RWE from Southwest Finland shows readiness to adopt new medications, utility of medical data

Utilising the data lake of the Hospital District of Southwest Finland (TYKS), Medaffcon, Pfizer and TYKS have examined the clinical adoption of new medications for advanced renal cell carcinoma and the utility of data extracted without extensive manual extraction.

The introduction of new medications has resulted in a change in the treatment of advanced renal cell carcinoma in Southwest Finland, reveals a real-world evidence (RWE) study conducted by Medaffcon, Pfizer and the Hospital District of Southwest Finland (TYKS).

The study shows that:

• The share of suntinib as the first-line treatment for patients with advanced renal cell carcinoma has decreased from 79 per cent to 52 per cent after 2018
• The share of cabozantinib has jumped to 31 per cent
• The share of ipilimumab-nivolumab has jumped to 10 per cent

Cabozantinib has been available in the country as a reimbursable medication since 2018 and ipilimumab-nivolumab since 2019.

“These medications have already been proven useful in large-scale prospective studies – that’s not new,” reminds Dr. Kalle Mattila, medical oncologist at the Urology Research Unit at TYKS.

“But what’s new is how these medications behave in real life, in patients who haven’t been chosen based on strict criteria,” he adds. “This study found, for instance, that a large share of patients didn’t receive any treatment, which of course indicates that there are patients who wouldn’t benefit from treatment due to their general health or co-morbidity status.”

In Finland, renal cell carcinoma is the third most common urogenital cancer after bladder and prostate cancer. The disease has an annual incidence of 18 cases per 100,000 and accounted for nearly three per cent of all cancer-related deaths in the country in 2020.

Structured or unstructured, real-world evidence provides key insights

The study had two primary objectives: to determine the extent to which the new treatment options have been adopted to clinical practice and to assess the utility of automatically extracted medical data from Auria, the data lake of TYKS.

Auria, the researchers found, contained altogether 1,122 patients with renal cell carcinoma in 2010–2021. About a third, or 366, of them were identified as having advanced renal cell carcinoma and 191 as having received systemic treatment for advanced renal cell carcinoma.

“We managed to find the necessary information on patients, their treatments and treatment outcomes from the patient data system without extensive manual extraction. And we managed to produce relevant results on how the medications have been used and how they’d affected the survival of patients,” tells Dr. Mattila.

He points out that the study relied principally on data that healthcare professionals entered into patient records during everyday clinical work. Although the records are not strictly structured, the extraction methods produced data that enabled the researchers to assign each patient to a risk category, as defined by the International Metastatic RCC Database Consortium (IMDC).

“We were then able to group treatment outcomes based on the IMDC risk score,” he says. “It demonstrated neatly that the score predicts the survival rate of patients very accurately – as it should – also in our data. The risk category was clearly a predictor of treatment outcomes, making it a useful clinical marker.”

The data did not, however, contain any structured entries on disease progression.

“We couldn’t extract or search for progression-free survival because the relevant dates are not defined in any structured way in the records. What we did instead was extract the time until the initiation of next-line treatment, which is a surrogate for progression-free survival,” recounts Dr. Mattila.

Risk score correlates with survival rate, advancement to next-line treatment

The median overall survival time was 61.9 months for patients with a favourable risk profile, 28.6 months for patients with an intermediate risk profile and 8.1 months for patients with a poor risk profile. Also healthcare resource utilisation correlated with the risk score:
poor-risk patients had two times as many inpatient days as intermediate-risk and four-times as many as favourable-risk patients.

“We saw that the new medications worked roughly as we expected based on prospective studies, meaning they helped extend the survival rate of patients,” he tells.

What was somewhat surprising, though, was how few patients with a poor risk score graduated from the first treatment line to the second due to co-morbidities, the disease progression or overall health status. While the share stood at almost 75 per cent for both favourable and intermediate-risk patients, it plummeted to 35 per cent for poor-risk ones.

“I was surprised by how small a share of poor-risk patients receive a second line of treatment after the first,” he admits.

Although combination treatments are already the default first-line treatment for intermediate and poor-risk patients at TYKS, Dr. Mattila believes the findings will further encourage clinicians to resort to combination treatments early despite their greater adverse side-effects.

“You simply don’t have the option to try several lines of treatments with patients with a bleak prognosis – with a risk score of three or more,” he elaborates.

Dr. Liisa Ukkola-Vuoti, senior scientific advisor at Medaffcon, highlights that the results were published only two years after the research project began, a formidable feat given the time-consuming nature of securing permits, gathering data and conducting analyses.

“I want to thank the research team and Auria for smooth co-operation,” she says. “The study design and especially the utilisation of structured data sped up the research process.”

Real-world Evidence reveals shift in clinical practice in Southwest Finland

• Data from Auria, the data lake of the Hospital District of Southwest Finland (TYKS), reveal that new medications for advanced renal cell carcinoma have been widely adopted to clinical practice since becoming available in the late 2010s.
• Comprising largely structured patient records, the data were the foundation of a recent study by Medaffcon, Pfizer and TYKS.
• The study also found that the data lake can be a valuable resource even without the use of extensive manual data extractions, with the research team using the data to not only measure the adoption of new medications but also to assign a risk score to each patient.
  
• The risk scores, meanwhile, were shown to correlate with both overall survival rates and the likelihood of a patient advancing to the next line of treatment
• Dr. Kalle Mattila, medical oncologist at TYKS, views that the finding that only 35 per cent of poor-risk patients advance to second-line treatment will encourage oncologists to resort to combination medications early.