The study was conducted from 1 August 2016 to 11 November 2019. Sixty-four patients were enrolled in the SOC arm and 65 in the CMW arm. The median follow-up for both arms was 2.3 years (range, 0.2–4.0 years). The median follow-up in the CMW arm was 3.0 years (range, 0.2–3.0 years). The median follow-up in the SOC arm was 1.6 years (range, 0.7–3.6 years). As expected, the follow-up in the CMW group was longer given its initial allocation prior to commencement of the SOC arm. Baseline characteristics and treatment modalities are summarised in Table 1. None of the characteristics were statistically different between the two arms. Most patients (at least 63% in each arm) received 70Gy of radiation treatment, whilst the rest received 60-66Gy of radiation treatment. The majority (> 97% in both arms) received intensity-modulated radiation therapy or volumetric modulated arc therapy. Most patients received concurrent systemic treatment with cisplatin chemotherapy as radiosensitiser.
Table 1
Baseline characteristics and treatment modalities between the 0.5% cocaine mouthwash arm and the standard of care arm.
| CMW (n = 65) | SOC (n = 64) | p-value |
Age, mean (years) | 63.3 | 62.3 | 0.818 |
Male (%) | 80.0 | 89.9 | 0.155 |
ECOG performance status 0 or 1 (%) | 96.6 | 98.3 | 0.788 |
Tumour location – oropharynx (%) | 44.6 | 57.8 | 0.429 |
Surgery performed prior to radiotherapy/ chemoradiotherapy (%) | 29.2 | 21.9 | 0.338 |
Non-smokers (%) | 21.5 | 21.9 | 0.732 |
Alcohol consumption – none or < 20g/day (%) | 55.9 | 67.9 | 0.412 |
Charlson comorbidity index > 5 (%) | 24.6 | 14.1 | 0.129 |
Baseline weight, mean (kg) | 77.1 | 82.2 | 0.100 |
Disease stage IV A/B, AJCC 7th edition (%) | 63.1 | 56.2 | 0.099 |
Cisplatin administration (%) | 86.4 | 94.1 | 0.378 |
Radiotherapy delivery, 70Gy (%) | 63.1 | 78.1 | 0.061 |
AJCC- American Joint Committee on Cancer; CMW- 0.5% cocaine mouthwash arm; ECOG- Eastern Cooperative Oncology Group; SOC- standard of care arm.
Rate of grade 2 or 3 mucositis did not differ between the SOC and CMW arms at various time points (Table 2A). Overall, although differences were noted between arms at different time points, these were not considered clinically meaningful when considered in isolation and particularly when examined in the context of other assessed variables. As expected, grade 2 and 3 mucositis occurred with higher prevalence towards the last week of treatment and resolved with longer follow up. More patients in the CMW arm had no mucositis compared to the SOC arm at the end of treatment and during follow- up. In comparison, in weeks 4 and 5 of radiotherapy or chemoradiotherapy, the CMW arm had a higher incidence of grade 2 mucositis (51.7% in the CMW arm vs. 42.4% in the SOC arm), but not grade 3 (5.0% in the CMW arm vs. 6.8% in the SOC arm). However, in the last week of treatment, these observations were reversed. That is, there was higher rate of grade 2 mucositis in the SOC arm but higher rate of grade 3 mucositis in the CMW arm.
Pain score ≥ 7 reported on the Numeric Rating Scale was not statistically different between the CMW and SOC arms, although numerically higher in the SOC arm in the last week of treatment (Table 2A).
Table 2
A. Rate of grade 1–3 mucositis and Numeric Rating Scale over time in the cocaine mouthwash arm and the standard of care arm.
Variable | Arm | T1 (week 1/2 of CRT) | T2 (week 4/5 of CRT) | T3 (last week of CRT) | T4 (1-month follow-up) | T5 (3-month follow-up) |
No mucositis (%) | SOC | 80.0 | 10.2 | 4.1 | 47.2 | 79.1 |
CMW | 76.3 | 23.3 | 21.2 | 68.1 | 91.8 |
Grade 1 mucositis (%) | SOC | 18.0 | 40.7 | 28.6 | 41.5 | 16.3 |
CMW | 18.6 | 20.0 | 9.6 | 12.8 | 2.0 |
Grade 2 mucositis (%) | SOC | 2.0 | 42.4 | 46.9 | 11.3 | 4.7 |
CMW | 5.1 | 51.7 | 40.4 | 14.9 | 6.1 |
Grade 3 mucositis (%) | SOC | 0 | 6.8 | 20.4 | 0 | 0 |
CMW | 0 | 5.0 | 28.8 | 4.3 | 0 |
p-value# | 0.685* | 0.046* | 0.009* | 0.008* | 0.053* |
Pain score ≥ 7 on NRS (%) | SOC | 0 | 3.4 | 17.7 | 0 | 0 |
CMW | 3.4 | 5.2 | 11.8 | 0 | 0 |
p-value | 0.609* | 0.432* | 0.402* | 0.374* | 0.180* |
# - comparison of all grades of mucositis per time point. *- aggregated p-value at the specified time points. CMW- 0.5% cocaine mouthwash arm; CRT- radiotherapy or chemoradiotherapy arm; NRS- numeric rating scale; SOC- standard of care arm; T- time point.
