Cancer-Specific Outcomes in the Elderly with Triple-Negative Breast Cancer: A Systematic Review

Round 1

Reviewer 1 Report

Treatment of older women with breast cancer has become of utmost relevance, due to the aging of population. This group of patients is underrepresented in clinical trials, making it difficult to give evidence-based recommendations about treatment, either surgery or adjuvant treatments (chemotherapy, radiotherapy). This is especially true for the subgroup of patients with high risk breast cancer subtypes, such as triple negative disease. Outcomes in this specific population have been underreported in clinical literature, and the optimal treatment strategy is far from being set.

For these reasons, I think that the paper “Cancer Specific Outcomes in the Elderly with Triple Negative 2 Breast Cancer: A Systematic Review” is of great interest and should be considered for publication with some minor revisions.

The paper is well written, although I recommend to recheck the spelling of some words all over the manuscript.

The authors chose breast cancers specific survival as the main outcome measure for their systematic review, thus excluding many studies which considered more inclusive outcome measures (i.e. overall survival). I agree with the authors that their decision is motivated by the purpose to avoid confounding due to competing comorbidities. I think that this should be discussed more clearly, at least in the discussion section, and put in context with the evidence from other studies, such as those cited by the authors (Haque et al. 2019 or Algan et al. 2018) and others (by example, Crozier JA, Pezzi TA, Hodge C, Janeva S, Lesnikoski BA, Samiian L, Devereaux A, Hammond W, Audisio RA, Pezzi CM. Addition of chemotherapy to local therapy in women aged 70 years or older with triple-negative breast cancer: a propensity-matched analysis. Lancet Oncol. 2020 Dec;21(12):1611-1619. doi: 10.1016/S1470-2045(20)30538-6. PMID: 33271091). The benefit of adjuvant therapies on overall survival has been demonstrated in these studies, so the results of the present review need to be more deeply discussed. Do patient characteristics (comorbidities, performance status) differ in the studies reviewed by the authors and in other studies demonstrating a survival benefit with more intensive treatment?

Chronological age has multiple limitations when treating older adults with cancer, as it does not provide useful information on the biological characteristics of the patient that could be more relevant to treatment tolerance and survival. These limitations should be more clearly stated.

Author Response

Please see attachment 

Author Response File: Author Response.docx

Reviewer 2 Report

Abstract

The statement is made in the abstract – ‘These treatment differences were associated with a decrease in BCSS (HR 0.250, p<0.001) among women ≥70 years of age compared to a younger cohort’. As stated this appears to apply to all studies but actually looks to be derived just from the Zhu paper and relates in patients >70 to the difference between those who received surgery and those that did not.

The more relevant general statistic to quote from this study would be ‘after correcting for confounding factors, advanced age at diagnosis (>70 years) for CSS, HR= 2.125; 95%CI, 1.664 to 2.713; p<0.001; for OS, HR, 3.042; 95%CI, 2.474 to 3.740; p><0.001’, OR ‘more than doubling of cancer specific mortality in the elderly cohort (5.9% vs. 2.7% in ≤70, 127 p<0.0001).’

Introduction

The statement is made that TNBC is more common in young women – but there is little demographic detail. As this review is about TNBC in the context of age suggest more detail for the interest/education of the reader. For example the higher TNBC rate does not appear to have a threshold age where it suddenly becomes less common but rather becomes relatively less prevalent with each decade (eg ref Bauer 2007 DOI: 10.1002/cncr.22618).

Accepting that TNBC is a smaller % of breast cancers in older age groups, if we consider individual prevalences, does actually TNBC become less common per 100,000 of the population in older cohorts, or is the falling proportion more driven by an increase in non-TNBCs diluting the proportion that have TNBC.
If TNBC absolute prevalences are lower, are postulates made about why it is less common in older people - ? based on known risk factors: alcohol, obesity etc.

Methods

BCSS is an appropriate choice of outcome as described.

Metastatic studies were presumably not sought, although stage of disease is not mentioned in the search methodology (though it should be) – I note  it is stated that patients with stage IV disease were excluded.
The relative response rates, PFS and BCSS by age in MBC would have been of interest here if metastatic specific studies were available as it would add data on treatment tolerance or response that could both extrapolate to adjuvant treatment utility and help explain survival differences . In this regard, outcomes by age group are often included in sub-group analyses of metastatic breast cancer treatment trials, even though I would guess often none of the listed search terms would come up. Would such data have been captured? If not to be included I would probably make the title ‘Triple Negative Early Breast Cancer’.

