcript/protein in multiple cancer types (Supplementary Fig. 6). We applied radar charts to exhibit various indicators under different measurement systems. The performance of ERBB2 on BRCA and BLCA is the most prominent among all target-indication combinations, although it is moderately expressed in many normal tissues (Fig. 7c). Compared with the ERBB2-BRCA or ERBB2-BLCA combination, performance of EGFR on LUAD falls behind in terms of the expression profiling. But its performance is superior to FGFR3-BLCA with regard to cancer stem cell features. Regardless of the two factors of genetic basis and cancer stem cell features, the performance of NECTIN4 on BLCA is perfect; so is the performance of MET in LUAD, except for its expression heterogeneity (Fig. 7c).

The selection of surface membrane protein targets is critical for ADCs to reach an acceptable therapeutic index. Target antigens should ideally be homogeneously expressed at an absolutely high level in tumours but expressed at a low level in normal tissues, or high expression limited to a small fraction of tissue types. The roles target antigens play in tumourigenesis and the extent to which they are oncogenic drivers are elucidated to explain additional clinical benefits of ADC targeting independent of cytotoxic drugs. The elucidation of the molecular basis underlying resistance to trastuzumab emtansine and brentuximab vedotin further implicates the importance of the discovery of ADC targets, since one of the mechanisms of resistance related to the naked antibody moiety is the decreased level in target antigens HER2 and CD30. In this study, we generated a discovery platform that used large datasets of transcriptome, proteome, and genome of 19 types of solid cancer and normal tissues to search for ideal ADC targets. This study constitutes the first attempt to explore pan-cancer ADC targets.

Research on ADC target antigens usually focuses on comparing expression in tumour tissues to their normal counterparts, rarely considering the systemic expression of target candidates, thus underestimating the risk of toxicity across normal tissues. To address body-wide molecule expression, we combined extensive proteomics databases to draw the human proteome map. The combination of databases based on immunohistochemical analysis and mass spectrometry increases confidence in the reliability for low-level expression. In addition, lessons from clinical experience suggest that the balance of on-target toxicity between the normal and tumour cells is critical for the success of clinical trials. We thereby systematically analysed the differential expression profiles of candidates between their paired indication and not only the normal counterparts, but also other normal tissues. Absolutely high expression of antigens in tumour tissues together with low expression in normal tissues allows the target to obtain considerable differential expression profiling, while relatively homogeneous expression among patients with a given indication allows the target to have a larger beneficiary population. We utilized large oncology repositories and removed those antigens whose gene expression levels dramatically decrease in a large proportion of patients bear a gene mutation. We further annotated the tumour heterogeneity of the target antigen based on functional genomic mRNA profiling. Target selection should also take into consideration the expression in HSCs and MPPs, the restricted expression could enable the cell populations depleted by ADCs to consistently replenish. Overall, starting from 20,242 HUGO gene symbols across 19 solid cancer types, our algorithm identified 75 candidate antigens with features potentially suitable for ADC targeting and their 165 target-indication combinations.

In the RTK/RAS/MAP-Kinase signaling pathway, several particularly interesting receptor tyrosine kinases in our selection list (Supplementary Figure 7), including ERBB2, ERBB3, EGFR, FGFR3, and MET, are recurrently altered by mutation, amplification and/or overexpression in a large proportion of patients with various types of cancer. Cancer type-specific alterations of these oncogenic genes have different levels of clinical actionability. As in the biomarker-drug association case for ERBB2 amplification in BRCA, the first-line treatment is ‘dual HER2 blockade’ with trastuzumab and pertuzumab plus a taxane. The mechanism of action of T-DM1 is relevant to not only the DM1 payload, but also the trastuzumab moiety since it retains all the mechanism of action of the naked antibody [26]. The results of DESTINY-Lung01 showed that the overall confirmed ORR of DS-8201 in the HER2 overexpression cohort was 24.5%, while in the HER2 mutation cohort, the confirmed ORR was as high as 61.9% [27]. Furthermore, cancer stem cells that are tumourigenic may be resistant to conventional cancer therapies, and efforts are now being directed towards exploring precision medicine to target these cell types. Several target antigens in our target atlas, including LGR5, EGFR, ERBB2, CSF1R, MET, IGF1R, and TPBG (5T4), are cell surface antigens with characteristics of cancer stem cells, and thus might enable the ADCs directed to these targets to suppress cancer relapse and metastasis.

