Six factors are associated with the eventual development of invasive breast cancer after an initial diagnosis of ductal carcinoma in situ (DCIS), an often benign condition, according to a new Dutch meta-analysis.
The researchers reviewed 17 studies and found a total of 26 prognostic factors; only six factors were statistically significant: African American race (pooled estimate [ES], 1.43), premenopausal status (ES, 1.59), detection by palpation (ES, 1.84), involved margins (ES, 1.63), high histologic grade (ES, 1.36), and high p16 expression (ES, 1.51).
There may be others, as the literature search yielded some factors that had only been identified once and thus were ruled out because they needed confirmation, the authors comment. They were led by Jelle Wesseling, MD, PhD, professor of breast pathology, Netherlands Cancer Institute and Leiden University Medical Center in the Netherlands.
Although the study results also need validation, the authors hope that the data can advance the understanding of DCIS.
“We hope our work will help reduce the burden of intensive treatment that thousands of women with low-risk DCIS undergo annually,” Wesseling said in a press statement.
The authors explained that the majority of DCIS, if left untreated, is not destined to progress to invasive disease and thus will never become life-threatening. Nevertheless, all cases of DCIS are currently treated, because at present there is no way to predict which cases will become invasive.
“This implies that many women are overtreated,” they write.
There is a need for strategies to improve breast cancer screening to detect only hazardous disease, the researchers emphasize. But “progress in this field has been slow,” write Wesseling and colleagues.
Indeed, in the past, “numerous prognostic factors have been reported, but none have shown to be of sufficient value for implementation into the clinic,” they say.
On a positive note, the authors indicate that major studies are underway in the management and treatment of DCIS that should eventually “translate promising prognostic factors to clinical practice.”
One of these initiatives is the PRECISION (PREvent ductal Carcinoma In Situ Invasive Overtreatment Now) study, funded by Cancer Research UK and the Dutch Cancer Society. Also, noninferiority trials — like LORD, LORIS, and COMET — are underway and “will be important in prospective validation of prognostic factors,” say the study authors.
But what can be done right now when seeing DCIS patients?
Eileen Rakovitch, MD, medical director, Louise Temerty Breast Cancer Centre, University of Toronto, Canada, tells Medscape Medical News that nomograms are available and are used for assessing local recurrence risk, but these tools have not been well validated.
She added that “most experts will estimate the risk of developing a local recurrence on the presence of clinical and pathological risk factors such as age at diagnosis, size of the DCIS lesion, the nuclear grade, and resection margin status (positive/negative).”
Another available tool, said Rakovitch in an email to Medscape Medical News, is the Oncotype DX Breast DCIS score (DS), one of the first molecular expression assays in DCIS.
The ECOG-ACRIN E5194 study initially reported the DS as an independent predictor of local recurrence (DCIS or invasive breast cancer) in women treated by breast conserving surgery (BCS) alone. The tool has since also been validated as predictive after BCS and radiation.
Rakovitch said that she and other academic researchers have also worked with Genomic Health (a provider of genomic-based diagnostic tests that address overtreatment and optimal treatment of early-stage cancer) to integrate clinico-pathological predictors such as tumor size and age at diagnosis with the molecular-based DS to “provide a more accurate estimate of local recurrence risk following BCS compared to estimates based on the DS alone or clinico-pathological features alone.” ( Breast Cancer Res Treat. 2018;169:359-369 ).
The meta-analysis performed by the Dutch team began with a systematic review of 1781 studies from PubMed (1970 to 2018) to assess the risk of ipsilateral invasive breast cancer recurrence in women diagnosed and treated for DCIS. Only 40 studies met the eligibility criteria, which included having at least 10 invasive breast cancer events and at least 1 year of follow-up.
Next, the researchers used a tool to assess the studies for risk of bias, which resulted in 23 of the studies being excluded. This left 17 studies, with participants in each ranging from a small cohort of 52 to a very large group of 37,692; mean follow-up ranged from 3.2 to 15.8 years.
In the last step of the study, meta-analyses were performed on all factors associated with the recurrence of invasive breast cancer reported by more than one of the 17 studies. The authors then calculated the average effect size for each factor.
The study authors called out a general problem with DCIS research. “New studies need to capture information about whether the cancer recurrence was DCIS or a subsequent invasive cancer, and whether these are true recurrences or new, primary lesions,” added Wesseling.
Cancer Epidemiol Biomarkers Prev. Published online April 25, 2019. Abstract