Understanding Breast Cancer Research
Understanding breast cancer research can be hard for someone who isn’t a scientist. The language in research papers is very technical and complex. The point of the research may not seem clear at first glance. And you may be left wondering if the study was successful or not.
Our Breast Cancer Research News program provides summaries of the latest breast cancer studies that could make a difference in your treatment, follow-up care, or general health and lifestyle habits. Our goal in sharing breast cancer research is to help empower you with the information you need to get the best care possible.
Understanding the latest research on breast cancer can help you:
better communicate with your doctors
play a more active role in your treatment decisions
be more aware of the possible benefits and side effects of a treatment
make a more informed decision about being part of a clinical trial testing a new treatment
On this page, you can learn about the types of research studies we cover in Breast Cancer Research News, key terms you’ll see in breast cancer research, and important concepts that can help you read a research paper and understand what the findings mean for you.
Types of breast cancer research studies
There are many different types of research studies. The type of study done depends on the question the research is trying to answer. The following list summarizes the main types of studies done on breast cancer and its treatments, as well as the advantages and disadvantages of each.
Randomized controlled trials are considered the gold standard when studying new treatments. These types of studies provide the best answers when doctors want to know how effective a new treatment or test is.
In a randomized controlled trial, a new treatment can be compared to no treatment or to the treatment people usually receive (called “standard of care”). A new test can be compared to a different test or no test.
“Randomized” means the people in the study are divided randomly into different groups. One group may get the new, experimental medicine. Another group may get the current standard of care. A third group may get the current standard of care plus a placebo — a harmless substance that looks exactly like the treatment being studied and is given exactly the same way, but contains no medicine.
“Controlled” means there is a control group for comparison. In a study looking at a new cancer treatment, the control groups would receive the standard of care or the standard of care plus the placebo. The results of the control groups are compared to the results of the group getting the new treatment.
When planning a randomized controlled trial, the researchers need to figure out several things before the study starts:
how long the study should last
how many people need to be in the study
how the effect of the treatment or test will be measured
the characteristics of the people in the study
Many randomized controlled trials are double-blinded. This means neither the researchers nor the people in the study know who is in each group. This is done to help reduce bias and improve the accuracy of the results. For example, if the researchers know who got the study medicine, they might expect those people to have better outcomes, and that could influence their findings.
Advantages of randomized controlled trials:
They can measure of the effect of a new drug or test and offer results that are unbiased and consistent.
The statistical significance of the results is relatively easy to calculate. Statistical significance is a math calculation that shows the difference in results between groups in the study is actually because of the difference in treatments and not just due to chance.
Disadvantages of randomized controlled trials:
The new drug is tested under conditions that may be different from the real world. For example, a study may find that a new treatment works better than the current standard of care. Still, after the drug is approved, side effects may mean that most people don’t take the new drug as prescribed. This can mean the new drug won’t be as effective as it was in the study.
The goals or endpoints of most randomized controlled trials of cancer treatments are overall survival or progression-free survival. Overall survival is how long a person lives, whether or not the disease comes back. Progression-free survival is how long a person lives without the disease becoming worse or growing. Goals that may be just as important to people taking a new drug, such as quality of life and side effects, may not be part of the results.
Observational studies let researchers look at the effect that a risk factor, treatment, or test has on a group of people without trying to change who is or is not exposed to it. Cohort studies and case-control studies are two types of observational studies.
Cohort studies: A cohort is a group of people who are followed and observed for many years. For example, the Women’s Health Initiative (WHI), which has been following more than 161,600 postmenopausal women since they joined the study between 1993 and 1998, is an example of a cohort study. The WHI is looking for links between health, diet, and lifestyle factors and diseases such as cancer.
In a cohort study, researchers compare the health outcomes of two or more groups of people who have been exposed to different things. For example, one group may smoke and the other may not. Or one group may have taken a certain medicine and the other has not.
Cohort studies can have a forward-looking design, which researchers call prospective, or a backward-looking design, which researchers call retrospective.
In a prospective study, the result the researchers are interested in, such as cancer diagnoses, hasn’t happened when the study starts — all the people are healthy. So the researchers follow the healthy people and see how the differences between the groups affect the outcomes.
In a retrospective study, the result has already happened when the study starts. So the researchers look at people’s health history — usually through medical records or self-reporting — to find risk factors.
