ADCs: a more effective, less harmful, targeted approach

Antibody-drug conjugates could hold the key to delivering highly potent anticancer drugs more accurately to the tumour site with less damage to healthy cells

Antibody-drug conjugates (ADCs) have generated a lot of interest in recent years because of their potential to target precisely highly potent cytotoxic drugs to tumour cells. Not only should they be more efficacious, the hope is that the side-effect profile will also be improved as they are less likely to damage healthy cells en route to the tumour.

The first of these to reach the market, Mylotarg (gemtuzumab ozogamacin) from Wyeth (now Pfizer), was launched in 2000 but withdrawn a decade later because of lack of efficacy. Since then, however, two more ADCs have reached patients – lymphoma treatment Adcetris (brentuximab vedotin) from Seattle Genetics, and Kadcyla (ado-trastuzumab emtansine) from Genentech, which is indicated in HER2+ breast cancer.

An ADC comprises an antibody designed to target tumour cells and a highly potent API, held together by a linker moiety. Two linkage technologies have predominated thus far, one developed by Seattle Genetics, and the other by ImmunoGen and used in the Genentech product. Both have their advantages and disadvantages, and various alternative linking technologies are under development.

Dozens of new ADCs are currently in the clinical pipeline, predominantly for treating various forms of cancer. Unsurprisingly, as one of the early players in the market whose technology is already proven in a commercial product, ImmunoGen has multiple ADCs in the clinic, both in-house and with partners. These include two in-house products that are in the late stages of development.

One of these, mirvetuximab soravtansine, is under development for solid tumours that are folate receptor alpha-positive. The lead indication is ovarian cancer, for which both the US FDA and EMA in Europe have granted the product orphan status. Trials are also under way in peritoneal, fallopian tube and endometrial cancers.

Early Phase I evaluation data indicated that doses of 6mg/kg every 3 weeks had antitumour activity in heavily pretreated patients with platinum-resistant epithelial ovarian cancer or relapsed/refractory endometrial carcinoma.1 Half of the 10 patients assessed achieved some clinical benefit.

It is in Phase III as a single agent, and is also being assessed in a Phase Ib study in combination with other chemotherapy agents, including carboplatin, doxorubicin and bevacizumab (Avastin). The company is also embarking on a joint trials programme with Merck where the ADC is dosed in combination with pembrolizumab (Keytruda) in FRa+ ovarian cancer.

ImmunoGen’s second well-advanced product, coltuximab ravtansine, was previously being developed in collaboration with Sanofi, but ImmunoGen regained the rights in April 2015. The antibody targets CD19, and is designed to treat diffuse large B-cell lymphoma (DLBCL) and other B-cell malignancies. The CD19 antigen is expressed in the majority of B-cell lymphomas. The ravtansine moiety is a microtubule-binding maytansinoid, DM4.

An ADC comprises an antibody designed to target tumour cells and a highly potent API, held together by a linker moiety

Phase II results of a Sanofi-funded study in diffuse large B-cell lymphoma (DLBCL) were reported at ASCO in 2014, with the ADC already having shown clinical activity in patients with both indolent and aggressive lymphomas in a Phase I trial.2 Patients who had relapsed or refractory DLBCL after at least one standard treatment including rituximab, and who were not transplantation candidates, were given 55mg/m2 of the ADC weekly for 4 weeks, and then biweekly until disease progression (or other discontinuation criterion) occurred. Of the 41 evaluable patients, a third of whom had received at least three prior regimens, the objective response rate was 44%, including five complete responses. It had an acceptable safety profile.

Although discontinued by Sanofi, ImmunoGen says it is doing preclinical evaluations of this ADC in combination with marketed anticancer agents, and hopes to initiate further clinical trials some time in 2016. As well as multiple further trials on brentuximab vedotin in a range of other indications, Seattle Genetics has two additional ADCs in the later stages of clinical development, and several more in Phase I. The furthest advanced, vadastuximab talirine, is in Phase II trials for acute myeloid leukaemia, with a Phase III study as a frontline treatment planned. It is also being evaluated in myelodysplastic syndrome.

The antibody part of this ADC targets the antigen CD33, which is expressed on the majority of AML cells. It was engineered so that it has two site-specific cysteine residues for linker attachment, giving an average of two HPAPIs on each antibody. The HPAPI portion is a pyrrolobenzodiazepine dimer that acts as a DNA binding agent, and the two are joined using a maleimidocaproyl linker.