The strict internalization requirement in ADC development has been challenged in the past decade [28]. Non-internalizing ADC were shown to induce a potent lineage ablation for hematologic malignancies and solid tumours [29,30,31]. Target antigen candidates that lack high degree of credibility regarding endocytosis and endocytic trafficking routes were therefore retained in our selection list.

Post-translational modifications appear to influence the expression levels of a fraction of proteome; therefore, performing differential analysis based on proteome is essential for the identification of ideal ADC targets. Currently, the Cancer Proteome Atlas (TCPA) that is based on the Reverse Phase Protein Array (RPPA) only covers 260 protein markers. The National Cancer Institute’s Clinical Proteomic Tumour Analysis Consortium (CPTAC), however, lacks proteomic data in the adjacent normal tissues. In addition, although this study covers a wide range of solid cancers, hematologic malignancies and many other rare solid tumour types are not in the scope of this study, since we only processed tissue types studied jointly by TCGA and GTEx for proper batch bias correction. Tissues that are not available or in a sufficient number of normal samples in TCGA were not analyzed (eg., SARC, OV, SKCM). Taken together, future studies should expand these analyses to larger sample sets, additional data types, such as post-translational modifications of proteins and functional proteomics.

Heterogeneous bulk tumour includes a diverse collection of tumour cells harbouring distinct subclonal signatures with various levels of resistance to anticancer strategies. It is therefore an accurate assessment toward cancer heterogeneity essential for the successful identification of ideal ADC target antigens. Multiregion sampling and sequencing together with single-cell dissociation and sequencing are emerging technologies with great potential to dissect the complex subclonal architecture of cancers.

This present study using a high-throughput annotation database based on transcriptomics, proteomics, and genomics constitutes the first attempt to explore pan-cancer ADC targets. This comprehensive ADC target landscape is meant to provide a valuable resource for cancer researchers, clinical oncologists, and for a broad scientific and pharmaceutical community interested in the fast-growing ADC oncology therapeutic for solid cancers.

Clinical trials search strategy

To identify the most comprehensive target atlas from clinical trials, ClinicalTrials.gov and PubMed were searched on 31 December 2022 with the following search terms ‘antibody-drug conjugate’, ‘cancer’, ‘tumour’, and ‘oncology’ in various combinations. Just interventional studies were included, and the early phase I trials were grouped with phase I trials. In addition, abstracts and posters from ASCO 2022, ECCO 2022 and ESMO 2022 congresses were included for ADC searching with the terms ‘antibody-drug conjugate’ or ‘ADC’.

Data acquisition of the normal tissue transcriptome and proteome

Expression profiles for human tissue proteins based on IHC were retrieved from the Human Protein Atlas (HPA) under the entry of ‘Normal tissue data’ (https://www.proteinatlas.org/about/download, normal_tissue.tsv.zip). Proteomic sequencing data based on LC-MS/MS was retrieved from the Human Proteome Map (HPM) (http://www.humanproteomemap.org/, HPM_normal protein_level_expression_matrix_Kim_et_al_052914 - LC-MSMS.xlsx). Consensus transcript expression profiling integrated from the HPA, GTEx and FANTOM5 was retrieved from the HPA under the entry of ‘RNA consensus tissue gene data’ (https://www.proteinatlas.org/about/download, rna_tissue_consensus.tsv.zip).

Retrieval and compilation of protein subcellular location data

Subcellular localization of proteins was retrieved from the HPA under the entry of ‘Subcellular location data’ (https://www.proteinatlas.org/about/download, subcellular_location.tsv.zip), and the knowledge channel of COMPARTMENTS (https://compartments.jensenlab.org/Downloads, human_compartment_knowledge_full.tsv). The membrane protein annotation dataset (includes 6176 entries) was compiled by extracting all cell surface membrane protein information first through the R language (version 4.0.3), and then merging the above information.

