2020 was a remarkable year in science, not only for COVID research and but also in the cancer research community’s quest to prevent, detect and treat cancers. Below is my ‘top ten’ list of recent notable advances:
(1) COVID-19 Vaccines:
Although this is not cancer-specific, the development of vaccines against COVID-19 unquestionably helps cancer patients, whose illnesses and often immune suppression make them vulnerable to serious complications and possible death from the virus. [1, 2]
(2) 2020 is Also the Year for Advances in CAR-T Therapy:
New T cell therapy: Current T cell therapy (CAR-T) involves collecting, modifying and expanding a patient’s T cells to attack their specific cancer. The treatment must be customized to each patient because the T cell receptor (HLA), which enables T cells to detect cancer cells, varies within the population. A recent study has identified a T cell receptor that may result in a new T cell therapy. Similar to HLA, this molecule (MR1) recognizes cancer cells, but does not vary among individuals. This T cell receptor does not respond to healthy cells but confers HLA-independent recognition to a wide range of cancer cells. This finding opens opportunities for T cell–mediated cancer immunotherapy and perhaps therapeutic vaccinations in a variety of cancer. [3]
Use of upfront CAR-T in Lymphoma: An interim report of the phase II ZUMA-12 trial showed that treatment with axicabtagene ciloleucel resulted in a high response rate in patients with high-risk large B cell lymphoma (LBCL) who showed early resistance to first-line chemoimmunotherapy. Of 27 LBCL patients, 85% responded to the anti-CD19 CAR-T therapy, including 74% being complete responders. [4]
(3) Machine Learning (ML) Achievements in Cancer Care and Cancer Research:
At last count, the FDA has authorized approximately 30 ML-enabled devices. Most of these are related to image analysis like reading mammograms, but also some very interesting uses of ML for more basic investigation:
AlphaFold: Machine learning was used successfully for the first time to deduce the structure of proteins from their genetic code in a way that approximates to what experimental techniques can achieve. Structural biologists say the AlphaFold system, developed by London-based AI powerhouse DeepMind, a sister company of Google, could be a game-changer for drug discovery. [5]
Use of Deep Learning to predict drug responses: Drugs often fail in clinical trials due to lack of intricate mechanistic understanding. Machine learning algorithms enable learning from large-scale data collected from such complex biological responses. DrugCell is a ML model of human cancer that was trained on the response of more than 1200 tumor cells to almost 700 drugs which gathers information on drug structure and therapy response and learns about the mechanisms fundamental to those responses. This technology enables the development of improved models for pre-clinical drug evaluation. [6]
Use of AI-based model to make optimal tyrosine kinase inhibitor therapy selection in CML: Although the majority of CML patients have long-lasting response to their initial BCR-ABL1 tyrosine kinase inhibitor (TKI) therapy, a significant number require shift to different TKI at some point during the course of their treatment. In this study, data from 500 newly diagnosed patients were used to train and validate the LEAP (Leukemia Artificial intelligence Program) model, including a wide range of patient-related factors and laboratory results. [7]
Use of AI to study RAS structural biology: Although published in 2019, this work became more impactful in 2020. RAS gene mutations drive colon, pancreas and lung cancers, and are associated overall with 30% of all cancers. Even so, much remains to be understood related to its molecular mechanisms and interactions with downstream signaling effectors in the complex cellular membrane environment. In a joint NCI/DOE effort to advance computing and data technologies to accelerate cancer research, the goal of Pilot Two, “Improving Outcomes for RAS-related Cancer,” is to generate a model of RAS biology which captures lipid environment changes, lipid-lipid interactions, protein behavior and protein-lipid interactions at the molecular level. This model will enable the largest and most refined computational investigation of RAS structural biology to date. [8]
(4) New Cancer Blood Tests:
New cancer blood tests have been developed that may improve early detection efforts:
cfDNA detection: Researcher reported a blood test that was able to detect more than 50 different cancer types, with a very low (< 1%) false positive rate. The test predicted the tissue in which the cancer originated in 96% of samples, with an accuracy of 93%. The test works by analyzing aberrant methylation patterns of circulating cell-free DNA (cfDNA) — which subsequently changes gene expression to drive tumor growth — to detect and localize multiple cancer types across all stages of disease. [9]
