I would order the assay for this 70-year-old patient with node-negative, ER-positive disease. The information is useful to me in my discussions with patients regarding their receiving chemotherapy or not.

For this 70-year-old postmenopausal patient with negative nodes, I would discuss the uncertainties around continuing an AI, and the results from the ATLAS and aTTom studies. Any perceived or real toxicities of the AI, primarily bone loss, would deter me from continuing anastrozole. If the patient were concerned about worsening of osteopenia or had perceived or real bone or joint pain, I would stop therapy in the absence of proven benefit.

I would recommend the Oncotype DX assay in this scenario of the 70-year-old patient. I discuss treatment options with patients who are older or who have comorbidities.

I would stop anastrozole for the 70-year-old postmenopausal woman because we don’t have data about the effect of continuing therapy. I struggle with these patients who have received 5 years of an AI alone. I believe the idea of a perpetual risk is real. Whether it matters to have received 5 years of an AI versus 5 years of tamoxifen in terms of that residual ongoing risk is uncertain. I believe we’ll have more data regarding that in the near future.

I would order an Oncotype DX assay for this 70-year-old patient with node-negative disease, especially if the tumor was high grade. If the tumor was intermediate or high grade and had a low expression of hormone receptors I believe that endocrine therapy alone would have less benefit, and I’d lean more toward considering the addition of chemotherapy.

I would stop anastrozole for the 70-year-old postmenopausal woman. We don’t have data on the effect of AIs administered for an extended duration. Currently I believe that the standard is to stop AI therapy at 5 years.

I would order an Oncotype DX assay for the 70-year-old postmenopausal woman.

For the 70-year-old patient with node-negative disease, I would have a long discussion with her regarding what is generally known about the duration of hormone therapy versus the duration of AI therapy. My bias is toward continuing anastrozole for the 70-year-old patient, but no data are available for that yet.

Although we don’t have data yet, my belief is that 10 years of an AI will be more effective than 5 years, as 10 years of tamoxifen is more effective than 5. The toxicities of AIs over the long term are modest and well known. I would rather turn out to be wrong with a little overtreatment in patients who tolerate it than turn out to be wrong with undertreatment in patients who might have benefited.

I would not recommend Oncotype DX for this 70-year-old patient with a node-negative IDC because it would be unlikely to change my recommendation. In general the benefits of chemotherapy are not likely to be significant enough to warrant the potential toxicity in a 70-year-old. Because of competing risks of mortality, an elderly person with comorbidities would be likely to have a very small absolute benefit from chemotherapy even with a high-risk score.

I would stop the anastrozole for the 70-year-old postmenopausal woman who had received 5 years of an AI without experiencing problems. We don’t have any data yet to say that additional therapy would be beneficial. I believe for a patient at high risk you could consider administering an additional 5 years of tamoxifen. But in the absence of supportive data, I stop therapy after 5 years of an AI for virtually all patients such as this.

I would definitely recommend an Oncotype DX assay for this 70 year-old patient because my decision whether or not to administer chemotherapy may change depending upon the score.

I would recommend that the 70-year-old with negative nodes stop anastrozole. We don’t have data to support use of an AI beyond 5 years. We certainly have good reasons to believe that longer-duration therapy will be better, but we don’t know that. I would recommend that the patient stop because we don’t know about the benefits and potential toxicities. If the patient were inclined to continue I would not object, but she would need to know that it’s not supported by data.

For a 70-year-old patient with a node-negative tumor I would decide whether to order an Oncotype DX assay depending on the tumor characteristics. If the biology is favorable with concordant results from biopsy and lumpectomy samples, I don’t order the assay. I would administer endocrine therapy alone. If any adverse features were present I’d order the assay.

If the patient were 70 years old with negative nodes, I would consider the tumor biology. A patient with slow-growing cancer and a low Recurrence Score would have favorable biology. In that situation, I would allow the patient to make the choice. If she wanted to stop for a couple of years until we have more data on a longer duration of therapy, I would be comfortable with that. On the other hand, if the tumor had unfavorable characteristics, such as Ki-67 staining of greater than 15%, I would be concerned that the tumor would recur if anastrozole were discontinued, and I would suggest that the patient continue treatment.

For this 70-year-old woman, my decision would be greatly influenced by what the patient was willing to do and by the tumor grade. If the tumor was Grade I, I would have a discussion about what the patient was interested in doing and what degree of benefit would motivate her to receive chemotherapy. For a patient in excellent health who wishes to be aggressive and know as much as possible, I would order the assay. If the patient had a Grade I tumor with any comorbidities and was not interested in receiving chemotherapy, I wouldn’t send for a Recurrence Score. I believe that having the information from the Oncotype DX assay in some cases can lead to a much more complicated situation in which the patient experiences a long period of recovery from toxicity for marginal benefit.

I’m more likely to extend hormone therapy for higher-risk disease than for lower-risk disease. This would hold true whether a patient were continuing tamoxifen, switching from tamoxifen to an AI or continuing an AI. Because of the absence of data on continuing an AI for more than 5 years, I’m more likely to continue with AI therapy for 7 to 8 years and then stop for a patient at moderate risk. For a patient with high-risk disease who is receiving an AI and tolerating it well, I continue past 5 years. For patients who are not tolerating it well, I switch to tamoxifen.