Rearrangement and mapping of the human tissues

Since the human tissue nomenclature differs among source repositories, each data set was mapped to a set of consensus tissue labels. In cases when mapping multiple tissues in one repository to a single tissue label in another source, the maximum expression value was selected. For example, the caudate, cerebellum, cerebral cortex, choroid plexus, dorsal raphe, hippocampus, hypothalamus, pituitary gland, and substantia nigra were collapsed into a single tissue category, “brain”. The cervix, uterine, endometrium, ovary, fallopian tube, vagina, epididymis, seminal vesicle, testis, and prostate were collapsed into internal genitalia. In addition, the adult adrenal, adult colon, adult esophagus, adult frontal cortex, adult gallbladder, adult heart, adult kidney, adult liver, adult lung, adult ovary, adult pancreas, adult prostate, adult rectum, adult retina, adult testis, and adult urinary bladder in the HPM were mapped to the adrenal gland, colon, esophagus, brain, gallbladder, heart muscle, kidney, liver, lung, internal genitalia, pancreas, internal genitalia, rectum, eye, internal genitalia, and urinary bladder, respectively. To maintain consistency, fetal tissues were also discarded, resulting in 32 unique tissue categories.

Normal tissue expression and binning

In the interest of facilitating target screening, the expression values were classified into five categories, including ‘High’, ‘Medium’, ‘Low’, ‘Not Detected’, and ‘Not Available’. To accomplish the binning, we first perform log₁₀ conversion on HPM dataset, and then temporarily correct it for the purpose of abundance distribution estimation. In order to best fit the normal curves to the observed distributions, we applied the Broyden-Fletcher-Goldfarb-Shanno algorithm [32], and subsequently obtained the peak maximum and standard deviation measure. Expression values in the range of one standard deviation above the peak maximum were recorded as ‘Medium’, and expression values above this threshold were recorded as ‘High’. Similarly, expression values in the range of one standard deviation below the peak maximum were recorded as ‘Low’, and for those falling below one standard deviation were recorded as ‘Not Detected’. Proteins without expression values were recorded to be of ‘Not Available’ abundance. The natural format of the expression profile of human tissue proteins obtained from IHC is the five aforementioned categories, so there is no adjustment. While for the RNA consensus expression profiling integrated from the HPA, GTEx and FANTOM5, the consensus normalized expression (NX) values between 20 and 40 were recorded as ‘Medium’, and the NX values above this threshold were recorded as ‘High’. Similarly, the NX values in the range of 1-20 were recorded as ‘Low’, and for those falling below 1 were recorded as ‘Not Detected’. The NX values in other cases were recorded as ‘Not Available’.

Differential expression analysis of genes between tumour and its adjacent normal tissue

We gathered the uniformly processed TCGA and GTEx RNA-sequencing data from the RNAseqDB (https://github.com/mskcc/RNAseqDB). All together there were 9,109 high-quality samples covering 14 normal tissues and 19 types of solid cancer. The DESeq2 package [33] was used to identify HUGO genes that are differentially expressed between tumours and their adjacent normal tissues. By setting a threshold of Benjamini-Hochberg adjusted p-values of 0.01 and log₂FoldChange of 1.0, those HUGO genes that were significantly upregulated in the tumours were retained.

Differential expression analysis of genes between tumour and other normal tissues

The read count data of RNA-sequencing gathered from the RNAseqDB was first transformed to TPM (transcripts per million) format that can be directly used to compare gene expression. To transform read counts into TPM format, we need to normalize for gene length, and then normalize for gene depth, in that order. For the gene length normalizing step, we fist calculated gene length from GTF files (GDC.h38 GENCODE v22 GTF (used in RNA-Seq alignment and by HTSeq)) by counting the longest transcript of each gene (sum of exons) or the sum of all exons, then divided each count by the length of its respective gene. For the gene depth normalizing step, we performed in the order as follows: 1) sum all counts within each sample column; 2) divide each column sum by the desired depth (1,000,000) to yield scaling factors; 3) divide each sample count within a column by its respective scaling factor.

The TPM values of each gene in its paired indication and normal tissues were used as input data for differential expression analysis. The non-parametric Mann–Whitney U test analysis was applied to calculate the differential expression ratios. The differential expression patterns of target genes between their paired indication and normal tissues were visualized via the OmicCircos package [34].

