Introduction
Early-stage non-small cell lung cancer (NSCLC) refers to stage I and stage II disease. 1 In recent years, there have been increasing numbers of patients diagnosed with early lung cancer compared to advanced disease. 2,3 This is likely to be due to improved public awareness, Reference Kennedy, Cheyne and Darby4 lung cancer screening Reference Bansal, Okoli, Cole and Jeannerat5 and the increase in incidentally found lung cancers due to increased utilisation of imaging in non-cancer specialities. Traditionally, surgery has been the standard treatment of early-stage NSCLC. A large percentage of lung cancer, however, is diagnosed in elderly patients with multiple comorbidities and poor lung function. Reference Froesch, Martucci, Györik, Dutly and Cafarotti6
Stereotactic ablative radiotherapy (SBRT) is associated with higher rates of tumour control and tolerability compared to conventionally fractionated radiotherapy and has become the standard of care for patients with inoperable early-stage peripheral lung cancer. 7,Reference Ball, Mai and Vinod8 SBRT is not routinely used for more central lesions due to safety concerns from earlier studies. Reference Timmerman, McGarry and Yiannoutsos9,Reference Fakiris, McGarry and Yiannoutsos10 A central lesion is defined as a lesion within 2 cm of the proximal bronchial tree. Reference Timmerman, Mcgarry and Yiannoutsos11 Patients with inoperable, centrally located lesions receive a radical dose of radiation therapy in either conventional or moderate hypofractionation of 20–30 daily fractions. Reference Rowell and Williams12 50 Gy in 15 daily fractions delivered over 3 weeks is a moderately hypofractionated radiotherapy regime that has been used in radiotherapy studies Reference Faria, Souhami and Portelance13 for early-stage NSCLC and shown good safety and efficacy profile. This study investigates the local control, survival outcomes and safety profile of this radiotherapy regimen at a large cancer centre in the UK.
Methods
Patients were eligible for the study if they were diagnosed with centrally located, non-metastatic NSCLC and received 50 Gy in 15 daily fractions of radiotherapy between 01 January 2014 and 30 December 2017 at the Leeds Cancer Centre. Cancer diagnoses were made on either histological or radiological grounds. Radiological diagnoses were based on positron emission tomography (PET) scan findings, and all cases had multidisciplinary team consensus.
All radiotherapy treatments were planned using 4D simulation computerised tomography (CT) scans. Radiotherapy volume definition includes gross tumour volume (GTV), which encompassed only the visible tumour as seen on simulation CT scan. Clinical target volume (CTV) was GTV plus a 5–7 mm margin in all directions. Planning target volume was CTV plus a 5 mm margin in all directions. Organs at risk (OARS) include the oesophagus, pericardium, brachial plexus, spinal cord and percentage of lung receiving more than 20 Gy (V20) as well as the mean lung dose (MLD). OARs doses are kept to a minimum, with threshold established by Radiation Therapy Oncology Group (RTOG). Reference Kong, Ritter and Quint14 Treatment verification was performed with using cone beam CT scans daily in the first week, then on alternate days for the remainder of treatment.
Electronic case notes and radiotherapy records of lung cancer patients were reviewed. Information extracted included patient and tumour demographics and Adult Comorbidity Evaluation-27 (ACE-27) score, which was used to evaluate comorbidities. Reference Piccirillo, Creech, Zequeira, Anderson and Johnston15 Post treatment radiation pneumonitis was recorded; it was graded using Common Toxicity Criteria version 5. 16 30- and 90-day post-treatment mortality was calculated. Overall survival (OS) was taken from date of diagnosis to date of death. Date of diagnosis is the date of histological confirmation for histologically diagnosed tumours or date of PET for radiologically diagnosed tumours. Progression-free survival (PFS) was taken from date of diagnosis to date of progression or death.
Statistics
Survival outcomes were estimated using Kaplan–Meier curves. Descriptive statistics was performed for categorical variables including sex, performance status, ACE-27 score, disease stage and histological subtype. Age is a continuous variable and is presented using median value and range. MLD is presented as a mean value with interquartile range. V20 is presented using a median value and interquartile range. Cox regression analysis was used to analyse factors which influenced survival outcomes. Statistical significance was set at p < 0.05. SPSS version 22 was used for statistical calculations.
Results
Fifty-three patients were identified and included in the study. Median follow-up was 20.2 months. Thirty-seven (70%) patients were female. The median age was 74.0 years, with a range of 69.5–79 years. 16 (30%), 33 (62%) and 4 (8%) patients had World Health Organisation performance status (WHO PS) of 1, 2 and 3, respectively. Thirty-seven (70%) patients had an ACE-27 score of 2 or above. Forty-six (87%) patients had stage I disease. Four (8%) of patients had stage II, lymph node negative disease. Thirty-seven (70%) patients had radiologically diagnosed disease. For patients with histologically confirmed disease, squamous cell carcinoma diagnosed in ten (19%) patients. A full summary of patient and tumour demographics is presented in Table 1.
