Scientifically, 2015 was the year of DNA repair.

In October, the Royal Swedish Academy of Sciences awarded the Nobel Prize in Chemistry for 2015 to Tomas Lindahl, Paul Modrich and Aziz Sancar "for mechanistic studies of DNA repair."

It was the second major research prize of the year to be awarded for DNA repair – in September, the Lasker Foundation awarded the 2015 Albert Lasker Basic Medical Research Award to Stephen Elledge and Evelyn Witkin "for discoveries concerning the DNA-damage response — a fundamental mechanism that protects the genomes of all living organisms."


The DNA damage response is a broader phenomenon that encompasses cell cycle checkpoints as well as DNA repair. But both awards honored the area of DNA repair.

Depending on who's doing the counting, there are five or six major DNA repair pathways. And "to a first approximation, each of them is a series of enzymatic steps," Alan D'Andrea told BioWorld Insight. D'Andrea is the Alvan T. and Viola D. Fuller American Cancer Society Professor of Radiation Oncology at Harvard Medical School.

Each of those enzymatic steps, in turn, is a potential drug target. Because cancer cells are both defective in DNA repair and hyperdependent on what capacities they have left.

Normal cells are fastidious about keeping their DNA in good working order. But cancer cells, D'Andrea said, "often unload some of those repair mechanisms."

That unloading is what enables further mutations, including what are generally considered driver mutations. Defects in DNA repair are early events in malignant transformation.

Still, D'Andrea said, "even the cancer cell doesn't completely want to go without DNA repair." And so, with one pathway defective, it becomes "hyperdependent" on other such pathways.

With a slight tolerance for fuzzy math, one could argue that the year of DNA repair began in December 2014 with the FDA's approval of Lynparza (olaparib, Astrazeneca plc.) for the treatment of advanced ovarian cancer with germline BRCA mutations.

Lynparza is the first drug targeting one of the major DNA repair mechanisms, namely base excision repair. More specifically, the drug inhibits the enzyme poly-ADP ribose polymerase (PARP). Part of the effect of PARP "inhibitors" also appears to stem not so much from inhibiting the enzyme as from trapping it on the DNA.

"Normal cells can do reasonably well without base excision repair for a period of time," D'Andrea explained. But cells with certain BRCA mutations are already defective in homologous recombination, another major DNA repair pathway. Knocking out base excision repair is lethal to cells with certain BRCA mutations.


Cells whose homologous recombination abilities are compromised become highly dependent on base excision repair to keep themselves going and, ultimately, individuals with germline BRCA mutations have a high risk of developing breast or ovarian cancer.

Lynparza targets that dependence, blocking the base excision repair that is temporarily dispensable for normal cells but is critical for the survival of cells with BRCA lesions.

There are a half dozen PARP inhibitors in clinical trials. (See BioWorld Today, Nov. 10, 2015.)

But with hundreds of proteins being important for DNA repair, there are plenty of other opportunities for targeting vulnerabilities, exploiting other synthetic lethal interactions.

Famously, DNA is a double helix. And when it breaks, it can break on one strand, or two.

DNA repair-targeting strategies work by trying to make breaks worse – turning the single-stranded breaks that can be dealt with by base excision repair, nucleotide excision repair or mismatch repair into double-stranded breaks that are dealt with by homologous recombination or nonhomologous end joining. The ultimate goal is not to repair the DNA, but to damage it to the point that the cell throws in the towel, succumbing to what's colorfully known as "mitotic catastrophe."


Homologous recombination is used mainly during mitosis because it depends on the presence of a template strand of DNA that has the correct information on it.

When no such template is present, cells use another form of DNA repair, the more error-prone nonhomologous end-joining or alternative end-joining.

Speaking at the joint Molecular Targets meeting of the American Association for Cancer Research, the National Cancer Institute and the European Organisation for Research and Treatment of Cancer (AACR-NCI-EORTC) in November, D'Andrea told the audience that with alternative end-joining, "you get repair, but you get repair at a cost."

That cost is insertions and deletions, because in the absence of a template, the process is highly error-prone.

D'Andrea and his group are investigating the potential of inhibiting polymerase Q (PolQ), which is strongly up-regulated in tumors that are deficient in homologous recombination, including some ovarian cancers.

There is synthetic lethality between homologous recombination defects and PolQ because, D'Andrea said, "these cells cannot sustain a double hit. They cannot sustain the loss of [homologous recombination] and alternative end-joining."


Clinical-stage drugs targeting nonhomologous end-joining are Celgene Corp.'s CC-115 and Merck KGaA's M3814 (MSC 2490484A), both of which target the catalytic subunit of DNA-dependent protein kinase (DNA-PKc).

M3814 is in phase I trials as a monotherapy for chronic lymphocytic leukemia (CLL) and in combination with radiation for non-small-cell lung cancer and lymphoma, while CC-115 is in phase I trials in CLL, non-Hodgkin lymphoma and advanced solid tumors.

Astrazeneca plc.'s AZD-6738 and Vertex Pharmaceuticals Inc.'s VX-970 target homologous recombination. Both drugs target the ATR kinase, a damage-signaling kinase whose activity is increased at stalled DNA replication forks. VX-970 is in phase I and II trials in multiple tumors, both alone and in combination with radiation or chemotherapy.

AZD-6738 is being tested alone and in multiple combinations, including with Lynparza and with cisplatin, a chemotherapy that induces DNA damage by crosslinking DNA bases to each other, a type of damage that is dealt with by nucleotide excision repair.

Beyond the development of new drugs, a better understanding of DNA repair might also help predict the response to existing nontargeted cancer drugs, which work by inducing DNA damage. Platinum-based drugs are a mainstay of chemotherapy, but no good biomarkers exist to determine which patients will respond to them and which will get nothing but toxicity.

Recent data from The Cancer Genome Atlas (TCGA) project has shown that many patients with ovarian cancer have defects in genes involved in Fanconi anemia, an autosomal recessive bone marrow failure that also predisposes patients to certain types of cancer.

Fanconi anemia is characterized by defects in DNA repair, although those defects cannot be neatly equated with one DNA repair pathway. Nevertheless, in TCGA studies, defects in the Fanconi anemia pathway predicted sensitivity to cisplatin, and of tumors with no defects in that pathway but nevertheless sensitive to cisplatin, another fraction had defects in nucleotide excision repair.

The data also showed that successful synergy of targeting DNA damage response pathways will be more sophisticated than "pick any two pathways." Tumors with defects in the nucleotide excision repair pathway, for example, are hypersensitive to platinum-based chemotherapies but are resistant to PARP inhibitors.

Editor's note: Next week, the second part of this feature will focus on targeting DNA repair through chaperones or the cell cycle.