Abstract
Tamoxifen exposure is a recognised risk for primary endometrial cancer. This case serves as a reminder to meticulously check the past medical history and inform patients of the risk-benefit of treatment as part of a shared-decision making process.
Keywords
Background
Breast and endometrial cancer share similar epidemiological characteristics, with oestrogen strongly implicated in the pathogenesis of the majority of cases. Tamoxifen, an oestrogen receptor (ER) antagonist, has been effectively used in the treatment of ER-positive breast cancer and is extensively prescribed as adjuvant therapy. More recently, the National Institute for Health and Care Excellence further recommended tamoxifen as a prophylactic agent in women at high-to-moderate risk of familial linked breast cancer.
Paradoxically, at the level of the endometrium, tamoxifen is associated with oestrogen receptor agonism and subsequent promotion of endometrial hyperplasia (a pre-cancerous condition when atypical). Under tamoxifen’s influence, there is progression of endometrial hyperplasia without atypia to atypical hyperplasia in approximately 50% of cases. 1 In addition, evidence suggests an 8.3% progression rate to become endometrial cancers, far higher than the background expected rate of 1.6%. 2 Notably, tamoxifen-associated endometrial tumours are recognised as having higher mortality than those without exposure to tamoxifen. 3 Given that breast cancer has an annual incidence of over 55,000 in the United Kingdom and oestrogen receptor positivity in 80% of cases, this potentially exposes a large population to the endometrial effects of tamoxifen. 4
This is important because a tamoxifen prescription in patients with pre-existing risk factors for endometrial cancer, including history of endometrial cancer, remains a ‘grey area’ in clinical guidelines. The British National Formulary does list ‘endometrial changes’ (including cancer) as a side effect of tamoxifen but does not list history of any endometrial proliferative disease as a caution or contraindication. Curiously, there is also a discrepancy between National Institute for Health and Care Excellence guidance on prophylactic therapy which recommends tamoxifen ‘unless patients have a past history or may be at increased risk of endometrial cancer’ and guidance on adjuvant prescription which does not. 5
Here we report a case of a patient who developed an ER-positive breast cancer two years after hysterectomy for an endometrial endometrioid carcinoma. She received adjuvant tamoxifen and a further two years later developed a vaginal recurrence of endometrial cancer. To the best of our knowledge, there is no report of the recurrence of endometrial cancer in hysterectomised patients subsequently prescribed tamoxifen for breast cancer.
Case presentation
In 2003, a 65-year-old nulliparous woman presented with post-menopausal bleeding. Hysteroscopy and curettage revealed atypical hyperplasia of the endometrium. Subsequently, the patient underwent a total abdominal hysterectomy, bilateral salpingo-oophorectomy and omental biopsy with peritoneal washing. Histopathology returned as a FIGO stage 1B grade 1 endometrioid adenocarcinoma (Figure 1) for which she underwent adjuvant external beam radiotherapy to the pelvis, with the field incorporating the upper two-thirds of the vagina.
Original 2003 ER+ endometrial adenocarcinoma.
In 2005, she presented with a mass in the upper outer quadrant of the left breast. Core biopsy showed an invasive ductal carcinoma and the patient elected to undergo mastectomy. The tumour was 35 mm in maximum diameter, grade 3, strongly ER-positive, Progesterone receptor (PR) and HER-2 negative. One of six lymph nodes excised was positive. After completing six cycles of adjuvant chemotherapy, she was commenced on tamoxifen therapy. Importantly, the patient was not receiving oestrogen-only hormone replacement therapy and had no history of obesity, diabetes or polycystic ovarian syndrome identified as the additional modifiable risk factors for endometrial cancer in tamoxifen’s Summary of Product Characteristics.
She remained in remission from both tumours until 2007 when she presented with vaginal discharge. Examination revealed a pea-sized nodule in the distal sub-urethral area. This was excised and histology demonstrated a highly ER-positive (Allred Score 8) endometrioid adenocarcinoma, identical in appearance to that of the original endometrial tumour (Figure 2). Subsequent metastatic survey revealed no other signs of disease. Her tamoxifen was stopped and she was commenced on an aromatase inhibitor (Anastrozole), an alternative anti-oestrogen blocking systemic conversion of circulating androgens into oestrogens. The patient was ostensibly free of either malignancy on follow-up eight years on from management of her endometrial recurrence.
2007 Sub-urethral recurrence of endometrial adenocarcinoma displaying strong and diffuse ER positivity.
Discussion
Although the precise aetiology of any cancer in any single case cannot be definitively proven, the combined weight of evidence outlined in this case report suggests tamoxifen to be the most likely cause of recurrence.
Key evidence of relapse is that the morphological and immunohistochemical findings confirmed an endometroid adenocarcinoma, identical in histology to the first endometrial lesion. In addition, the sub-urethral site of the isolated recurrence is a recognised clinical problem, with local vaginal spread a common site of endometrial cancer metastasis. Of note, in this case is that the tumour was just outside the radiotherapy field.
The tumour showed a strong and diffuse expression for ER, making it highly suspectable to tamoxifen-driven growth. The recurrence occurred <1.5 years after tamoxifen initiation, well within the 0.7–8.1 years mean lag period between commencement of tamoxifen therapy and diagnosis of tamoxifen-associated endometrial cancer documented in the literature. 7 However, the timeframe separating the original March 2003 presentation and the December 2007 re-presentation was 56 months. This far exceeds the median time to recurrence of 20.2–32 months acknowledged in endometrial cancer and is incongruous with evidence suggesting 92% of relapses would be expected within 36 months.8–10 Furthermore, in a trial following up 2385 women surgically managed for endometrial cancer, less than 1% of recurrences occurred after 5 years, only a few months on from the recurrence we observed. 10 While endometrial cancer recurrence is itself not an overly rare event, with lifetime relapse rates for stage 1 cancers around 6.5%, that a recurrence of such low probability transpired favours the idea of an external stimulant. 10
Formal assessment using the Naranjo Adverse Drug Reaction Probability Score, a tool to assess the causality of all adverse reactions, was also performed. A score of 4 was obtained, corresponding with tamoxifen being a ‘possible’ cause of relapse. Finally, there was no further recurrence, a definite risk in a patient with known metastatic spread, on switching to an aromatase inhibitor. The speculative inference is that if there were any residual malignant endometrial cells Anastrozole conferred successful treatment of the breast cancer without replicating the oestrogen receptor agonism of tamoxifen liable to propagate recurrence.
The authors recognise a single case report is not a sufficient evidence base from which to draw any firm conclusions about the risk–benefit profile of tamoxifen therapy in women with a history of endometrial cancer. However, we do believe this case serves as a reminder to meticulously check the past medical history, inform patients about the risks of treatment as part of the shared decision-making process and of the need to further investigate whether previous endometrial cancer should be a caution in all tamoxifen prescriptions.