Advantages of cohort studies:
Cohort studies, especially prospective cohort studies, reduce the risk that the results will be affected by selection bias. Selection bias means that the sample of people in the study do not accurately represent the larger population. Because all the people are healthy when the study starts, it’s unlikely that they would be more or less likely to develop the disease being studied.
Cohort studies allow researchers to calculate the rate of disease in people exposed to something, so they can then calculate absolute risk and relative risk. Absolute risk is one person’s risk of developing a disease, such as breast cancer, over a certain period of time. Relative risk is the risk of a disease happening in one group of people compared to a different group of people. In most cases, the two groups of people have been exposed to different risk factors, such as smoking cigarettes vs. not smoking cigarettes, for example.
Disadvantages of cohort studies:
Large numbers of people may have to be studied for very long periods of time.
Cohort studies can be expensive and time-consuming.
Cohort studies aren’t very helpful for learning about rare diseases.
Cohort studies aren’t good for diseases that have a long latency period, meaning that a person contracts the disease, but no symptoms show up until years later.
Case-control studies: These types of observational studies compare people who have a medical condition, such as cancer, with people who don’t have the condition. But the two groups are otherwise as similar as possible, including age, ethnicity, and sex. The people are interviewed or their medical records are analyzed to look for things that may be risk factors for the medical condition.
Advantages of case-control studies:
They can be good for learning about rare diseases.
They are less expensive and less time-consuming than randomized controlled trials and cohort studies.
Disadvantages of case-control studies:
It can be difficult to figure out which people are the most similar to each other.
Most case-control studies are retrospective, which means the researchers ask about things people did or may have been exposed to in the past. If people don’t remember things correctly, it can affect the accuracy of the results.
Surveys are the most common type of cross-sectional study. A representative group of people are interviewed to find out their opinions or facts about their lives. Representative means the group is diverse enough in age, sex, race, or other factors so that the researchers can make conclusions about a bigger population of people based on the study. The strongest cross-sectional studies use randomly selected groups of people.
Advantages of cross-sectional studies:
Cross-sectional studies only collect information once, so they’re quick and inexpensive compared to other types of studies.
They can provide information on how common a particular health condition is.
Disadvantages of cross-sectional studies:
They don’t provide information on the cause of an illness or what the best treatment is.
The accuracy of cross-sectional studies depends on how truthful people are when they answer the survey questions.
Qualitative studies gather information on people’s opinions, motivations, or how they perceive their lives. For example, a study asking people receiving a specific cancer treatment about their quality of life is a qualitative study. Qualitative research doesn’t rely on numbers and data. Qualitative studies use diaries, questionnaires, and in-person interviews to collect information from people who have a particular medical condition and possibly their loved ones.
Advantages of qualitative research:
The results can offer a detailed picture of people’s experiences and help researchers understand why people act the way they do.
Disadvantages of qualitative research:
Because qualitative research is generally more time-consuming, the studies usually involve fewer people.
The quality of the information collected depends on the skills and observations of the researcher.
A meta-analysis is a study that combines and analyzes the results of a number of earlier studies. Researchers do meta-analyses to increase the statistical power of the results and to resolve controversy when the results of individual studies disagree.
Advantages of a meta-analysis:
Meta-analyses can offer a tidy review of a large, complex, and sometimes conflicting body of research.
Disadvantages of a meta-analysis:
The results of a meta-analysis are only as good as the studies included in it.
The researchers must do a thorough and detailed search for existing research. If some relevant studies aren’t included, the results of the meta-analysis are weaker.
Phases of clinical trials
Clinical trials looking at new breast cancer treatments and tests are conducted in a series of four steps, or phases, from phase I to phase IV. Each phase builds on the one before it:
Phase I trials look at the best way to give a new treatment, as well as the safest dose.
Phase II trials look at how effective the new treatment is.
Phase III trials compare the safety and effectiveness of a new treatment to the current standard of care.
Phase IV trials look at whether the treatment has benefits or long-term side effects that were not studied or seen in the phase II and phase III trials.
To learn more about each clinical trial phase, read How Clinical Trials Are Conducted.
How to read breast cancer research journal articles
At first glance, reading an article in a research journal such as the Journal of Clinical Oncology or JAMA (formerly known as the *Journal of the American Medical Association*) can be a little intimidating. Even the titles can be filled with unfamiliar, very long words. The body of the paper is likely to have many abbreviations for words that aren’t usually abbreviated.
It also can be frustrating when you can’t read the full text of a paper unless you pay for it. You might be able to access full journal articles through your local library. Most libraries have online subscriptions to many scientific journals.
