Monthly Archives: June 2015

Where have all the promising treatments gone? The challenges of clinical trials.

Identifying the right patients to recruit for clinical trials of a new drug is a key component of a trial's success.

Identifying the right patients to recruit for clinical trials of a new drug is a key component of a trial’s success.

Many of the scientific breakthroughs we read about in the media discuss discoveries in the lab that could be translated to promising therapies, or exciting new drugs that have just entered clinical trials. Months or years pass with no further mention of this amazing treatment. What happened?

It’s tempting to hope that a newly discovered, promising drug could be used by patients within the next year. However before a drug can be used in the clinic, researchers need to evaluate its performance in small cohorts of volunteers and patients. I will briefly discuss the three phases of clinical trials below; you can read more about them here.

Why are cancer treatments so expensive?

In 2009, the estimated average cost of a clinical trial in the US was US$1.3-1.7 billion, and those costs continue to rise. With so many new drugs being developed, and many new advances in cancer treatments, it’s getting harder to prove that a new drug is significantly better than the existing options. In order to detect these incremental improvements, trials are recruiting more patients, for longer periods of time, and are using more advanced (and expensive) monitoring techniques, such as MRI or genome sequencing. All of this contributes directly to high drug costs once a treatment hits the market.

Why do clinical trials fail?

Considering the time and cost that goes into establishing a clinical trial, it is surprising to learn that over 62% of Phase III trials in the US fail – meaning that the investigated drug is not better than the current standard of care. There are many reasons for this, including a trial that may not have been properly designed. For instances, a drug might be effective in a small percentage of patients with specific characteristics, however if a clinical trial is conducted in a broader population, the effectiveness of the drug on the small percentage of “responders” will be lost within the larger population, resulting in failure of the trial.

What does the future of clinical trials look like?

In an effort to design clinical trials that have higher rates of success (and are therefore more beneficial for patients), researchers are moving away from large scale trials with diverse patient populations, and more towards smaller trials of targeted patient groups. For example, by categorizing cancer patients by the genomic profile of their tumours, and recruiting only a defined subset of patients that are likely to benefit from treatment, current clinical trials are starting to improve their success rates.

The high costs and high failure rates of clinical trials are certainly cause for concern, however it’s important to remember that drugs do succeed, and many have dramatically changed the face of cancer treatment. For example, the drug Herceptin is a monoclonal antibody that targets a specific protein called HER2, which is thought to drive tumour growth. This oncogenic protein is over-expressed in the tumours of 25-30% of breast cancer patients. When the Herceptin antibody was evaluated in clinical trials, it was tested on patients whose tumours over-expressed HER2. Because of this effective design of the clinical trial, researchers were able to capture the positive effects of this drug that would have otherwise been lost had it been tested on patients with and without HER2 over-expression. Since its development in 1990 and approval by the FDA in 1998, Herceptin has had a tremendous impact on the treatment of patients with HER2+ breast cancer.

New cancer treatments are being approved every year, some of which have met with great success. The hope is that as more drugs make their way successfully through clinical trials, patient outcomes will continue to improve.

This article was written by Ashley Hickman. Ashley is in the second year of a Masters program at the University of Toronto where she studies how to regulate a very important cancer causing gene called myc. To learn more about Ashley and her research check out her bio on our members page.


Amiri-Kordestani, L. & Fojo, T. Why do phase III clinical trials in oncology fail so often? J. Natl. Cancer Inst. 104, 568–569 (2012).

Collier, R. Rapidly rising clinical trial costs worry researchers. CMAJ 180, 277–278 (2009).

Phases of Clinical Trials.

Webcast: Latest Advances in Cancer Research

Watch this great webcast from several RIOT members who outline some of the most exciting advances in cancer research today. Natalia Ruiz will explain what personalized medicine is and how it is growing. Mathew Hall will talk about immunotherapy and its effectiveness against melanoma. Sue Li will finish things up by talking about how old drugs are being re-purposed to fight against cancer.

If you want more information all of the presenters addressed some questions that came up during the webcast. You can read the questions and answers in this document: Latest Advances in Cancer Research Questions and Answers.

Personalized Medicine: A New Approach to Target the “Silent Killer”

Personalized medicine blog picturePersonalized Medicine, or Precision Medicine, is an emerging strategy to treat all forms of cancer. But what exactly is personalized medicine? It essentially means using the right drug for the right patient. This doesn’t mean that personalized medicine CREATES drugs for patients, rather, it implies tailoring medical treatment to the individual characteristics of each patient’s tumour. Cancer is complex, it is actually over 200 different diseases, and research uncovered that ovarian cancer itself is not one single disease. Thus, we should not use one treatment for all patients.

