Improved Characterization of Antibody-Drug Conjugates

Antibody-drug conjugates (ADCs) are an emerging class of therapeutics, comprising a monoclonal antibody linked to a small molecule drug. The ability to target cytotoxic effects to a subset of cells while sparing others makes them an attractive treatment option for diseases such as cancer. However, analysis of ADCs – a key step in both pre-clinical development and quality control – can be difficult and there is a need for improved analysis methods. 

In a new research study published in Analytical Chemistry, scientists from the University of Wisconsin showcase a novel mass spectrometry-based method that enables high-throughput characterization of ADCs.

Technology Networks spoke to Eli J. Larson, lead author of the paper and a PhD candidate in the Ge Lab, to learn more about the method and some of the benefits it offers over existing analytical techniques. In this interview, Eli also explains how ADCs are being applied as anticancer therapies and highlights key challenges faced when ensuring their safety and efficacy.

Ash Board (AB): How are ADCs being applied as anticancer therapies and what are the key benefits?

Eli J. Larson (EL): At this time, the most common application for ADCs in anticancer treatment is as a drug of last resort. Ten of the 11 currently FDA approved ADCs are used to treat patients not responding to traditional anticancer therapies or patients with relapsed or refractory cancers. ADCs are the combination of two therapeutic moieties – small molecule drugs and monoclonal antibodies (mAbs). Small molecule drugs display high levels of cytoxicity and effectively kill cancerous cells; however, they often produce deleterious side effects for patients. Conversely, mAbs exhibit lower cytoxicity levels than small molecule drugs, but the targeted nature of mAbs limits patient side effects. Combining these moieties in an ADC enables the strengths of both therapeutic entities to be preserved while the weaknesses are diminished, making ADCs “magic bullets” for anticancer applications.

AB: What are some of the challenges of ensuring safety and efficacy of ADCs?

EL: Two key challenges to the analysis of ADCs are the molecular complexity and size of ADCs. Any antibody-based therapeutic must assess the myriad of factors known to effect pharmacokinetics, binding efficacy and stability. These include primary sequence variants, C-terminal lysine clipping, deamidation, glycosylation variants, oxidation and many others. The drug conjugation process adds complexity resulting from varied drug-to-antibody ratios (DAR), drug positional isomers due to stochastic drug conjugation, and possible conjugation in unintended regions of the mAb such as the complimentary determining region. For analyses using mass spectrometry (MS) detection, the signal-to-noise of a given analyte decreases with increasing analyte size. ADCs are relatively large molecular entities, approximately 150 kDa in mass, which can make analysis of ADCs difficult and specifically hinder detection of low abundance variants of the ADC.

AB: What are some of the common methods used to assess the quality attributes of ADCs. Are there any drawbacks associated with these methods?

EL: The most commonly used methods for ADC analysis are bottom-up liquid chromatography coupled to tandem MS (LC-MS2) and middle-down LC-MS2. Bottom-up analysis uses extensive enzymatic digestion with proteases such as trypsin or Lys-C to produce peptide fragments of the ADC (< 3 kDa) prior to LC-MS2 analysis. The bottom-up digestion process is generally a time-consuming overnight reaction and may induce artifactual modification to the ADC during the digestion and sample preparation process. Additionally, ionization efficiency differences between modified and unmodified peptides makes quantitation of global molecular features – such as glycoform variants and average DAR – difficult to achieve. Middle-down LC-MS2 offers a more global perspective than bottom-up, opting to digest using a specific protease like IdeS or KGP followed by chemical reduction to produce subunits ~ 25 kDa in size. This allows facile quantitation of global modification variants such as glycoforms and average DAR by LC-MS, but the sample preparation time and LC-MS2 analysis time is still ~ 3 hours long. Other methods such as hydrophobic interaction chromatography coupled to UV detectors enable determination of DAR from the intact ADC, foregoing the lengthy sample digestion procedure, but buffer conditions are MS-incompatible and further quality attributes cannot be monitored easily.

AB: Your group has developed a rapid and high-throughput method for the multi-attribute analysis of cysteine ADCs. Can you explain how this was achieved?

EL: Our goal when designing this method was to develop a rapid diagnostic for ADCs to inform users whether or not further characterization of the ADC may be needed. In order to make this method truly rapid we needed to: 1) Reduce or eliminate sample preparation time, 2) Reduce or eliminate time needed for front-end separations, 3) Improve MS sensitivity and signal-to-noise (S/N), and 4) Increase the breadth of information collected during the MS analysis. To achieve this, we chose to use top-down mass spectrometry, which does not use enzymatic digestion prior to MS analysis, eliminating the single greatest contribution to sample preparation time. Because the model cysteine-linked ADC is held together entirely by non-covalent interactions for certain DAR values, our only sample preparation step is dilution of the ADC in non-denaturing, MS-compatible buffer. To eliminate the need for LC prior to MS, we used gas-phase separation in the form of trapped ion mobility spectrometry (TIMS). TIMS allows the collisional cross section (CCS) of the analytes to be measured and provides a boost in sensitivity through improved S/N ratio. While the use of TIMS already increases the breadth of information collected relative to conventional methods, increasing the in-source collision-induced dissociation (CID) and collision cell CID energies enable detection of the intact ADC mass (MS1), intact ADC subunit mass (MS2), and primary sequence analysis (MS3). TIMS combined with a three-tiered MS approach provides greater breadth of information than traditional methods, specifically through collection of intact ADC mass and determination of CCS values for the ADC. Although previous techniques can provide greater primary sequence coverage than our developed method, our technique provides the information needed for users to determine whether or not a more lengthy analytical method is warranted.

AB: What benefits does this provide over existing analytical techniques used to analyze ADCs?

EL: The major advantages our method has over traditional techniques are speed of analysis, and breadth of information provided. While previous techniques have used lengthy digestion procedures and one to two hour LC-MS, our method can be applied by simply diluting the analyte in MS-compatible buffer like ammonium acetate, making the analysis duty cycle functionally equal to the instrumental analysis time. The addition of CCS and intact ADC mass to the subunit mass and primary sequence characterization performed in traditional methods offer greater breadth of information in less time than traditional techniques. A distinctive minor advantage of our method is the use of an unmodified, commercially available instrument. This instrument is widely being used to perform bottom-up LC-MS2 analyses and our method shows that the capabilities of this instrument are much greater than many users may be aware of.

AB: How can this approach be used for the characterization of antibody-based therapeutics in pre- and post-clinical quality control?

EL: The technique can be used in both pre- and post-clinical quality control to rapidly monitor batch-to-batch variability in antibody-based therapeutics The results generated from this method provide a roadmap of therapeutic attributes which shows practitioners whether further molecular detail is needed. I envision that someday this method may be used in an online continuous bioprocessing workflow thanks to the speed, breadth of information provided, and ease of interfacing with automated sample handling devices.

Eli J. Larson was speaking to Ash Board, Editorial Director for Technology Networks.