Tumour tissue transcriptome, genome, and phenotype compilation

To integrate transcriptome and genome and phenotype information of the gene set of interest. We downloaded the phenotype data of TCGA patients with various solid cancers from UCSC Xena (https://xenabrowser.net/datapages/), and then extracted and organized the information about gender, neoplasm histologic grade, pathologic stage, and TNM staging. The non-silent mutations (SNP and INDEL) for each gene in individual cancer type were determined through mining the MC3 (“Multi-Center Mutation Calling in Multiple Cancers”) TCGA MAF (mutation annotation format) file. The gene-level transcription estimates (in log₂(x+1) transformed RSEM normalized count format) were transformed to TPM format that can be directly used to compare gene expression. Thereafter, we compilated a comprehensive dataset via integrating the aforementioned tumour tissue transcriptome and genome information and the organized phenotype data. The heterogeneous expression pattern analysis was performed using the ggstatsplot package in batch mode (https://github.com/IndrajeetPatil/ggstatsplot).

Annotation of functionally relevant mutation

OncoKB (https://www.oncokb.org/) contains annotation information about the impact and therapeutic significance of 5616 specific alterations in 682 cancer genes. It combines multiple resources, including FDA, NCCN (National Comprehensive Cancer Network) and other guidelines, ClinicalTrials.gov and scientific literatures. We utilized the annotation information about oncogenic and clinically actionable alterations from the OncoKB and discarded somatic mutations that were labeled as likely oncogenic or predicted oncogenic, resulting in the set of driver mutations, which were not contaminated by passenger mutations. The collected information was applied to analyze the therapeutic significance of potential target genes that altered in a large proportion of patients.

Predicting overexpression rates of target antigens

We applied the method called functional genomic mRNA profiling to predict overexpression rates of target antigens [35]. Typically, principal component analysis (PCA) was used to analyze the mRNA transcriptome and n eigenvalues and n corresponding eigenvalues (transcriptional components) were subsequently obtained. We identified these subsets of transcriptional components that describe non-genetic regulatory factors (physiological and experimental factors) and used them as covariates to in multiple linear regression to correct the original gene expression data (the so-called functional genomic mRNA profiles that capture the effects of genomic alterations on gene expression levels). The overexpression percentages of each target antigens in samples per cancer type were determined based on the threshold that was defined in the set of functional genomic mRNA profiles of normal tissues by calculating the 97.5^th percentile for the functional genomic mRNA signal.

Gene set enrichment score analysis

We first downloaded the pan-cancer phenotype data and expression matrix from XENA (https://xenabrowser.net/datapages/?dataset=GDC-PANCAN.basic_phenotype.tsv&host=https%3A%2F%2Fgdc.xenahubs.net&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443, GDC-PANCAN.basic_phenotype.tsv) and the TCGA PanCanAtlas (https://gdc.cancer.gov/about-data/publications/pancanatlas, EBPlusPlusAdjustPANCAN_IlluminaHiSeq_RNASeqV2.geneExp.tsv), respectively. We then extracted the paired samples and their corresponding expression matrix according to the rules of TCGA sample barcode (https://docs.gdc.cancer.gov/Encyclopedia/pages/TCGA_Barcode/). After which the enrichment score of gene set of interest was calculated based on the ssGSEA [36], and the z-score was transformed to evaluate the expression similarities and differences of gene set in TCGA pan-cancer.

Statistical analysis

All statistical analyses described above within the context of individual analyses in the Methods section were carried out using R statistical environment. The non-parametric Mann–Whitney U test and non-parametric Kruskal–Wallis one-way ANOVA were carried out for the analyses of two groups and more than two groups, respectively.

Code availability

The custom computer codes utilized during the current study are available from the corresponding author upon reasonable request.

Acknowledgements: The authors would like to acknowledge the American Association for Cancer Research and its financial and material support in the development of the AACR Project GENIE registry, as well as members of the consortium for their commitment to data sharing. Interpretations are the responsibility of study authors. The authors thank Dr. Simon Wang at the Language Centre, HKBU, for help with an improvement to the linguistic presentations of the manuscript.

Competing interests: Authors declare that they have no competing interests.