Still, getting to the library and spending a couple of hours scanning journal articles can be difficult for some people. But you can get a basic understanding of a paper and decide whether it applies to your unique situation by reading the abstract, which is the first section of a scientific paper.
The abstract gives you a summary of the entire paper.
Here are the most common sections in an abstract and what they will tell you. It’s important to know that different journals may have different names for the sections of a paper covered in the abstract. In JAMA, specific sections are called “objective” and “interventions.” In the Journal of Clinical Oncology, similar sections are called “purpose” and “patients and methods.”
The purpose or objective of the paper tells you why the study is being done and may include background information on why the study is important.
In most cases, the objective of the study will tell you whether the research applies to you because the researchers will say which type of breast cancer — early-stage, hormone-receptor-positive disease, for example — is included in the study.
Some journals split this section, so you may see methods, participants, and design as individual sections. This section (or sections) tells you how many people were in the study and gives you information about the type of breast cancer being studied.
For example, “This trial enrolled 476 women with HER2-positive early or locally advanced breast cancer,” lets you know the study applies to women diagnosed with early-stage HER2-positive disease or HER2-positive breast cancer that has spread to tissue near the breast, but not to parts of the body away from the breast.
Methods/design will tell you the type of study that was done and any treatments the researchers used in the study.
The results section tells you the results of the study. While this section seems like it would be the easiest to understand, the results also may include quite a bit of statistical and mathematical information, including terms such as “mean,” “median,” “hazard ratio,” “standard deviation,” “confidence interval,” and “P value.” Those terms are explained in detail under “Reading the full paper” below. In most cases, it’s easier to skim the results section and move on to the final section of the abstract to learn what happened in the study.
The conclusion or relevance section summarizes the results of the study in a few sentences, leaving out all the math and statistics. This section also may explain how the results support, build on, or conflict with earlier research.
Some scientific journals may offer free access to full papers. The full paper has many more sections than the abstract. In many cases, it may make sense to read the sections in a different order than they are presented in the paper.
For example, you may want to read the “Conclusion” section first to see if the research applies to your situation. If it doesn’t, then you may not want to read the paper.
Here are the most common sections in a research journal article and what they will tell you.
As explained above, the abstract is the first section of research papers and gives you a summary of the entire paper.
The introduction of a research paper gives some history of the problem the research is trying to solve and explains how the paper fits in with other research that has been done. The last part of the introduction usually explains why the researchers wanted to do the study.
The methods section of a research paper will likely have a number of subsections that explain exactly how the research was done:
Study design tells you how people were assigned to treatment groups, the types of treatments that were given to each group, and the type of breast cancer being looked at in the study. This section also will tell you which institutional review boards and ethics committees approved the study and that the people in the study gave written consent to be in the study.
Study population tells you exactly who was in the study, including:
the age of the people
the type of cancer they have
any previous treatments they received
any other specific characteristics that are important to the study, such as brain metastases or time before treatment started
Note: In some journals, details about the study population may be in the results section of the paper.
Data collection tells you what information the researchers collected about each person in the study and when they collected it. For example, if the researchers wanted to see if a new treatment was effective at stopping metastatic breast cancer lesions in bones from growing, they might have imaging done on all the people in the study before the new treatment starts, again 3 months after the new treatment started, and again 6 months and 1 year after the treatment started.
Statistical analysis is likely to be the most complex section and probably the one that is most difficult for non-scientists to understand. This is where the researchers explain the statistical methods they used. You may see terms such as “Kaplan-Meier method” and “Cox proportional hazards regression model.” The Kaplan-Meier method is a statistical tool often used to measure the percentage of people living for a certain amount of time after a treatment. A Cox proportional model is a statistical tool that links the survival time of a person to another factor. In breast cancer research, this other factor is usually a treatment. Cox models also allow researchers to adjust for other factors that may affect survival. So if researchers wanted to know how a new treatment improved survival, but wanted to take out the effects of another factor — say, whether the people in the study smoked or not — they would use a Cox model. If the statistical analysis section of the paper is hard to understand, don’t worry. The important information is explained in the results section of the paper.
The results section of a research paper is where the most important information is, including:
how many people were in the study
the treatments being studied
which treatments the people received
when the people received the treatments
when information on the effectiveness of the treatments was collected
the outcomes or endpoints the researchers used to figure out how effective the treatments were, such as progression-free survival or overall survival
how effective the treatments were
side effects the treatments caused
In many cases, the results section will include a written summary of the results and more detailed information will be presented in tables and figures.
