CAMBRIDGE, Mass.—For nearly three decades, chemotherapy drug cisplatin has been a key component in doctors' arsenals as a first line of defense against tumors, especially those of the lung, ovary and testes.
While cisplatin is often effective when first given, it has a major drawback: Tumors can become resistant to the drug and start growing again.
A team of MIT cancer biologists have shown how that resistance arises, a finding that could help researchers design new drugs that overcome cisplatin resistance. The team, led by Tyler Jacks, director of the David H. Koch Institute for Integrative Cancer Research at MIT, reported the results in the April 15 issue of the journal Genes & Development.
Cisplatin and other platinum-based cancer drugs destroy tumor cells by binding to DNA strands, interfering with DNA replication. That activates the cell's DNA repair mechanisms, but if the damage is too extensive to be repaired, the cell undergoes programmed suicide.
Eventually, cancer cells learn to fight back. The new study shows that tumor cells treated with cisplatin ramp up their DNA repair pathways, allowing them to evade cell death, says Trudy Oliver, a postdoctoral fellow in Jacks' lab and lead author of the paper.
According to Oliver, the significance of the discovery is that it is one of the first autochthonous mouse tumor models of acquired cisplatin resistance.
"Cisplatin is a standard chemotherapy drug used to treat many types of human cancer, including lung, ovarian, head and neck, testicular and bladder cancer," she says. "It functions by binding DNA and forming lesions, or adducts, that inhibit the ability of the cell to replicate."
A major limitation to its successful use is that tumors commonly acquire resistance to cisplatin after repeated doses.
Oliver points out that many previous studies investigating the mechanisms of cisplatin resistance have used cell lines or xenograft models, which cannot recapitulate the complexity of the in vivo context.
"We have used endogenous lung tumors that are initiated by the expression of an oncogene and the loss of a tumor suppressor gene—Kras and p53, respectively—that occur in human cancer," she notes. "Like human cancer, the mice have lung tumors that acquire resistance to cisplatin after repeated dosing. Using this in vivo model, we show that resistant tumors acquire the ability to rapidly repair the toxic lesion, and we identify genes involved in DNA damage repair that are more highly expressed in these tumors."
Previous studies had suggested several possible mechanisms for resistance development, including enhancement of DNA repair pathways, detoxification of the drug, and changes in how the drug is imported into or exported out of the cell. However, those studies were done in cancer cells grown in the lab, not in living animals.
"Many mechanisms have been identified but it's not clear what happens in vivo because the in vivo environment is so much more complicated than in cell lines," says Oliver.
"We also show that the p53-p21 tumor suppressor pathway is not required for a significant therapeutic benefit of cisplatin," Oliver points out. "The role of p53 in human cancer has been controversial—some studies suggesting its loss contributes to resistance, others to its sensitivity. In this context, we find that loss of p53 activity contributes to poor prognosis, but does not prevent therapeutic benefit. "
Finally, the researchers show that chronic cisplatin treatment can actually promote tumor progression.
"This demonstrates that detrimental effects can occur after repeated treatment with agents whose mechanism of action is DNA damage," Oliver says. " It highlights the need for additional therapies whose function is not based on DNA damage."
To do this, Oliver and her colleagues set out to study cisplatin resistance in mice with a mutation in a gene called Kras, which leads the animals to develop lung cancer. About 30 percent of human lung cancer patients have mutations in Kras. Some of the mice also had defective versions of the tumor suppressor gene p53, which is mutated in about half of human lung cancers.
The researchers found that cisplatin was effective against lung tumors in both sets of mice, though it was more potent in mice that still had functional p53. In those mice, tumors actually shrank, while the drug only slowed tumor growth in mice with defective p53. Those results are consistent with findings in human patients.
After four doses of cisplatin, mice with normal p53 developed resistance to the drug, and tumors started growing faster. To figure out why, the researchers analyzed which genes were being transcribed more as resistance developed, and identified several that are involved in DNA repair pathways.
One gene that particularly caught the researchers' attention is PIDD (p53-induced protein with a death domain), which is turned on by p53 and has been implicated in programmed cell death, though its exact function is not known. When PIDD levels are artificially increased in human lung cancer cells, they become more resistant to cisplatin.
"The genes we identified that are altered in cisplatin-resistant tumors have been shown to play a role in multiple DNA repair pathways," Oliver says. "It is not yet clear which repair pathways are the most important, but given that some of these pathways are already targetable by current drugs, such as Parp inhibitors and Chk kinase inhibitors, this may provide rationale for testing combination therapies with cisplatin and these agents, which may improve efficacy and/or inhibit resistance."
Oliver says researchers are pursuing the mechanism of action of particular genes involved in cisplatin–resistance in these tumors, such as PIDD, which we demonstrated can promote cisplatin-resistance in human lung tumor cell lines.
Oliver is now studying tumors in which the PIDD gene has been knocked out, to see if its absence hinders drug resistance.
"It is likely that PIDD is just one of many genes, in many pathways, involved in the drug resistance process," says Oliver. "It's not a simple phenomenon."
Moving forward, Oliver said the team would like to be able to target particular genes and pathways in tumors in vivo and show that this can be therapeutically beneficial in combination with cisplatin.
"The characterization of this model also makes it possible to identify additional signaling pathways important in cisplatin-resistance, because we identified heterogeneity in tumor response, suggesting that tumors can find multiple ways to avoid cisplatin response," she concludes.