Also in Phase II is denintuzumab mafodotin. This uses a CD19-targeting antibody and, like the ImmunoGen product, is being investigated in a range of B-cell malignancies. The HPAPI is the synthetic tubulin inhibitor monomethyl auristatin F (MMAF), again attached via a maleimidocaproyl linker. Early results from a Phase I, open-label, dose-escalation study in patients with relapsed or refractory B-cell non-Hodgkin lymphoma indicated that eight of 20 evaluable subjects had achieved an objective response, with six complete responses, two partial. Three further patients achieved stable disease. No dose-limiting toxicities were observed.3

Seattle Genetics is also working with various other companies to apply its ADC technology, including Agensys, which is an affiliate of Astellas. The furthest advanced product in this collaboration is enfortumab vedotin, which is in Phase I trials to treat solid tumours. The antibody portion targets nectin-4, found on various different tumours, including bladder, breast, lung and pancreatic, and the HPAPI is momomethyl auristatin E (MMAE), the same microtubule disrupting agent found in Adcetris. They are joined using the enzyme-cleavable maleimidocaproyl-valyl-citrullinyl-p-aminobenzyloxycarbonyl linker technology. Also in Phase I is ASG-15ME, which combines MMAE with an antibody that targets SLITRK6. This is found in bladder cancers.

Numerous big pharma companies are also investing in the ADC arena

Genentech is continuing to investigate ADCs, following the successful launch of Kadcyla. The same HPAPI, MMAE, is incorporated in polatuzumab vedotin. Again, this is constructed using the Seattle Genetics technology, only here it is combined with a monoclonal antibody directed against CD79b on haematologic cells.

A Phase II trial in combination with rituximab for relapsed or refractory follicular non-Hodgkin lymphoma and diffuse large B-cell lymphoma is under way.4 Preliminary results showed six complete and six partial responses out of 20 evaluable follicular lymphoma patients, and five complete and 14 partial out of 54 patients with diffuse large B-cell lymphoma. This trial is also evaluating pivolumab vedotin, which incorporates an antibody that targets CD22. In those preliminary results, the objective response rates of the two ADCs were similar, but the complete response rate was greater for polatuzumab vedotin.

Numerous big pharma companies are also investing in the ADC arena. Bayer, for example, is working on anetumab ravtansine as a mesothelioma treatment. The antibody targets mesothelin, and it is joined via a SPDB linker to the maytansine microtubule inhibitor DM4. It is currently in Phase I trials. Another ADC project, BAY79-3620, using monomethyl auristatin as the HPAPI, targets CA9 on various solid tumours, with a valine-citrulline linker.

AbbVie’s ABT-414 is another MMAF-containing ADC. Its targeting antibody is directed towards EGFR, and this product is in Phase II trials for recurrent glioblastoma multiforme. Early Phase I results showed promise.5 Among 18 patients given the ADC plus temozolomide, there was one complete response and four partial ones; in the trial arm where the patients were given the ADC as monotherapy, there was one complete and one partial response in 28 patients. All of the patients who exhibited a confirmed response had EGFR amplification.

Millennium’s indusatumab vedotin is in clinical trials for gastrointestinal, pancreatic and colorectal cancers. Here, the antibody is targeting malignant cells that express the surface antigen guanylyl cyclase C. It is conjugated to MMAE. Initial results of a first in human study in patients with a variety of gastrointestinal malignancies have been reported.6 Subjects were given the ADC by intravenous infusion on day 1 of 21 day cycles in a dose escalation study, and once the maximum tolerated dose was established patients with metastatic colorectal cancer were enrolled into an expansion cohort. Of the 33 patients who had been dosed and evaluated, most of whom had colorectal cancer, one with gastric carcinoma achieved a partial response, and overall 13 achieved stable disease. It was generally well tolerated.

Immunomedics has developed its own linker technology, termed CL2A. This includes a short polyethylene glycol chain to improve solubility, and the attachment to the HPAPI is pH sensitive, giving detachment at a rate of about 50% a day in vivo in acid or alkaline conditions. It was designed for the delivery of SN38, a derivative of irinotecan that is approximately three orders of magnitude more potent than the parent compound. The ADCs created using this technology have a relatively high average drug-antibody ratio of seven.

The antibody bevacizumab

Two ADCs with SN38 are in Phase I. Sacituzumab govitecan combines it with the antibody hRS7, which targets the cell surface receptor TROP-2. This is overexpressed in a range of tumours, including breast, colon and lung cancers. Phase I/II trials have been reported in gastrointestinal cancers, metastatic lung cancer, and triple-negative breast cancer, and in all cases the results were sufficiently good to warrant further investigation – in the breast cancer study, for example, 34 patients were assessed for response, and seven had an objective response, including one complete response.7

Labetuzumab govitecan, meanwhile, incorporates an anti-CEACAM5 antibody and is being investigated in metastatic colorectal cancer. In a Phase I/II trial, subjects with relapsed or refractory disease plus elevated CEA were given once or twice-weekly escalating doses as monotherapy on weeks one and two of 3-week cycles.8 Of the 58 evaluable patients, three experienced dose-limiting toxicities, and a third experienced tumour reductions, including one partial response. And in the GI cancer trial, one of 26 colorectal cancer patients achieved a partial response, seven of 14 with pancreatic cancer achieved stable disease, there were two partial responses and seven stable disease outcomes out of 14 patients with oesophageal cancer, and all three with gastric cancer achieved stable disease.9

