Conference Coverage

Early infection-related hospitalization portends poor breast cancer prognosis

Key clinical point: Particularly close monitoring for adverse breast cancer outcomes is warranted for women with an infection-related hospitalization during the first year after breast cancer diagnosis.

Major finding: Women hospitalized for sepsis or a respiratory or skin infection during the first year after being diagnosed with stage I-III breast cancer were up to 4.8 times more likely to subsequently die of breast cancer than those without an infection-related hospitalization.

Data source: This was a prospective observational study of 8,338 Stockholm-area breast cancer patients followed for a median of 4.9 years post diagnosis.

Disclosures: The study presenter reported having no financial conflicts regarding this university-funded study.


 

AT SABCS 2015

References

SAN ANTONIO – Hospitalization for an infection within the first year following diagnosis of primary nonmetastatic breast cancer is a red flag for increased risk of subsequent development of distant metastases or breast cancer–related death, Judith S. Brand, Ph.D., reported at the San Antonio Breast Cancer Symposium.

Judith S. Brand, Ph.D.

Judith S. Brand, Ph.D.

This increased risk for future adverse breast cancer outcomes was statistically significant and clinically meaningful for women hospitalized for respiratory tract or skin infections or sepsis. Those are the patients for whom particularly close monitoring for recurrence of breast cancer is warranted in the next 5 years, said Dr. Brand of the Karolinska Institute, Stockholm.

In contrast, hospitalization for a gastrointestinal or urinary tract infection didn’t achieve significance as an independent predictor of increased risk of adverse breast cancer outcomes.

Dr. Brand presented a prospective population-based study of 8,338 women diagnosed with stage I-III breast cancer in the Stockholm area during 2001-2008. During a median 4.9 years of follow-up after diagnosis, 720 women had an infection-related hospitalization, with the great majority of these events occurring during the first year.

The incidence of hospitalization for sepsis among breast cancer patients in their first year post diagnosis was 14-fold greater than in the general Swedish female population matched for age and year. Respiratory infections resulting in hospitalization were fourfold more frequent than in the general population, skin infections were eightfold more common, and GI infections were twice as common.

Infection-related hospitalizations had a strong and independent association with breast cancer mortality during follow-up, as seen in a multivariate analysis adjusted for age at diagnosis, comorbid conditions, infectious disease history, type of breast cancer therapy, and tumor characteristics including size, grade, hormone receptor status, and lymph node involvement.

Moreover, the risk of developing distant metastases was 50%-78% greater in breast cancer patients hospitalized for respiratory infection, sepsis, or skin infection, compared with breast cancer patients who didn’t have an infection-related hospitalization.

“We think the sepsis results are the most interesting findings,” Dr. Brand said in an interview. “Sepsis could be an expression of an immunosuppressed state. And sepsis itself can induce immunosuppression for a long time, which could trigger tumor growth. Animal studies have shown that in postseptic mice, tumor grows faster.”

Infection-related hospitalizations didn’t increase the future risk of locoregional recurrences.

Independent predictors of infection-related hospitalization included older age, comorbidities, markers indicative of greater tumor aggressiveness, and treatment with chemotherapy or axillary radiotherapy.

“This shows that the risk of infection-related hospitalizations is not only due to immunosuppression caused by chemotherapy, but that the characteristics of the tumor itself play a role, as do patient characteristics. This is the first epidemiologic study to show with very extensive data that all three elements contribute to the risk,” Dr. Brand said.

bjancin@frontlinemedcom.com

Next Article:

   Comments ()