This quasi-experimental study was conducted at the Department of Radiation Oncology, National Institute of Cancer Research & Hospital (NIRCH), Dhaka, Bangladesh from January 2021 to December 2021. This study was approved by the Institutional Review Board (IRB) of the NIRCH, Dhaka, Bangladesh.

A total of 66 adult patients (age > 18 years) with histopathologically or cytologically diagnosed and radiologically proven locally advanced stage III NSCLC, who could not be treated surgically were included following purposive sampling technique.

For determination of sample size following formula was applied:

$n=\frac{P_1\left(1-P_1\right)+P_2\left(1-P_2\right)}{{\left(P_1-P_2\right)}^2}\times {\left(Z_{\alpha }+Z_{\beta }\right)}^2$

P₁ = Proportion of patients developed outcome in control arm = 45% [11] .

P₂ = Proportion of patients developed outcome in experimental arm = 70.6% [12] .

Z_α = Z-value (two tail) at a definite level of significance e.g., 1.96 at 5% level of significance.

Z_β = Z-value (one tail) at a definite power e.g., 0.85 at 80% power.

Conventional fractionated radiotherapy was expected to cause outcome 45% in unresectable locally advanced non-small cell lung cancer patients and 70% in hypofractionated arm. So, we calculated the sample size to examine which treatment is better at 5% level of significance with 95% confidence interval.

Here, P₁ = 45% (0.45), P₂ = 70.6% (0.7), Z_α = 1.96, Z_β = 0.85, n = Sample size

$n=\frac{\left(0.45\times 0.55\right)+\left(0.7\times 0.3\right)}{{\left(0.45-0.70\right)}^2}\times {\left(1.96+0.85\right)}^2=58.14\approx 59$

With 10% allowance for loss to follow up, final sample size = 59 + 5.9 = 64.9 ≈ 65.

Total of 66 patients included in this study, distributed in two arms (A and B), 33 patients in each arm.

Inclusion criteria were: patients with stage III NSCLC, age between 18 to 70 years with Eastern Co-operative Oncology Group (ECOG) performance status (PS) ≤ 2. Exclusion criteria were: patients having age less than 18 years or above 70 years, patients with histology other than squamous and adenocarcinoma of lung, patients with history of initial surgery (excluding diagnostic biopsy) of the primary site or radiotherapy to chest, patients with features of superior vena caval obstruction, patients having pleural or pericardial effusion, patients having concurrent diseases like- history of pulmonary fibrosis, myocardial infarction within last 12 months, grade II and above heart failure, uncontrolled heart failure, uncontrolled chronic obstructive pulmonary disease, impaired renal function (serum creatinine level > 1.4 mg/dl or calculated creatinine clearance rate < 60 ml/minute when needed) and abnormal liver function.

1) Case selection

Initially total 76 patients were enrolled. After eligibility assessment, 10 patients were excluded, and a total number of 66 patients were included in the study according to the selection criteria. After selecting the patient, informed written consent was taken from each patient. Then a complete medical history, relevant clinical examination findings and necessary investigations reports were documented in a case record form. Location and size of the tumor was recorded prior to treatment. All patients were divided into two arms: Arm A for conventional fractionated radiotherapy and Arm B for hypofractionated radiotherapy. Patients for Arm A and Arm B were allocated by using systematic randomization, assigning odd numbers to Arm A and even numbers to Arm B.

2) Pretreatment evaluation

A thorough history had been taken from the patient or attendant to obtain maximum possible information including demographic characteristics, Eastern Co-operative Oncology Group (ECOG) performance status and details of his/her illness. The mode of onset, progression, duration of chief complaints had been noted.

3) Investigations

Laboratory investigations

Imaging

Special investigation: Histopathology by biopsy or fine needle aspiration cytology (FNAC) report of the lesion.

4) Intervention

Medical and supportive care

Patients were managed symptomatically with antibiotics, steroid, analgesics, antiemetic and conservative management was given according to need throughout the treatment period.

Chemotherapy protocol

A total of 66 patients were included in this study and were equally sub-grouped into two Arms, Arm A and Arm B. All received 4 cycles of chemotherapy with Inj. Paclitaxel 175 mg/m² in 500 ml normal saline within 3 hours at Day-1 and Inj. Carboplatin (AUC = 6) in 500 ml D/A within 1 hours at Day-1 [13] .