Detecting cancer four years earlier than standard diagnoses: A non-invasive blood test has been developed that is able to detect whether a person has one of five common types of cancer (stomach, esophageal, colorectal, lung and liver) four years before the condition can be diagnosed with other standard diagnostic methods. This test works by using targeted methylation analysis of cancer-specific methylation signatures on circulating tumor DNA. [10]
(5) Cryo-EM:
The technique of cryo-electron microscopy (cryo-EM) reached single-atom resolution, approaching the precision of the well-established — but more cumbersome — X-ray crystallography. This breakthrough will make it easier to understand the structure of proteins, which will improve drug discovery efforts. [11]
(6) Targeting KRASG12C in Non-Small Cell Lung Cancer:
Thirteen percent of non-small cell lung cancers (NSCLC) contain the KRASG12C mutation. A recent phase I clinical trial in NSCLC patients of sotorasib, a selective small molecule that irreversible binds to and inhibits KRASG12C, showed almost complete absence of side effects due to the level of drug specificity. This is a game changer for lung cancer patients. [12]
(7) Multiple Myeloma:
The FDA approved selinexor (XPOVIO; Karyopharma) for combination treatment with bortezomib and dexamethasone for patients with refractory or relapsed multiple myeloma. Selinexor is a new class of drug that selectively inhibits nuclear export, resulting in accumulation of tumor suppressor proteins. In the phase III BOSTON trial, adding selinexor to weekly bortezomib (Velcade) plus dexamethasone improved progression-free survival over standard of care twice-weekly bortezomib plus dexamethasone (13.9 vs 9.5 months). [13]
(8) Her-2/TLR8 ADC Proof-of-Concept in Phase I Trials:
Antibody-drug conjugates (ADCs) are novel drugs composed of a monoclonal antibody for an antigen expressed on a cancer cell linked to a cytotoxic agent. ADCs exploits the specificity of antibodies to target chemotherapy delivery. A first-in-human Phase I trial of SBT6050, a novel HER2-TLR8 ADC, in being tested in in adult breast cancer patients with moderate or high HER2-expressing solid tumors. [14]
(9) Combination of Venetoclax plus Ibrutinib for CLL:
In a phase II trial presented at the American Society of Hematology (ASH) virtual meeting, the addition of venetoclax to ibrutinib as a consolidation therapy was associated with high rates of undetectable minimal residual disease (MRD) in bone marrow and complete response within 12 months among patients with high-risk chronic lymphocytic leukemia. [15]
(10) Triple Combination Therapy for CRC:
Metastatic colorectal cancer has a poor prognosis. The BEACON CRC study compared a triplet regimen of encorafenib (ENCO) + binimetinib (BINI) + cetuximab (CETUX) and a double regimen (ENCO + CETUX) to standard of care (irinotecan + CETUX or FOLFIRI + CETUX) in patients with BRAF V600E metastatic colorectal cancer. The 3-drug combination significantly improved overall survival (hazard ratio: 0.52) and objective response rates (26% vs 2%), as well as patient reported quality of life variables. [16; 17]
References:
- U.S. Food & Drug Administration. COVID-19 vaccines. https://www.fda.gov/emergency-preparedness-and-response/coronavirus-disease-2019-covid-19/covid-19-vaccines
- Centers for Disease Control and Prevention. Facts about COVID-19 Vaccines. https://www.cdc.gov/coronavirus/2019-ncov/vaccines/facts.html
- Michael D. Crowther MD, et al. Genome-wide CRISPR–Cas9 screening reveals ubiquitous T cell cancer targeting via the monomorphic MHC class I-related protein MR1. Nature Immunology. 2020; 21:178–185.
- Neelapu SS, et al. “Interim analysis of ZUMA-12: A phase 2 study of axicabtagene ciloleucel (axi-cel) as first-line therapy in patients with high-risk large B cell lymphoma.” ASH 2020; Abstract 405.
- Callaway E. “It Will Change Everything”: AI makes gigantic leap in solving protein structures. Nature. 2020; December 10; Vol 588.
- Kuenzi BM, et al. Predicting drug response and synergy using a deep learning model of human cancer cells. Cancer Cell. 2020; 38(5): 672-684.e6.
- Sasaki K, et al. The LEukemia Artificial Intelligence Program (LEAP) in chronic myeloid leukemia in chronic phase: A model to improve patient outcomes. Am J Hematol. 2020; doi: 10.1002/ajh.26047. Online ahead of print.
- Bhattacharya T, et al., AI Meets Exascale Computing: Advancing cancer research with large-scale high performance computing. Front Oncol. 2019; 9:984. doi: 10.3389/fonc.2019.00984. eCollection 2019.
- Lui MC, et al. Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Annals of Oncology. 2020; 31: 745-759.
- Chen, X., et al. Non-invasive early detection of cancer four years before conventional diagnosis using a blood test. Nat Commun. 2020; 11: 3475.
- Callaway E. Revolutionary microscopy technique sees individual atoms. Nature. 2020; June 11; Vol 582.