WHO. World cancer report 2020 (Imprimerie Faurite, France).
Davis C, Naci H, Gurpinar E, Poplavska E, Pinto A, Aggarwal A. Availability of evidence of benefits on overall survival and quality of life of cancer drugs approved by European Medicines Agency: Retrospective cohort study of drug approvals 2009-13. BMJ. 2017;359:j4530.
Beck A, Goetsch L, Dumontet C, Corvaïa N. Strategies and challenges for the next generation of antibody-drug conjugates. Nat. Rev. Drug Discov. 2017;16:315-337.
Chau CH, Steeg PS, Figg WD. Antibody–drug conjugates for cancer. Lancet. 2019;394:793-804.
Modi S, Saura C, Yamashita T, Park YH, Kim SB, Tamura K, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Positive Breast Cancer. N. Engl. J. Med. 2020;382:610-621.
Lyon R. Drawing lessons from the clinical development of antibody-drug conjugates. Drug Discov. Today Technol. 2018;30:105-109.
Van Geel R, Wijdeven MA, Heesbeen R, Verkade JMM, Wasiel AA, Van Berkel SS, et al. Chemoenzymatic conjugation of toxic payloads to the globally conserved N-glycan of native mabs provides homogeneous and highly efficacious antibody-drug conjugates. Bioconjug. Chem. 2015;26:2233-2242.
Junutula JR, Raab H, Clark S, Bhakta S, Leipold DD, Weir S, et al. Site-specific conjugation of a cytotoxic drug to an antibody improves the therapeutic index. Nat. Biotechnol. 2008;26:925-932.
Panowski S, Bhakta S, Raab H, Polakis P, Junutula JR. Site-specific antibody drug conjugates for cancer therapy. MAbs. 2014;6:34-45.
Strop P, Delaria K, Foletti D, Witt JM, Hasa-Moreno A, Poulsen K, et al. Site-specific conjugation improves therapeutic index of antibody drug conjugates with high drug loading. Nat. Biotechnol. 2015;33:694-696.
Axup JY, Bajjuri KM, Ritland M, Hutchins BM, Kim CH, Kazane SA, et al. Synthesis of site-specific antibody-drug conjugates using unnatural amino acids. Proc. Natl. Acad. Sci. U. S. A. 2012;109:16101-16106.
Tsuchikama K, An Z. Antibody-drug conjugates: recent advances in conjugation and linker chemistries. Protein Cell. 2018;9:33-46.
Donaghy H. Effects of antibody, drug and linker on the preclinical and clinical toxicities of antibody-drug conjugates. MAbs. 2016;8:659-671.
Anami Y, Yamazaki CM, Xiong W, Gui X, Zhang N, An Z, et al. Glutamic acid-valine-citrulline linkers ensure stability and efficacy of antibody-drug conjugates in mice. Nat. Commun. 2018;9:2512.
Devay RM, Delaria K, Zhu G, Holz C, Foletti D, Sutton J, et al. Improved Lysosomal Trafficking Can Modulate the Potency of Antibody Drug Conjugates. Bioconjug. Chem. 2017;28:1102-1114.
De Goeij BECG, Vink T, Ten Napel H, Breij ECW, Satijn D, Wubbolts R, et al. Efficient payload delivery by a bispecific antibody-drug conjugate targeting HER2 and CD63. Mol. Cancer Ther. 2016;15:2688-2697.
Tolcher AW. Antibody drug conjugates: lessons from 20 years of clinical experience. Ann. Oncol. 2016;27:2168-2172.
Saleh MN, Sugarman S, Murray J, Ostroff JB, Healey D, Jones D, et al. Phase I trial of the anti-Lewis Y drug immunoconjugate BR96-doxorubicin in patients with Lewis Y-expressing epithelial tumors. J. Clin. Oncol. 2000;18:2282-2292.
Tijink BM, Buter J, De Bree R, Giaccone G, Lang MS, Staab A, et al. A phase I dose escalation study with anti-CD44v6 bivatuzumab mertansine in patients with incurable squamous cell carcinoma of the head and neck or esophagus. Clin. Cancer Res. 2006;12:6064-6072.
Damelin M, Zhong W, Myers J, Sapra P. Evolving Strategies for Target Selection for Antibody-Drug Conjugates. Pharm. Res. 2015;32:3494-3507.
Arora S, Pattwell SS, Holland EC, Bolouri H. Variability in estimated gene expression among commonly used RNA-seq pipelines. Sci. Rep. 2020;10:2734.
Wang Q, Armenia J, Zhang C, Penson AV, Reznik E, Zhang L, et al. Data Descriptor: Unifying cancer and normal RNA sequencing data from different sources. Sci. Data. 2018;5:180061.
Bagger FO, Kinalis S, Rapin N. BloodSpot: A database of healthy and malignant haematopoiesis updated with purified and single cell mRNA sequencing profiles. Nucleic Acids Res. 2019;47:881-885.
Uhlen M, Zhang C, Lee S, Sjöstedt E, Fagerberg L, Bidkhori G, et al. A pathology atlas of the human cancer transcriptome. Science. 2017;357:eaan2507.
Sweeney SM, Cerami E, Baras A, Pugh TJ, Schultz N, Stricker T, et al. AACR project genie: Powering precision medicine through an international consortium. Cancer Discov. 2017;7:818-831.
Junttila TT, Li G, Parsons K, Phillips GL, Sliwkowski MX. Trastuzumab-DM1 (T-DM1) retains all the mechanisms of action of trastuzumab and efficiently inhibits growth of lapatinib insensitive breast cancer. Breast Cancer Res. Treat. 2011;128:347-356.
Smit EF, Nakagawa K, Nagasaka M, Felip E, Goto Y, Li BT, et al. Trastuzumab deruxtecan (T-DXd; DS-8201) in patients with HER2-mutated metastatic non-small cell lung cancer (NSCLC): Interim results of DESTINY-Lung01. J. Clin. Oncol. 2020;38:9504.
Staudacher AH, Brown MP. Antibody drug conjugates and bystander killing: is antigen-dependent internalisation required. Br. J. Cancer. 2017;117:1736-1742
Perrino E, Steiner M, Krall N, Bernardes GJL, Pretto F, Casi G, et al. Curative properties of noninternalizing antibody-drug conjugates based on maytansinoids. Cancer Res. 2014;74:2569-2578.
Gébleux R, Stringhini M, Casanova R, Soltermann A, Neri D. Non-internalizing antibody–drug conjugates display potent anti-cancer activity upon proteolytic release of monomethyl auristatin E in the subendothelial extracellular matrix. Int. J. Cancer. 2017;140:1670-1679.
Giansanti F, Capone E, Ponziani S, Piccolo E, Gentile R, Lamolinara A, et al. Secreted Gal-3BP is a novel promising target for non-internalizing Antibody–Drug Conjugates. J. Control. Release. 2019;294:176-184.
Perna F, Berman SH, Soni RK, Mansilla-Soto J, Eyquem J, Hamieh M, et al. Integrating Proteomics and Transcriptomics for Systematic Combinatorial Chimeric Antigen Receptor Therapy of AML. Cancer Cell. 2017;32:506-519.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.
Hu Y, Yan C, Hsu CH, Chen QR, Niu K, Komatsoulis GA, et al. Omiccircos: A simple-to-use R package for the circular visualization of multidimensional Omics data. Cancer Inform. 2014;13:13.
Fehrmann RSN, Karjalainen JM, Krajewska M, Westra HJ, Maloney D, Simeonov A, et al. Gene expression analysis identifies global gene dosage sensitivity in cancer. Nat. Genet. 2015;47:115-125.
Hänzelmann S, Castelo R, Guinney J. GSVA: Gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics. 2013;14:7.