Hospitalisations due to any cause occurred in approximately 50% of patients with the mean length of hospital stay between 5–12 days highlighting the significant morbidity in patients with head and neck cancer undergoing radiotherapy or chemoradiotherapy (Table 2B). Although the length of stay was statistically longer in patients in the CMW arm, it is difficult to draw clinical conclusions from the observation. Enterostomy tube was required in up to 37% of patients. However, these results were not statistically different between the treatment arms despite numerically higher rates in the CMW arm, noting numerically more patients in the SOC arm received 70Gy dose of radiation treatment with concurrent cisplatin chemotherapy. Weight loss occurred in both arms and the mean weight was 71kg at the 3-month follow-up period in both arms. The weight loss was not significantly different between groups, with or without adjustment for the baseline weight. However, the reduction of weight overall was significant within each of the arms when compared to the baseline weight.
Table 2
B. Hospitalisation rate, hospitalisation length of stay, enterostomy feeding tube rate and weight at 3-month follow-up in the cocaine mouthwash arm and the standard of care arm.
Variable | Arm | Value | p-value |
Hospitalisation rate (%) | SOC | 45.3 | 0.332 |
CMW | 53.8 |
Hospitalisation length of stay, mean (days) | SOC | 5.9 | 0.015 |
CMW | 11.5 |
Enterostomy feeding tube rate (%) | SOC | 21.9 | 0.061 |
CMW | 36.9 |
Weight at 3-month follow-up, mean (kg) | SOC (baseline, 82.2kg) | 71.5 | 0.969* |
CMW (baseline, 77.1kg) | 71.4 |
*- reduction of weight overall was significant within each of the arms when compared to the baseline weight but not between arms at any of the specified time points (T1-5). CMW- 0.5% cocaine mouthwash arm; SOC- standard of care arm.
With respect to pain management, rate of analgesia prescription at steps 1–3 of the analgesic ladder did not differ significantly between the two arms (Table 3). Unsurprisingly, the rate of analgesia prescription was highest during the last week of radiotherapy or chemoradiotherapy and subsequently dropped with longer follow-up.
Table 3
Rates of analgesic prescription use in the cocaine mouthwash arm and the standard of care arm.
Variable | Arm | T1 (week 1/2 of CRT) | T2 (week 4/5 of CRT) | T3 (last week of CRT) | T4 (1-month follow-up) | T5 (3-month follow-up) |
Rate of analgesia prescription, step 1 (%) | SOC | 26.9 | 88.3 | 92.2 | 36.5 | 15.6 |
CMW | 37.3 | 85.7 | 89.1 | 50.0 | 24.5 |
p-value | 0.244* | 0.666* | 0.589* | 0.174* | 0.281* |
Rate of analgesia prescription, step 2 (%) | SOC | 1.9 | 25.0 | 37.3 | 7.7 | 2.2 |
CMW | 8.5 | 56.5 | 51.9 | 18.8 | 4.1 |
p-value | 0.128* | < 0.001* | 0.133* | 0.100* | 0.608* |
Rate of analgesia prescription, step 3 (%) | SOC | 0 | 23.3 | 41.2 | 17.3 | 2.2 |
CMW | 3.4 | 25.8 | 42.6 | 21.3 | 12.2 |
p-value | 0.180* | 0.751* | 0.883* | 0.617* | 0.064* |
*- aggregated p-value at the specified time points. CMW- 0.5% cocaine mouthwash arm; CRT- radiotherapy or chemoradiotherapy arm; SOC- standard of care arm; T- time point.
Quality of life was assessed using the EORTC QOL-C30 and H&N 35 questionnaires. Compliance for completion of the 65 questions on both questionnaires was reasonable. From baseline to week 6 of treatment, the median compliance was 67%, but in week 7, reflecting the peak of toxicity, compliance dropped to 36% but returned to 53% at months 1 and 3 follow-up. No statistically significant differences were noted between the treatment arms for any domain assessed by the two QoL questionnaires at any time point, using unadjusted and adjusted analyses which took into account the baseline value (data not shown), or using a mean 10-point difference to define clinically meaningful change. Figure 1 demonstrates the global health status on the EORTC QOL-C30 questionnaire demonstrating no difference in the mean QoL scores at any of the assessed time points. In week 5 of the radiotherapy or chemoradiotherapy, the mean reduction in global QoL score on the EORTC QOL-C30 was highest at 18.2 points. Following that, the mean scores improved until the 3-month follow-up, however, did not reach the baseline level at the last follow-up.
A summary of results including categorisation by 10-point reduction in scores, for both questionnaires is provided as supplementary information in Online Resource 2 (EORTC QLQ-C30 questionnaire) and Online Resource 3 (H&N 35 questionnaire). For some domains, at specific time points statistically significant difference was reached between arms on this assessment (e.g., EORTC QLQ-C30 pain domain - difference between week 6 and 7 and baseline categories, p < 0.05) of which the significant differences were not confirmed on the H&N 35 questionnaire (pain domain, p > 0.05), pain reporting using the visual analogue scale or by physician assessed toxicity grading or medication prescription. Overall, there were no consistent or significant differences in patient reported outcomes assessed by either questionnaire noted between the treatment arms using multiple methods of analyses.