Background. Inflammation, a common result of anticancer therapies, correlates with poor survival and cardiovascular issues in cancer survivors. High levels of C-reactive protein (CRP), D-dimer, and serum amyloid A (SAA) indicate cancer recurrence, mortality, and cardiotoxicity. CRP-lowering clinical trials have proven that the magnitude of the reduction in cardiac events in patients with residual inflammatory risk is directly proportional to the achieved reduction in CRP, with the greatest benefit in those achieving a threshold below than 2 milligrams per litre (mg/L).
Cancer survivors may benefit from non-pharmacological anti-inflammatory treatments that do not exacerbate cardiotoxicity burden. Evidence suggests the air purifiers lower inflammatory biomarkers in high-risk cardiovascular groups. Air filtration presents a promising alternative to inflammation-inhibiting drugs, but its anti-inflammatory and cardioprotective effects in cancer survivors and interactions with medications remain unclear.
Design, setting and participants. A series of N-of-1 randomised, adaptive, blinded, placebo-controlled trials will be conducted at the home sitting of community-dwelling cancer survivors who are registered in primary care practices of the city of Valencia, Spain. Participants eligible are aged 18 years or older, have a history of breast, colorectal, prostate, lung and haematologic cancer, completed curative cancer treatment with evidence of remission and screened blood CRP level of at least 3 mg/L (within individual SD CRP of \~0.20 mg/L at baseline is accepted).
Screening for CRP levels. Up to two home visits could be carried out to collect a 5 microlitres (µl) blood microsample and analyse the CRP levels in real-time of the potentially eligible participants. These screening visits will be done no more than 28 and 7 days prior to blinded N-of-1 phase, respectively.
Recruitment. Participants will be identified via 3 strategies: advertisements on social media, snowball sampling and in-primary care clinics referrals. Staff from local primary care setting participating in the study will identify eligible potentially participant. The selection criteria for primary care settings were based on neighborhood air quality, specifically urban areas with the highest index impact of pollutant on population \> 125 and NO2 levels exceeding WHO and EU Directive 2008/50/EC limits: Malilla, Russafa, and Arrancapins.
Intervention. Each participant will be randomly allocated (1:1 ratio per cycle) to 3 treatment sets, each comprising a 14-day period of active therapy (portable air filtration unit \[PAFU\] at 275 m3/h) and a 14-day period of placebo (unit with sham filters - standard of care) in the bedroom. Active therapy and placebo will be nightly administered during at least 7-hours consecutive within a time window between 10:00 p.m. and 10:00 a.m. The blinded N-of-1 trial enrolment will be between 4 and 12 weeks per participant, depending on the treatment set(s) received to produce evidence of clinical benefit (CRP \< 2 mg/L or CRP reductions ≥ 35%) .
After 2-weeks wash-out period of the last blinded treatment set, participants who do not achieve a clinically meaningful change will be given an open-label phase to take 14-days of no treatment and 14-days of nightly and daily active therapy - the PAFU will be continuously operated (filtered air at 275 m3/h) in bedrooms with the door and windows closed. The nightly active therapy keeps the same as the blinded treatment sets and participants will be asked to remain in the bedrooms for as long as possible (e.g., nap time). Time exposed to the in-bedroom air quality will be recorded.
Outcomes measures. The primary outcome is the level change in CRP after air filtration treatment compared with placebo. The secondary outcomes are changes in D-dimer, SAA and HbA1c concentrations, and systolic and diastolic BP. Primary and secondary outcomes will be collected at baseline and in the day 7 and 14 of each period using at-home minimally invasive medical devices.
Self-reported questionnaires. Four short versions of self-reported questionnaires will be weekly completed as potential inflammation modifying confounding factors: Godin Leisure-Time Exercise Questionnaire (GLTEQ), Food Frequency Questionnaire (FFQ), Pittsburg Sleep Quality Index (PSQI) and in-home Patterns \& Time-Activity Questionnaire (PTAQ).
In-bedroom and outdoor air quality measurement. The changes in airborne pollutants will be continuously measured in real-time throughout the trial period using two monitors that integrates a combination of sensors, including particulate matter.
Statistical plan. Following treatment set 1 and 2, interim analysis will be conducted to establish whether the active treatment, as compared with placebo, is clinically effective: achievement of CRP levels \< 2 mg/L or CRP reductions ≥ 35%. While participants who achieve this clinically meaningful change will be discontinued of the N-of-1 trial, those who do not achieve it in the treatment set 3 will be given the open-label phase.
Changes from baseline data for treatment and placebo per cycle will be determined. The means and 95% confidence intervals (CIs) for the differences will be calculated within treatment pairs for each outcome. The N-of-1 trials will be combined using a hierarchical Bayesian effects model to estimate the average effect that incorporate variance within- and between-individuals.
Sample size. A sample size of 8 participants will provide a power of 80% to detect a minimal clinically important difference (MCID) of change in CRP (≥ 35% reductions in this trial). A maximum of 10 participants will be enrolled (dropout rate of 20%). The estimation was based on the Yang et al. equations and implemented into an R Shiny app \[https://jiabeiyang.shinyapps.io/SampleSizeNof1/\].