While cisplatin is often effective when first given, it has a major drawback: Tumors can become resistant to the drug and start growing again.
A team of MIT cancer biologists have shown how that resistance arises, a finding that could help researchers design new drugs that overcome cisplatin resistance. The team, led by Tyler Jacks, director of the David H. Koch Institute for Integrative Cancer Research at MIT, reported the results in the April 15 issue of the journal Genes & Development.
Cisplatin and other platinum-based cancer drugs destroy tumor cells by binding to DNA strands, interfering with DNA replication. That activates the cell's DNA repair mechanisms, but if the damage is too extensive to be repaired, the cell undergoes programmed suicide.
Eventually, cancer cells learn to fight back. The new study shows that tumor cells treated with cisplatin ramp up their DNA repair pathways, allowing them to evade cell death, says Trudy Oliver, a postdoctoral fellow in Jacks' lab and lead author of the paper.
According to Oliver, the significance of the discovery is that it is one of the first autochthonous mouse tumor models of acquired cisplatin resistance.
"Cisplatin is a standard chemotherapy drug used to treat many types of human cancer, including lung, ovarian, head and neck, testicular and bladder cancer," she says. "It functions by binding DNA and forming lesions, or adducts, that inhibit the ability of the cell to replicate."
A major limitation to its successful use is that tumors commonly acquire resistance to cisplatin after repeated doses.
Oliver points out that many previous studies investigating the mechanisms of cisplatin resistance have used cell lines or xenograft models, which cannot recapitulate the complexity of the in vivo context.
"We have used endogenous lung tumors that are initiated by the expression of an oncogene and the loss of a tumor suppressor gene—Kras and p53, respectively—that occur in human cancer," she notes. "Like human cancer, the mice have lung tumors that acquire resistance to cisplatin after repeated dosing. Using this in vivo model, we show that resistant tumors acquire the ability to rapidly repair the toxic lesion, and we identify genes involved in DNA damage repair that are more highly expressed in these tumors."
Previous studies had suggested several possible mechanisms for resistance development, including enhancement of DNA repair pathways, detoxification of the drug, and changes in how the drug is imported into or exported out of the cell. However, those studies were done in cancer cells grown in the lab, not in living animals.
"Many mechanisms have been identified but it's not clear what happens in vivo because the in vivo environment is so much more complicated than in cell lines," says Oliver.
"We also show that the p53-p21 tumor suppressor pathway is not required for a significant therapeutic benefit of cisplatin," Oliver points out. "The role of p53 in human cancer has been controversial—some studies suggesting its loss contributes to resistance, others to its sensitivity. In this context, we find that loss of p53 activity contributes to poor prognosis, but does not prevent therapeutic benefit. "
Finally, the researchers show that chronic cisplatin treatment can actually promote tumor progression.
"This demonstrates that detrimental effects can occur after repeated treatment with agents whose mechanism of action is DNA damage," Oliver says. " It highlights the need for additional therapies whose function is not based on DNA damage."
To do this, Oliver and her colleagues set out to study cisplatin resistance in mice with a mutation in a gene called Kras, which leads the animals to develop lung cancer. About 30 percent of human lung cancer patients have mutations in Kras. Some of the mice also had defective versions of the tumor suppressor gene p53, which is mutated in about half of human lung cancers.
The researchers found that cisplatin was effective against lung tumors in both sets of mice, though it was more potent in mice that still had functional p53. In those mice, tumors actually shrank, while the drug only slowed tumor growth in mice with defective p53. Those results are consistent with findings in human patients.
After four doses of cisplatin, mice with normal p53 developed resistance to the drug, and tumors started growing faster. To figure out why, the researchers analyzed which genes were being transcribed more as resistance developed, and identified several that are involved in DNA repair pathways.
One gene that particularly caught the researchers' attention is PIDD (p53-induced protein with a death domain), which is turned on by p53 and has been implicated in programmed cell death, though its exact function is not known. When PIDD levels are artificially increased in human lung cancer cells, they become more resistant to cisplatin.
"The genes we identified that are altered in cisplatin-resistant tumors have been shown to play a role in multiple DNA repair pathways," Oliver says. "It is not yet clear which repair pathways are the most important, but given that some of these pathways are already targetable by current drugs, such as Parp inhibitors and Chk kinase inhibitors, this may provide rationale for testing combination therapies with cisplatin and these agents, which may improve efficacy and/or inhibit resistance."
Oliver says researchers are pursuing the mechanism of action of particular genes involved in cisplatin–resistance in these tumors, such as PIDD, which we demonstrated can promote cisplatin-resistance in human lung tumor cell lines.
Oliver is now studying tumors in which the PIDD gene has been knocked out, to see if its absence hinders drug resistance.
"It is likely that PIDD is just one of many genes, in many pathways, involved in the drug resistance process," says Oliver. "It's not a simple phenomenon."
Moving forward, Oliver said the team would like to be able to target particular genes and pathways in tumors in vivo and show that this can be therapeutically beneficial in combination with cisplatin.
"The characterization of this model also makes it possible to identify additional signaling pathways important in cisplatin-resistance, because we identified heterogeneity in tumor response, suggesting that tumors can find multiple ways to avoid cisplatin response," she concludes.