Evolutionary dynamics of cancer in response to targeted combination therapy

Ivana Božić, Johannes G. Reiter, Benjamin Allen, Tibor Antal, Krishnendu Chatterjee, Preya Shah, Yo Sup Moon, Amin Yaqubie, Nicole Kelly, Dung T. Le, Evan J. Lipson, Paul B. Chapman, Luis A. Díaz, Bert Vogelstein, Martin A. NowakeLife 2013 · DOI

TL;DR

This paper develops a mathematical model to predict how cancer evolves resistance to targeted therapies, particularly combination therapies. It reveals that even a single mutation conferring resistance to both drugs can doom combination therapy, and that simultaneous drug administration is far superior to sequential treatment for achieving long-term disease control.

What Problem Does This Paper Address?

Targeted cancer therapies have revolutionized treatment by precisely attacking cancer-driving mutations. However, a major hurdle is the rapid evolution of drug resistance, leading to tumor relapse. Combination therapy, using multiple drugs simultaneously, is a promising strategy to overcome this, but its effectiveness depends critically on the evolutionary dynamics of resistance. This paper aims to provide a quantitative framework to understand and predict the success of such combination therapies, considering the genetic diversity within tumors and the emergence of resistance mutations.

Methods

The authors employ a multitype branching process model to simulate tumor growth and evolution. This model accounts for cell division, death, and the acquisition of mutations that confer resistance to drugs. They specifically model two phases: pretreatment, where cells grow and mutate, and treatment, where sensitive cells are targeted. The model considers different types of resistance mutations: those conferring resistance to a single drug, and those conferring cross-resistance to multiple drugs. They also use Monte Carlo computer simulations to validate their analytical results, employing an enhanced version of the "Tool for Tumor Progression" (TTP) that handles continuous-time branching processes and multi-drug treatments.

Datasets

The study utilizes two main datasets:

  1. Melanoma patient data: Data from 20 melanoma patients treated with vemurafenib were used to estimate key model parameters, such as cell division and death rates, and to observe resistance dynamics in response to monotherapy.
  2. Metastatic cancer patient data: Data from 22 patients with pancreatic, colorectal, and melanoma metastatic disease were used to assess the model's predictions for combination therapy in a broader clinical context, quantifying lesion burden and number of metastases.

Key Insights and Ideas

  • The "single mutation" Achilles' heel: Even a single genetic alteration that confers resistance to both drugs in a combination regimen can severely limit the chances of long-term cure, especially in patients with a high tumor burden. This highlights the critical importance of developing drugs that target truly distinct pathways.
  • Simultaneous > Sequential: Administering drugs simultaneously is dramatically more effective than administering them one after another. Sequential therapy, even with drugs that have no common resistance mutations, is predicted to fail in most cases.
  • The power of distinct pathways: The model strongly suggests that the development of drugs targeting entirely different cellular pathways is paramount. If drugs share resistance mechanisms, even combination therapy struggles.
  • Tumor size matters: Larger tumors, with their greater cell populations, offer more opportunities for resistance mutations to arise and for resistant clones to establish themselves, making them harder to eradicate.

Key Results

  • Monotherapy failure: The model predicts, and patient data confirms, that monotherapy will eventually fail in almost all detectable lesions due to resistance.
  • Dual therapy success hinges on cross-resistance: For dual therapy, the probability of eradicating all lesions is high (>95%) only if there are no mutations that confer resistance to both drugs simultaneously (n₁₂ = 0). If even one such mutation exists (n₁₂ = 1), the probability of cure drops significantly, especially for patients with large tumor burdens. For example, in patients with a tumor burden of 10¹¹, dual therapy with n₁₂=1 has a failure probability of 0.74, while with n₁₂=0, it drops to 0.12.
  • Triple therapy is not a silver bullet: Similar to dual therapy, triple therapy is unlikely to succeed if there's a single mutation conferring resistance to all three agents.
  • Simultaneous vs. Sequential: Simulations show that simultaneous dual therapy can cure ~26% of lesions even when cross-resistance mutations are possible (n₁₂ = 1), whereas sequential therapy fails in all such cases. If there are no common resistance mutations (n₁₂ = 0), simultaneous therapy achieves >99% cure rates, while sequential therapy still fails in all cases.

Limitations and Future Work

  • Cancer Stem Cells: The authors acknowledge that their model's applicability could be improved by better understanding and quantifying the role of cancer stem cells, which may influence resistance dynamics.
  • Costly Resistance: While they explored the impact of "costly resistance" (where resistance mutations reduce cell growth rate), further work could refine these estimations.
  • Immune Therapies: The model is not directly applicable to immunotherapies, which involve complex interactions between the immune system and cancer cells, and whose mechanisms of resistance are not yet well understood.
  • Drug Development Guidance: The authors suggest their model could be used to develop in vitro assays to detect cross-resistance mutations, aiding drug development and clinical trial design. They also emphasize the need for drugs with very different chemical structures and mechanisms of action to circumvent resistance.

Summary updated

Abstract

In solid tumors, targeted treatments can lead to dramatic regressions, but responses are often short-lived because resistant cancer cells arise. The major strategy proposed for overcoming resistance is combination therapy. We present a mathematical model describing the evolutionary dynamics of lesions in response to treatment. We first studied 20 melanoma patients receiving vemurafenib. We then applied our model to an independent set of pancreatic, colorectal, and melanoma cancer patients with metastatic disease. We find that dual therapy results in long-term disease control for most patients, if there are no single mutations that cause cross-resistance to both drugs; in patients with large disease burden, triple therapy is needed. We also find that simultaneous therapy with two drugs is much more effective than sequential therapy. Our results provide realistic expectations for the efficacy of new drug combinations and inform the design of trials for new cancer therapeutics. DOI:http://dx.doi.org/10.7554/eLife.00747.001.
BibTeX
@inproceedings{bozic2013evolutionary,
  author = {Ivana Božić and Johannes G. Reiter and Benjamin Allen and Tibor Antal and Krishnendu Chatterjee and Preya Shah and Yo Sup Moon and Amin Yaqubie and Nicole Kelly and Dung T. Le and Evan J. Lipson and Paul B. Chapman and Luis A. Díaz and Bert Vogelstein and Martin A. Nowak},
  title = {{Evolutionary dynamics of cancer in response to targeted combination therapy}},
  booktitle = {eLife},
  year = {2013},
  doi = {10.7554/elife.00747},
}
Figure