For example, the text may say that “the disease characteristics were well balanced between treatment arms (Table 1).” When you look at Table 1 you see:
the number of people in each treatment group
the median age of the people in each treatment group
specific cancer characteristics in each treatment group, such as previous treatments, stage at diagnosis, and whether the cancer is hormone-receptor-positive or not
The results section will likely include many terms that are commonly used in research, but may not be familiar to you. Here are definitions of some of key terms you’re likely to see:
Median: First, it’s important to know that median does *not* mean average. The median is the middle of a list of numbers. So if the median age of the people in the study is 62, that means that half the people were older than 62 and half the people were younger.
Mean: Mean is the average of a group of numbers. So if the mean age of the people in the study is 55, it means that the researchers added up the age of everyone in the study and then divided that total by the number of people in the study.
Disease-free survival: How long a person lives without the cancer coming back.
Progression-free survival: How long a person lives without the cancer growing.
Overall survival: How long a person lives whether the cancer comes back or grows.
Hazard ratio, often abbreviated HR: How often a particular thing happens in one group of people compared to how often it happens in another group of people, over a certain period of time. Hazard ratios are commonly used to measure survival at a point in time in a group of people who have been given the treatment being studied compared to another group of people, called the control group, who are given a different treatment (usually a placebo). A placebo is a harmless substance that looks exactly like the treatment being studied and is given exactly the same way, but contains no medicine.
A hazard ratio of 1 means there is no difference in survival between the two groups.
A hazard ratio greater than 1 means survival was worse in the people given the treatment being studied compared to the control group.
A hazard ratio less than 1 means survival was better in the people given the treatment being studied compared to people in the control group.
Confidence interval, often abbreviated CI: How likely a statistical result would happen if the same study and calculations were done on a different sample of people with the same disease. In other words, if the confidence interval is 95%, it means that if the study were done 100 times, the results would be the same 95 times.
P value: The probability value, abbreviated as P value, is a number that shows how statistically significant the results are. If a result is statistically significant, it means that it was likely the result of the differences in treatment and not just because of chance. If a result is not statistically significant, it means that it was likely due to chance and not because of the difference in treatment. Most researchers consider a P value of less than 0.05 as being statistically significant, which means that the result has less than a 1 in 500 chance of being wrong.
When breast cancer research results are reported, most results are survival statistics. Nearly all studies will report the percentage of people surviving in each treatment group and then also give hazard ratio, confidence interval, and P value information.
For example, in a study on adding Xeloda (chemical name: capecitabine) to standard chemotherapy after surgery for early-stage triple-negative breast cancer, the results showed that 5-year disease-free survival rates were 86.3% for women treated with Xeloda and chemotherapy compared to 80.4% for women treated with chemotherapy alone. The hazard ratio was 0.66, the confidence interval was 0.99, and the P value was 0.044.
The hazard ratio means that disease-free survival was 34% better in the group treated with Xeloda and chemotherapy compared to the group treated with chemotherapy alone.
The confidence interval means that the results are 99% certain to be accurate.
The *P* value means that the results are statistically significant. A P value of 0.044 means that the difference between the groups was due to chance only 44 times out of 1,000. So, there is less than a 1 in 22 chance of the researchers’ conclusion being wrong.
The discussion section of a research paper is where the researchers summarize the results of the research. They also may talk about the history of research on the topic and why the study is important.
The researchers also may talk about any strengths or limitations of the study. If the strengths of the study outweigh the limitations, it’s likely that the results are accurate.
The conclusion is often the shortest section of the paper. The researchers summarize the results in one or two sentences and say why they’re important.
Here, the authors of the paper disclose if they have received money from any companies or organizations that may be interested in the results of the research and detail any conflicts of interest they may have. For example, if one of the authors has received money from the company that is making the medicine being studied, that could be a conflict of interest.
The affiliations detail each author’s title and home institution.
Considering all aspects of research results
Let’s say you’ve read the article on adding Xeloda to standard chemotherapy after surgery for early-stage triple-negative breast cancer. The results seem very promising, but there are still a few questions to consider as you make sense of the results.