The absence of co-morbidity data is an important confounder as the authors note. Is any data source available that gives such data for the population ages involved to make wither comment or estimate on what the impact of co-morbidities might be.

Results

3.2 Clinicopath features
Interesting data - grade and other tumour features all appear favourable suggesting less aggressive phenotypes. Possibly a small table showing the data from each of the studies on this?
I presume there is no data from the literature that looks at sub-typing TNBCs on detailed expression profile? As grade and LN status are different one wonders if older women have less aggressive sub-types as suggested by the authors in the discussion, which could also impact chemotherapy benefit.
It is interesting that tumours in older women are smaller also despite them being of an age where I presume mammographic screening is less prevalent than 50-70 yrs. Worth mentioning?

3.3 Zhu study
The statement ‘However, for patients with stage III disease, woman ≥70 years old were significantly less  likely to receive surgical excision (14.2% vs 9.1%, p=0.004).’ is confusing. This appears to say 14.2 and 9.1% DID undergo resection, which is very low. The paper states - A higher proportion of elderly women with stage III TNBC didn’t receive surgical excision compared with the younger patients (14.2% vs 9.1%, p = 0.004 by chi-squared test).
Patients with more advanced disease (i.e. stage II-III), were less likely to receive radiotherapy compared to those with stage I disease (p<0.001). – this would be routinely expected as Stage I disease often undergoes local excision which routinely then requires RT. A more interesting fact is that ‘for patients with relatively advanced diseases (stage II or III) more than a half of elderly women forewent radiation’ as these patients would often have had mastectomy after which RT is generally recommended to the chest wall in stage III disease.

 3.4 Bhoo-Pathy study

This paper is hard to analyze as presented data is patchy relative to age. That older patients received less RT is to be expected as there were more mastectomies which do not then routinely require RT.

3.5 Kaplan study

The presentation of data here is a little confusing although the paper is good. More data here could go into Table 2 I think.

The statement ‘Elderly patients with TNBC were less likely to receive chemotherapy and less likely to complete the full course of treatment (74% vs 84%, p=0.001).’ could be shortened to ‘Elderly patients with TNBC were also less likely to complete the full course of treatment (74% vs 84%, p=0.001) (less repetitive).

3.6 Kozak study - No issues except could be more concise.

3.7 Syed study – remarkable that 0% of older people received chemo for TNBC. At odds with other studies. Notably this study cohort starts in 1973, pre-dating adjuvant chemo use, especially in older women where tamoxifen was a focus even in TNBC. HER2 was not measured until around 2000 so TNBC as a concept did not exist for much of this cohort at time of therapy.

 Table 2
there is extra data in the papers that could be synthesized into a more comprehensive survival data table with possibly separate columns for surgery, RT and chemo applications as well as general overall survivals by age as well as the impact of treatments on age. If data is too much could have separate tables for treatment differences and survival impacts and differences.

At present although there is plenty of useful data extracted it is presented in a somewhat selective way with some data missing from some studies that is present from others.

Discussion

Nice first paragraph.
In the second paragraph the data after the sentence ‘Despite the paucity of data, our study demonstrated that elderly 202 patients with TNBC were less likely to receive adjuvant therapy, including both chemo-203 therapy and radiotherapy.’ I believe refers to breast cancer in general, not TNBC specifically (although it is still relevant),. Worth mentioning this as I was initially confused.
The point about possibly less aggressive TNBC sub-types occurring in the elderly is well-made with ref to the Syed data. Could comment/suggest that a study sub-typing TNBCs by age-groups into sub-types could further clarify as well as identify elderly patients with more chemosensitive sub-types.

Other unanswered questions that could be mentioned
Does early chemotherapy make a difference in older patients as seen in young TNBC ?

How are older women represented in cisplatin and capecitabine trials? Should these be considered?

How do metastatic chemotherapy benefits compare?

Author Response

Please see attachment 

Author Response File: Author Response.docx