There is NO conflict of interest to disclose.

Download PDF

Editorial decision: revise
04 Jul, 2023
Review #2 received at journal
23 May, 2023
Review #3 received at journal
23 May, 2023
Review #1 received at journal
13 May, 2023
Reviewer #4 agreed at journal
12 May, 2023
Reviewer #3 agreed at journal
10 May, 2023
Reviewer #2 agreed at journal
09 May, 2023
Reviewer #1 agreed at journal
08 May, 2023
Reviewers invited by journal
06 May, 2023
Submission checks completed at journal
05 May, 2023
First submitted to journal
04 May, 2023
Unknown event
02 May, 2023
Editor assigned by journal
02 May, 2023

You are reading this latest preprint version

The Target Atlas for Antibody-Drug Conjugates across Solid Cancers

Status:

Version 1

Abstract

Figures

1 Introduction

2 Results

2.1 Assembling a Comprehensive Dataset and Designing an Algorithm to Identify ADC Targets

2.2 Identified ADCs and Their Target Antigens in Clinical Trials

2.3 Differential Gene Expression Patterns Between Tumour and Normal Tissues

2.4 Heterogeneous Gene Expression Patterns in Solid Cancers

2.5 Predicted Protein Overexpression Rates of Target Antigens

2.6 Genetic Basis of Target Antigen Candidates

3 Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1