Table 1. Patient and tumour demographics.
MLD was 7.3 Gy with an interquartile range of 5.8–8.8 Gy. Median V20 was 12.9% with an interquartile range of 9.5%–15%. Grade 2 pneumonitis was seen in five (9%) patients. No grade 3 or 4 pneumonitis was observed.
The median PFS and OS were 18.5 months (95% CI 12.2–24.8) and 28.2 months (95% CI 14.4–42.1), respectively, demonstrated in Graphs 1 and 2. The estimated 1- and 2-year PFS rates were 62.3% and 41.3%, respectively, and 1 and 2 years OS rates were 774% and 566%, respectively. There were no deaths within 30 days of treatment. Ninety-day mortality after treatment was 3.8% (two patients).
Graph 1. PFS (months).
Graph 2. OS (months).
WHO PS had an impact on OS on cox regression analysis with poor performance status associated with worse survival (Table 2). Nineteen (36%) patients had relapsed disease. 4 (7.5%) had infield recurrence, 10 (19%) had out-of-field lung recurrence and the rest had distant metastases.
Table 2. Cox regression analysis of variables on OS in the cohort
Discussion
The study population is typical of the patient group diagnosed with early-stage NSCLC in the UK, with respect to age, sex, comorbidities and performance status. 17 Majority of patients had radiologically diagnosed stage I disease. This reflects the difficulty of obtaining tissue biopsy in small central lung cancers Reference Kothary, Lock, Sze and Hofmann18 and the risk of complications Reference Heerink, de Bock and de Jonge19 in an elderly population with prevalent cardiopulmonary conditions. V20 and MLD measurements were low in the cohort, reflecting the small radiotherapy treating volume for early lung cancer.
In the current cohort of patients, 50 Gy in 15 fractions was associated with good local control more than 90% at a median follow-up of 20.2 months. Regional and distant relapse occurred in less than 30% of patients. This is consistent with the published literature of SBRT to peripheral lesions, Reference Prezzano, Ma, Hermann, Rivers, Gomez-Suescun and Singh20 conventionally fractionated radiotherapy to early lung cancer Reference Rowell and Williams12 and SBRT to central lesions in the recently published results from the RTOG 0813 study. Reference Bezjak, Paulus and Gaspar21 Survival times in the cohort appear in line with SBRT and appear superior to that reported in an interim analysis of a study of accelerated, hypofractionated radiotherapy for non-metastatic lung cancer. Reference Iyengar, Westover and Court22 WHO PS was strongly correlated with survival; this is unsurprising and confirms the importance of patient selection before treatment. There has been no large-scale, randomised studies comparing radiotherapy to surgery for early-stage NSCLC. Due to the difference in patient demographics and comorbidities, patients receiving radiotherapy are more likely to die from non-cancer causes. Reference Spencer, Kennedy and Lummis23
50 Gy in 15 fractions is associated with a good safety profile. There were no deaths within 30 days of treatment and two deaths within 90 days of treatment. No pneumonitis above grade 3 occurred. This is consistent with published toxicity profiles of SBRT. Reference Guckenberger, Allgäuer and Appold24
Patients with medically inoperable centrally located lung cancer have limited treatment options, as they are not routinely amenable to receive SBRT. 50 Gy in 15 fractions is associated with good tumour control with low rates of toxicity. In terms of biological effective dose (BED), using the time-adjusted BED formula outlined in the Machtay paper, Reference Machtay, Bae and Movsas25 the time-adjusted BED for 55 Gy in 20 fractions is 68.16 Gy, and for 50 Gy in 15 fractions is 65.63 Gy, alpha/beta ratio was presumed to be 10. The regime spans 3 weeks and is 1 week shorter than the widely used 55 Gy in 20 fractions schedule. This is more convenient for patients, places less demand from radiotherapy services and decreases patients’ contact with the hospital environment. Reducing exposure to clinical areas could reduce the risk of contracting the COVID virus, which in an elderly population with many comorbidities is associated with a high risk of death. Reference Rakhsha, Azghandi and Taghizadeh-Hesary26,Reference Faivre-Finn27
This study has several limitations. It was carried out in a single cancer centre, and retrospective data were used for analysis. The cohort of patients is relatively small, and a large proportion of patients had radiological diagnosis of cancer. Not all radiotherapy parameters, including doses to other OARS, were analysed. The key strengths of this study are long follow-up duration, detailed documentation of patient demographics and pattern of disease relapse.
Conclusion
50 Gy in 15 fractions radiotherapy for central early-stage lung cancer is a feasible choice that requires further randomised trials.
Acknowledgements
None.
Financial support
No financial support was received for this project.