Rovalpituzumab tesirine, or Rova-T, is the furthest advanced ADC project at South San Francisco-based biotech StemCentrx. This ADC combines an antibody directed against DLL3 with a DNA-damaging pyrrolobenzodiazepine dimer toxin payload antibody, attached via a maleimide-containing linker containing an eight-carbon polyethylene glycol spacer, a cathepsin B-cleavable valine-alanine dipeptide, and a self-immolating group, with a mean drug-to-antibody ratio of 2.10

DLL3 is a ligand that is related to those involved in the Notch pathway which, unlike those other ligands, does not activate Notch signalling. Its expression is increased in small cell lung cancer and large cell neuroendocrine carcinoma, and preclinical studies indicated that the ADC caused in vivo tumour regression. A Phase II study to evaluate its efficacy, safety and pharmacokinetics for the third line and later treatment of patients with relapsed or refractory DLL3 expressing small cell lung cancer is under way. In addition, a Phase I/II trial in a range of solid tumours, including malignant melanoma, medullary thyroid cancer, glioblastoma, large cell neuroendocrine carcinoma and neuroendocrine prostate cancer, among others, is being planned.

Millennium’s indusatumab vedotin is in clinical trials for gastrointestinal, pancreatic and colorectal cancers

Millennium’s indusatumab vedotin is in clinical trials for gastrointestinal, pancreatic and colorectal cancers

Dutch biotech Synthon is developing SYD985, which targets HER2+ breast cancer. It uses the established antibody trastuzumab (the same antibody that forms part of Kadcyla), which is conjugated to the HPAPI duocarmycin via the company’s own proprietary linker technology. Duocarmycin is a natural product extracted from Streptomyces bacteria in the 1980s that binds to the minor groove of the DNA architecture, and causes irreversible alkylation of the DNA. In contrast to the microtubule disrupting HPAPIs, it can act at any point in the cell cycle, and not just during mitosis.

Initial Phase I results have been reported.11 Patients refractory to trastuzumab and trastuzumab emtansine with locally advanced or metastatic solid tumours were given the ADC. At the time of reporting 12 patients had been enrolled, including seven with breast cancer, three with colorectal cancer, one with gastric cancer and one with adenoid cystic carcinoma of the parotid gland. Of the nine patients who had received at least two cycles of treatment, two patients with HER2+ breast cancer who had been pretreated with multiple HER2+-directed therapies, including Kadcyla, showed a partial response. Five had stable disease as the best response.

Progenics’ PSMA ADC incorporates a monoclonal antibody that binds to prostate specific membrane antigen, present on the surface of most prostate cancers but very few healthy tissues. This is conjugated to MMAE. Activity and tolerability were exhibited in a Phase I trial, and early results from a Phase II study showed promise.12 In all, 119 patients with progressive metastatic castration resistant prostate cancer, 83 of whom had previously received taxane therapy, were given the ADC in three-week cycles for up to eight cycles. Most had also previously been treated with abiraterone and/or enzalutamide. In the 31 patients with measurable target lesions, four achieved a partial response and 19 stable disease; clinically significant adverse events included neutropenia and neuropathy.

Also targeting HER2+ cancers is Ambrx’s ARX788. This combines a HER2-specific monoclonal antibody that is site-specifically conjugated with the potent cytotoxic tubulin inhibitor amberstatin269. It has shown promise in preclinical studies, with efficacy in controlling xenograft tumours. The company is currently recruiting for a first in human Phase I study in patients with HER2+ breast cancer. This will be run in Australasia in collaboration with Chinese company Zhejiang Medicine.


1. K.N. Moore, et al., J. Clin. Oncol. 2015, 33(suppl), Abst. 5518.

2. M. Trneny, et al., J. Clin. Oncol. 2014, 32(suppl), Abst. 8506.

3. A. Forero-Torres, et al., J. Clin. Oncol. 2014, 32(suppl), Abst. 8505.

4. F. Morschhauser, et al., J. Clin. Oncol. 2014, 32(suppl), Abst. 8519.

5. H.K. Gan, et al., J. Clin. Oncol. 2015, 33(suppl), Abst. 2016.

6. C.C. Zambrano, et al., J. Clin. Oncol. 2014, 32(suppl), Abst. 3546.

7. A. Starodub, et al., J. Clin. Oncol. 2015, 33(suppl), Abst. 1016.

8. E. Dotan, et al., J. Clin. Oncol. 2015, 33(suppl), Abst. 2505.

9. M.D. Glassy, et al. J. Clin. Oncol. 2015, 33(suppl), Abst. 357.

10. L.R. Saunders, et al., Sci. Transl. Med.. 2015, 7: 302ra136.

11. C.M.L. van Herpen, et al., Eur. Cancer Congr. (Vienna, September 2015), Abst. 333.

12. D.P. Petrylak, et al., J. Clin. Oncol. 2015, 33(suppl), Abst. 144.