It’s important to know some things about the journal that has published the paper:
Is the journal peer-reviewed? This means that the paper is reviewed before publication by scientists who are experts in the paper’s subject matter. These reviewers look to see if the methods are sound, the data are factual, and the statistical analysis is accurate. In many cases, a paper may go through several rounds of peer review before it is published. Peer-reviewed journals are considered more scientifically valid than other publications. You can figure out if a journal is peer-reviewed by going to its website and looking at the “about” information. If that’s not clear, you also can look at the information for authors, which will explain the peer-review process, if one exists.
What is the journal’s impact factor? The impact factor is a measure of how often the articles in a journal are cited by other papers. The idea is that important papers are cited more often, which influences the research on a topic. So the higher the impact factor, the more prestigious a journal. An impact factor of 6 or higher is considered a good score.
What is the journal’s reputation? A journal published by a reputable professional association or society, such as the American Medical Association or the American Society of Clinical Oncology, is considered of higher quality than many other publications. Also, the reputation of the journal’s editor or editorial board can be a good indicator of the journal’s quality. You can usually find this information on the journal’s website in the “about” information.
While the results may seem promising, if they’re not statistically significant, it means they could have happened by chance and not because of a difference in treatment.
The number of people in a study, called the sample size, can tell you how applicable the results may be in the real world. For example, a study that includes 600 women diagnosed with early-stage triple-negative breast cancer would have stronger results than a study including only 20 women with the disease.
It’s also important to know how the people in the study were chosen. In a perfect world, the people in a study would be a randomly chosen representative sample of people with a specific illness. So people in a study on a new treatment for early-stage triple-negative breast cancer would represent, on average, people who are diagnosed with that particular type of breast cancer. The sample of people in the study would be of the same age ranges, ethnicities, and insurance status as people in the real world.
Selection bias means that the sample of people in the study do not accurately represent the larger group. So if the study were on early-stage triple-negative breast cancer and every person in the study had stage I disease, that would be selection bias because early-stage disease includes stage I to stage III. And since stage I breast cancer is more likely to have better outcomes than stage III breast cancer, the results of the study could be biased.
Randomized controlled trials are considered the gold standard when studying new treatments. If people are not randomly assigned to either the new treatment or the placebo, the results could be biased if more people with less aggressive disease are assigned to get the new treatment.
Studies on certain topics, such as people’s diets and how what they eat could be linked to cancer risk, have inherent bias. In these studies, people are asked to recall what they ate, sometimes more than 2 or 3 years ago. It’s really unlikely that people can accurately recall how much milk they drank in elementary school or how many servings of vegetables they had when they were in high school.
So the results of any studies that ask people to recall what they ate or what they did in the past should always be considered biased.
News coverage of research results and tips for wading through the hype
You may see news coverage of research results that make it sound like a vaccine for breast cancer or a cure for breast cancer is right around the corner. Sadly, neither is true.
What is true is that scientists are studying cancer vaccines using cells in petri dishes, and some of these extremely early results are promising. But it can take 15 or 20 years for a petri dish study to move up to a study in people, if at all.
So how does this distortion between the results in a scientific paper and the story you see in the media happen?
In most cases, the distortion happens in part because the aim of the news media is to sell papers or magazines or get eyeballs on their websites and clicks on stories. A sensational headline usually attracts more interest. So a headline like, “Scientists close in on cure for breast cancer,” is guaranteed to get more readers than a headline that reads, “Mouse cells seem to respond to cancer vaccine.”
Another factor is the speed of the news cycle. We live in a 24-hour society, and news websites must constantly develop new content. So journalists often don’t have time to read the entire scientific paper. They look at the headline and rely on a media release from the research institution to fill in the details. And institutions know that the punchier the media release, the more likely the story is to get picked up. It becomes a vicious cycle of overhyping research results to get media attention.
So how can you make sure that the breast cancer research you’re reading about in the media is accurate? Here are some tips:
Read more than the headline. See if the story actually explains the research.
Search the topic of the research for other news stories. If they all have the same wording, it’s likely the stories are based on a single media release and not much work has been done to verify its accuracy.
Find the original paper and read the abstract yourself. Do the original paper’s conclusions match the media stories? If not, it’s likely the story has been overhyped.
Check our Breast Cancer Research News section for our coverage of the research. Our aim is to give you accurate, easy-to-understand summaries of the latest research that can make a difference for you and your unique situation. If we haven’t summarized a study, it’s probably because the results are too early to be put into practice.
— Last updated on February 10, 2022, 3:55 AM
This content made possible in part through generous support from Lilly Oncology.