Novel App Informs Patients With Cancer of Unknown Primary

CURE, 2021 Rare Cancers Special Issue, Volume 9,

An online tool developed by oncologists for this rare diagnosis individualizes survival estimates, resulting in candid discussions and better decision-making.

A diagnosis of cancer of unknown primary — an umbrella term used to classify a diverse group of metastatic cancers based on the absence of an identifiable primary tumor — can pose a challenge for both patients and oncologists. However, a novel prognostic model to predict overall survival may provide more insight and even spark a conversation between patients and their oncologists.

Although the exact number of cases of cancer of unknown primary diagnosed annually is unknown, the American Cancer Society estimates that 32,880 cases of cancer of unknown primary will be diagnosed this year in the United States, which represents about 2% of all cancers diagnosed.

According to Dr. Kanwal P.S. Raghav, an associate professor of gastrointestinal medical oncology at The University of Texas MD Anderson Cancer Center in Houston, it is extremely challenging to treat cancer of unknown primary because most cancer research is done based on primary location of tumors — even when it comes to predicting overall survival.

He and his colleagues sought to change that. They conducted a study that enrolled nearly 900 patients across MD Anderson, Tennessee Oncology and the University of British Columbia to test a web tool, or nomogram, that individualizes overall survival predictions based on baseline characteristics like biological sex, performance status and histology (tumor description). These characteristics are available to every patient at diagnosis and are important because they can reflect the severity of disease in cancer of unknown primary.

In an interview with CURE ®, Raghav explained the trial results and how this app could positively affect the experiences of patients with cancer of unknown primary as well as increase communication with their health care providers.

Q: What led to the creation of this tool?

A: When somebody gets a diagnosis of metastatic cancer or cancer that has spread — such as cancer of unknown primary — one of the first questions in everyone’s mind is: How long do I have to live? We don’t really have a good answer to that.

The only way of estimating survival is with some older models that tell you whether you are in a good risk category or in a poor risk category, but they don’t give you your prediction individually. All they can tell you is that if you are good risk, you would live a certain amount of time, and if you are poor risk, you would live a certain amount of time — but that’s an average approximation. Therefore, we wanted to create a model that could provide an individualized survival estimate for patients with cancer of unknown primary.

The second problem with those models is that they’re old and based on patients who probably didn’t have access to some of the modern-day treatments that are out there now. So we designed the study to look at this individualized prediction using a more recent cancer of unknown primary group, and it is also one of the largest studies for patients with cancer of unknown primary, considering how rare this tumor is.

How exactly can this tool help patients when making treatment decisions?

The estimate of one’s survival probability is important in helping patients make an informed decision regarding their cancer care, advance directives and overall planning for the future as they go through treatment. It is also a helpful tool for physicians — since patients rely on them to make these estimates — and most often, they do not have well-developed tools like the nomogram.

What can data from the tool tell patients or health care providers?

The average survival for our entire cohort was about 14 months. About a year is the average survival for most patients with cancers of unknown primary. ... We then looked at patients who had lower scores and higher scores: the higher the score in the nomogram, the worse your survival. Our lowest-scoring people — the lower-third scores — had survival estimates as high as 40 months ... and the other group, which had the highest scores, had a median survival of only about four months.

You can see that patients fall under a very large spectrum, and it is kind of important to recognize which patients will do well and which will do poorly with the disease. For patients who would otherwise do well, you can tend to be more aggressive, and you can plan for a longer duration of treatment. These are also patients who might be good candidates for clinical trials. Patients who are really doing poorly may not be patients who are good clinical trial candidates for now, but it might be important to develop more aggressive therapies for them in the future. These are not patients who can wait two or three months to start treatment. These are patients for whom you need to act quickly.

Is there potential for the tool to cause panic in patients as they learn this information about overall survival?

Even though we think that estimate of survival in patients with cancer may cause panic, evidence has shown that not knowing one’s prognosis results in over-treatment, heightened patient and caregiver distress, poor quality of life and, ultimately, poor outcomes. That is why a discussion of prognosis is endorsed in key cancer group guidelines. This tool allows a more objective and candid discussion of this aspect of care in cancer of unknown primary.

How could the tool be further developed?

It would be important to keep working on this nomogram and add more and more factors that can adjust for newer treatments and newer prognostic factors that we figured out, especially genomic profiling. ... I think integrating the tool into clinical care of patients is definitely a good way of looking at how this would perform in the real world and also integrating it in clinical trials.

What we actually need is more clinical trials for this orphan disease. Cancer is pretty overwhelming to begin with, irrespective of where it is. It is more overwhelming when you don’t even know what the primary (cancer) is, and you cannot identify yourself with the 95% of other patients with cancer who know (where their disease originated). The uncertainty around prognosis, both from providers’ as well as patients’ perspectives, can make it very, very challenging. This nomogram is a tool
to help our patients and our providers feel more comfortable with what their disease is and give them a resource that can help them make better decisions in clinical care.

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.