Search

TPS12144 Background: Patients with locally advanced squamous cell carcinoma of the head and neck (LASCCHN) receive curative chemoradiation (CRT) with Cisplatin, as a standard of care. A meta-analysis of 52 randomized trials comparing Low Dose (LD) and High Dose (HD) schedules demonstrated differing toxicity profiles but hearing effects were not rigorously studied. Hearing loss associated with HD Cisplatin can result in survivorship challenges. A local study suggested a protective effect for LD Cisplatin in relation to ototoxicity and pharmacogenomic markers, MATE1 and COMT, to be associated with risk for ototoxicity. We hypothesize that LD cisplatin is associated with reduced frequency of hearing loss when compared to the standard HD cisplatin in LASCCHN patients on CRT and that differences in MATE1/COMT can predict for cisplatin-related ototoxicity and be identified prior to treatment. Our goal is to develop an innovative personalized treatment pathway incorporating predictive pharmacogenomics markers to improve the tolerability and survivorship outcomes of curative CRT for LASCCHN. Methods: This is a prospective, open-label, randomized clinical trial. Following informed consent, eligible LASCCHN patients planned for primary CRT will be stratified by tumor p16 status and then randomized in 1:1 fashion to either concurrent LD Cisplatin (40mg/m2 every week) or HD cisplatin (100mg/m2 every 3 weeks). The primary outcome is to measure the change in incidence of CTCAE grade ≥2 hearing loss and hearing-related quality of life (QOL) at 1 year. As part of secondary and exploratory outcomes, differences in survival, loco-regional control, global QOL and other toxicities (e.g. nephrotoxicity, neurotoxicity) will be assessed. The relationship between MATE1 and COMT, as predictors for cisplatin-related ototoxicity will be evaluated. Cost-effectiveness analyses comparing the two regimens will be assessed. Statistical plan: Based on rates of CTCAE grade ≥2 hearing loss in an earlier study (Winquist et al., 2016), assuming a conservative rate of hearing loss, amongst treated patients, of 60% with HD cisplatin and 30% with LD cisplatin, a total sample size of 92 patients would achieve > 80% statistical power, (two-sided, alpha = 0.05 test of two proportions) to detect these differences. 100 patients would be targeted to accrual for an assumed 5% noncompliance rate. For hearing related QOL, a two-sided, alpha = 0.05, two-sample t-test with 50 patients per group would achieve > 80% statistical power to detect an effect size of 0.60 and > 95% power to detect an effect size of 0.75. All analyses will be based primarily on the intent-to-treat population. An arms-length data and safety monitoring committee (DSMC) will review safety data bi-annually. Trial accrual status: 60 participants have been accrued. Clinical trial information: NCT03649048.

3530 Background: Blood based liquid biopsies examining circulating tumour DNA (ctDNA) have increasing applications in non-small cell lung cancer (NSCLC). Limitations in sensitivity remains a barrier to ctDNA replacing tissue-based testing. There is a paucity of data regarding the dynamics of ctDNA levels in the hours to days following a new treatment. We hypothesize that chemotherapy or radiation will yield an increased abundance of ctDNA in plasma by inducing tumor lysis, allowing for the detection of genetic alterations that were occult in baseline testing. Methods: Two prospective cohorts of 20 patients (pts) with stage III/IV NSCLC were enrolled. Cohort 1 (C1) contained pts starting the first cycle of platinum doublet chemoradiation (C1a, n=10) or the first cycle of platinum doublet cytotoxic chemotherapy ± immunotherapy without radiation (RT) (C1b, n=10). Cohort 2 (C2) contained pts receiving palliative RT alone. Two baseline samples were collected, the first ≤ 14 days prior to starting treatment and one immediately prior to treatment. In C1, subsequent samples were collected 3, 6, 24 and 48 hours post initiation of chemotherapy. Pts in C2 had samples collected immediately prior to RT fractions 2, 3, and 4. Samples were analyzed for ctDNA using the 36-gene amplicon-based NGS Inivata InVisionFirst-Lung assay. Results: Complete results were available for the first 35 of 40 enrolled pts, C1a – 10 pts, C1b – 9 pts, C2 – 16 pts. Detectable ctDNA was present at baseline in 27 pts (77%), 4 additional pts (11%) had detectable ctDNA in post treatment samples. Four of the patients with detectable ctDNA at baseline (15%) had new genetic alterations detected in post treatment samples. A total of 8/35 pts (23%) had new genetic alterations detected in the post treatment samples. Mutant molecule numbers increased with treatment in 23 of 31 (74%) pts with detectable ctDNA, C1 - 13 of 19 pts (68%) and C2 - 10 of 16 pts (63%). ctDNA levels peaked a median of 2.2 hours (IQR: 1.5 – 2.9 hours) after the initiation of chemotherapy and a median of 1 day (IQR: 1-2 days) after radiation was commenced. The percentage increase in ctDNA levels was a median of 29% (IQR: -18 to +112%) in C1. C2 had a median increase of 16% (IQR: 0 to +131%). Conclusions: ctDNA levels increase in the hours to days after starting treatment. ctDNA testing in the acute post treatment phase can yield results that were not evident in pretreatment testing. Application of this principle could improve ctDNA utility as an alternate to tissue-based testing and improve sensitivity for the detection of treatment-resistant clones.

e12532 Background: Early Breast Cancer Trialist Collaborative Group meta-analysis suggested significant reduction in bone metastasis and breast cancer mortality among post-menopausal women diagnosed with early breast cancer when treated with an adjuvant bisphosphonate. Cancer Care Ontario and American Society of Clinical Oncology clinical practice guideline published in 2017, recommends the use of an adjuvant bisphosphonate in postmenopausal women who are candidates for adjuvant systemic therapy. Our goal was to evaluate trends and factors associated with adjuvant bisphosphonate usage since the guideline’s publication. Methods: This study is a retrospective review of postmenopausal women treated for early breast at the London Regional Cancer Program from 2017-2018. Univariate and a multivariable logistic regression models were used to investigate factors potentially associated with adjuvant bisphosphonate use including age, stage, grade, estrogen receptor, progesterone receptor, HER2 receptor status and previous use of adjuvant chemotherapy. The percentage of patients offered, receiving and declining therapy was also recorded and the time interval from surgery to start of bisphosphonate therapy (< 6 months, 6-12 months or >12 months). Results: We identified 286 postmenopausal breast cancer patients, of whom 75 (28%) received adjuvant bisphosphonate therapy. In our multivariable model, cancer stage [odds ratio (OR) stage II vs. I=2.26, 95% confidence interval (CI) 1.08-4.74) and OR Stage III vs. I=4.94, 95% CI 1.84-13.17] and previous use of adjuvant chemotherapy (OR=2.76, 95%CI 1.37-5.55) were significantly associated with adjuvant bisphosphonate use. Among 133 patients who received adjuvant systemic chemotherapy, 51% were offered adjuvant bisphosphonate and of these 81% patients accepted therapy. Among a total of 75 patients receiving adjuvant bisphosphonate therapy 19% initiated therapy within 6 months of surgery, 48% within 6-12 months of surgery, and 33% after 12 months following surgery. Conclusions: Stage and previous use of chemotherapy were significantly associated with adjuvant bisphosphonate therapy. Our study observes the potential underutilization of adjuvant bisphosphonate therapy and possible need to start treatment earlier in some patients. Further education and awareness of the clinical practice guidelines regarding adjuvant bisphosphonate therapy may be warranted and additional population-based study investigating treatment patterns and real-world effectiveness.