Personalized medicine is now being used in an attempt to treat ovarian cancer, also known as the “silent killer”, since more than 80% of women present with late stage disease and thus have a poor survival outcome. One of the challenges in treating ovarian cancer is that the biology of these tumours is quite complex. Every ovarian patient’s tumour is different, due to the various underlying DNA changes (mutations) in their particular tumour. These mutations in turn determine how patients respond to treatment. So is it possible that personalized medicine can be used to improve treatment outcomes for this devastating disease by giving the right drug to the right patient? There are currently two emerging approaches that are being tested to try to answer this question:

  1. Reading” the genetic make-up of a tumour to identify drugs that target mutations found within the tumour

For decades, treatment decisions for cancer have been based on a patient’s tumour histology. In other words, pathologists analyze a small piece of tumour under the microscope, and categorize the patient’s tumour based on, for example, how the cancer cells look compared to normal cells. Ovarian cancer has multiple histological subtypes, each with differing responses to standard therapy. Through research, we now know that these subtypes are the result of different underlying genetic changes (mutations). Clinical research is now underway to determine the genetic make-up of a patient’s tumour and identify changes within it (1). By reading the tumour’s DNA (known as genetic profiling), there is a potential to identify drugs that selectively target the mutations found within the tumour. The DNA is taken from the tumour biopsy or, in the future, from a blood sample (2). Knowing a patients’ cancer alterations is highly valuable to cancer researchers and oncologists. In some cancers, where the mutations are well characterized, genetic profiling has helped improve outcomes because of more informed treatments selected by the oncologists. For example, ovarian cancer patients that have the hereditary form of the disease and carry a change in their BRCA genes usually respond much better to chemotherapy. Researchers are also investigating another drug, called olaparib (AZD-2281), in this same group of patients (ASCO, Abstract 5505; New England Journal of Medicine, PMID 22452356) (3). As more targeted drugs are approved for ovarian cancer, other genetic changes associated with this disease may become important targets for ovarian cancer treatment.

  1. Mouse avatars to determine treatment options

Oncologists are looking at new ways in which they can test multiple potential treatment strategies on a patient’s tumour to find the most appropriate one for that individual patient. For example, drugs have been tested on patients’ cells extracted from tumours, by placing them in a Petri dish and growing them in a laboratory. However, most cancer cells do not survive these conditions. A more successful, and widely used approach involves growing pieces of a patient’s tumour in mice (4). Some companies have even begun offering this method to patients in the United States (5). In this approach, the tumour piece is surgically removed from the patient and implanted under the skin of multiple mice to allow it to grow. Commonly used treatments as well as potential new drugs are given them to the mice with tumours and then compared. If the tumour shrinks when mice are treated with a certain drug, the idea is that the drug can potentially be effective in that patient. This approach has already been used for ovarian cancer patients (6). Although promising, this method has not been approved in humans using randomized clinical trials.

Challenges of personalized medicine:

Despite these successes, there are emerging challenges with personalized medicine. For example, in both examples I presented, the approach is quite costly. Sequencing a patient’s tumour and interpreting the results or, using multiple mice for a single patient to test various drugs can cost thousands of dollars. In addition, cancer drugs will not target every cancer change (mutation) and it is possible that the cancer cells will become resistant to these drugs over time.

Personalized medicine holds a lot of promise and researchers are continuing to find new ways, such as nanotechnology (7), to predict the best treatment option for patients. In the case of ovarian cancer, these newly emerging tests could potentially predict who will respond to chemotherapy or targeted therapy without sequencing the tumour DNA or using animal models. More research is needed to have personalized medicine become a standard approach to treating cancers such as ovarian cancer. I believe the future of personalized cancer medicine is bright and will lead to new and effective treatment options for patients with ovarian cancer.

This article was written by Dan Cojocari. Dan is a PhD candidate in the Department of Medical Biophysics at the University Toronto. He studies how pancreatic cancer cells are able to survive in harsh tumor environments. To learn more about Dan and his research check out his bio on our members page.

References and further readings:

  1. Cancer Genomics Program Clinical Trial Studies.
  2. An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage
  3. Ovarian Cancer Research
  4. The Littlest Patient: Cutting-edge mouse models fuel hope for understanding and treating cancer. Jennifer Couzin-Frankel. Science Magazine. 3 October, 2014.
  5. Hope in a mouse: Selling personalized mouse models to cancer patients, a firm draws thanks and reproach. Jennifer Couzin-Frankel. Science Magazine. 3 October, 2014.
  6. Tumorgrafts as In Vivo Surrogates for Women with Ovarian Cancer. Clinical Cancer Research. March 1, 2014.
  7. How to identify drugs that work best for each patient. MIT News. 2015.
  8. Princess Margaret Cancer Centre Personalized Medicine.