Studies of side effects related to adjuvant breast cancer regimens with focus on chemotherapy

Abstract

Radiotherapy, endocrine therapy, chemotherapy and targeted therapy adjunct to surgery have a critical role in the management of early breast cancer for outcome improvement. Anthracycline containing chemotherapy reduces 10-year incidence of breast cancer mortality by an absolute 6.5% compared to no adjuvant chemotherapy and addition of taxanes to anthracycline regimes leads to additional gain of 2.8% in 8-years breast cancer mortality. Dose dense chemotherapy improves breast cancer mortality by an estimated relative reduction of 40%. Trastuzumab, the first anti-HER2 agent among plenty, administered for one year, led to a paradigm shift in the management of HER2 positive breast cancer and significantly altered prognosis. Adjuvant trastuzumab improves disease free survival by a relative risk reduction of 40%. Despite beneficial effect of chemotherapy and anti-HER2 targeted therapy, acute or late toxicity can be dose limiting and even influence patients’ quality of life during treatment and long time after it is completed. Given the good prognosis of breast cancer and the millions of breast cancer survivors, therapy related toxicity affects a considerable number of women. Efforts to escalate or de-escalate treatment are ongoing and potential trade-offs in safety are monitored, resulting at best to improved benefit-risk balance. The aim of the current thesis was to examine heart failure outcomes and management after breast cancer diagnosis compared to a population of women with heart failure and toxicity outcomes related to tailored dose dense chemotherapy namely, cardiotoxicity after combination with trastuzumab, neutropenia related events and premature ovarian insufficiency.

The Swedish Registry for heart failure (SwedeHF) and the national health care registries were utilised for the purposes of Paper I. Patients enrolled in the SwedeHF registry between 2008 and 2013 were included and followed for a median period of two years. Patients with breast cancer history, identified through the National Cancer Registry, and age-matched controls (1:5) were investigated. Heart failure related baseline characteristics and outcomes did not differ amid presence of breast cancer history among women registered in the SwedeHF registry with incident heart failure. Differences in the history of myocardial infarction, administration of aspirin and device therapy were observed among women with prevalent heart failure, depending on previous breast cancer history. Breast cancer history did not alter heart failure outcomes but time from heart failure diagnosis did; women with prevalent heart failure had worse survival than those with incident heart failure.

Papers II-IV investigated different toxicity aspects related to the population enrolled and treated in the PANTHER phase III study comparing tailored (protocol predefined dose escalation or de-escalation) and dose dense (every two weeks) chemotherapy to standard dose three weekly chemotherapy. Paper II, investigated if tailored dose dense chemotherapy can further improve trastuzumab efficacy, compared to combination with standard chemotherapy and whether this combination would jeopardise cardiac safety; both parts of the study were predefined. The trastuzumab and tailored dose dense group demonstrated a 32% relative improvement in risk for breast cancer relapse but the results did not reach formal statistical significance. Despite small reductions of left ventricular ejection fraction at four- and sixyears follow-up, no clinically meaningful difference in the risk for cardiotoxicity was demonstrated between tailored dose dense chemotherapy and standard chemotherapy compared as administration per HER2 treatment or as per chemotherapy group. Compliance to the planned chemotherapy schedule is related to better breast cancer outcomes, underscoring the value of prophylactic granulocyte-colony stimulating factor (GCSF). Efficacy of G-CSF in preventing neutropenic events and delays in the delivery of the planned chemotherapy dose were examined in a secondary analysis in Paper III. Administration of G-CSF reduced the risk for neutropenic events defined as febrile neutropenia or infection with low absolute neutrophil count. Although a comparison between the two treatment groups was not possible, within group investigation in the standard chemotherapy group revealed improved adherence to planned schedule by fewer dose delays when G-CSF was administered.

Chemotherapy-induced amenorrhea is a therapy-related adverse event but it also improves breast cancer survival. Efficacy of dose dense chemotherapy has been assumed to derive from increased incidence of chemotherapy-induced amenorrhea. Thus, the exploratory, post hoc studies in Paper IV investigated the incidence of chemotherapy-induced amenorrhea and impact on treatment outcomes, excluding patients receiving gonadotropin releasing hormone agonists. Even though the delivered mean chemotherapy doses in the experimental treatment group were higher than in the standard chemotherapy group, amenorrhea incidence at two years of follow-up between the two treatment groups did not differ. Hence, benefit of dose dense chemotherapy is deemed to stem from chemotherapy effect per se and not aggravated gonadal toxicity. Breast cancer relapse events were too few to make any inference on the impact of amenorrhea on breast cancer relapse outcomes and longer follow-up is required. Menstruation status did not impact efficacy of allocated treatment although a non-significant benefit from tailored dose dense chemotherapy was demonstrated and persisted in most breast cancer subgroups. Sub-group analysis of the different breast cancer subtypes did not reveal any influence of the baseline menopause status on the efficacy of given chemotherapy schedule, except in triple negative breast cancer. There is no known biological plausibility to explain this interaction and the possibility of a chance finding cannot be excluded.

In summary, at median follow-up of two years, having previously been diagnosed with breast cancer did not alter heart failure outcomes, compared to controls, although some limited discrepancies in existing comorbidities and heart failure treatment were observed. Moreover, increasing dose intensity did not relate to excess cardiotoxicity when combined with trastuzumab, did not increase the risk for neutropenic events, provided G-CSF prophylaxis is administered, and did not affect risk for premature ovarian insufficiency. Conclusively, patients with early breast cancer should be offered efficient breast cancer therapy related to their risk level, with proper supportive therapy and assessment of potential comorbidities such as cardiovascular risk factors.