Proper hydration was maintained. Patients were monitored for any kind of hypersensitivity reactions. Pre and post - chemotherapy medication with anti-emetics, and other necessary drugs were given before and after chemotherapy.

Also, for Paclitaxel, patients received premedication for prevention of the incidence of hypersensitivity reactions i.e., Dexamethasone 20 mg PO at 12 and 6 hours before and Diphenhydramine 50 mg at 30 minutes before drug administration [13] . Patients were assessed before each cycle by clinical symptoms analysis, physical examination and hematological parameters.

After completion of 4 cycle sequential chemotherapy, all patients underwent evaluation by clinical symptoms analysis, physical examination and with a contrast enhanced computed tomography (CECT). The response criteria were based on cross-sectional diameter and tumor response, nodal response, and overall response according to response evaluation criteria in solid tumors (RECIST) response criteria [14] .

If there was no progression of disease after 3 weeks of completing 4^th cycle of chemotherapy, all patients proceeded for radiotherapy.

Radiotherapy

For both arms

Radiotherapy had been delivered to the patients using

Arm A

Patients of Arm A was treated with conventional fractionated radiotherapy, 60 Gy in 30 fractions, 2 Gy per fraction, 5 days in a week over 6 weeks.

Arm B

Patients of Arm B was treated with hypofractionated radiotherapy, 55 Gy in 20 fractions, 2.75 Gy per fraction, 5 days in a week over 4 weeks.

Dose constraints:

5) Patient assessment

Assessment during treatment

During Chemotherapy:

During Radiotherapy:

Assessment after treatment

After collection, data were checked and verified. Then, all data were tabulated in a master data sheet. Thereafter, data were entered into computer and coded. Data categorization and summarization were done. Continuous data were expressed as mean ± standard deviation (SD), whereas categorical data were expressed with rate and ratio. Statistical analysis was done according to the study’s objective by using Statistical Package for Social Sciences (SPSS) software version 25.0 for windows. The analysis was done using Unpaired t-test for continuous variables and Chi-square test (χ²)/Fisher’s exact tests for categorical variables. All reported p-values were two-sided, and a p value less than 0.05 was taken as statistically significant. The present study was designed to assess and compare the short-term response outcomes of conventional versus hypofractionated radiotherapy in locally advanced non-small cell lung cancer. As such, patient survival data beyond 6 months were not collected, and therefore reliable Kaplan-Meier analyses of progression-free survival (PFS) and overall survival (OS) could not be performed.

Analyzing the baseline characteristics of the study patients showed that, most of the patients belonged to age group 56 - 65 years. The mean age was 56.03 ± 5.8 years in arm A and that was 56.61 ± 5.8 years in arm B. Minimum age was 45 years in Arm A, it was 44 years in Arm B. Maximum age was 66 years in Arm A, which 65 years in Arm B. The difference was not statistically significant (p > 0.05) between two arms ( Table 1 ).

In Arm A, male and female patients were 26 (78.8%) and 7 (21.2%) respectively, and the ratio was 3.7:1. In Arm B, male and female patients were 27 (81.8%) and 6 (18.2%) respectively, and the ratio was 4.5:1. The difference was not statistically

COPD = Chronic obstructive pulmonary disease; SOB = Shortness of breath; p-values obtained from: a = Unpaired t-test, b = Fisher’s exact test, c = Chi-square test; *Multiple responses.

significant between the two arms (p > 0.05) ( Table 1 ). We evaluate the distribution of study patients according to their exposure to some risk factors. Smoker was quite common in both Arms; 26 (78.8%) in Arm A and 23 (69.7%) in Arm B ( Table 1 ). In this study, most patients had baseline ECOG performance status score 1 in each of the study Arms. ECOG performance status was not significantly different between the study Arms (p = 0.484) ( Table 1 ).

Among the study patients, most common presentation in both Arms was cough (78.8% and 87.9%). Weight loss was also common in both groups; 23 (69.7%) patients in Arm A and 22 (66.7%) in Arm B. Chest pain (40.9%) and shortness of breath (40.9%) were equally common presenting features in both Arms. There was no significant difference between the two Arms in context of clinical presentation (p > 0.05) ( Table 1 ).

In this study, majority of the study patients had right lung lesion, which was 66.7% in Arm A and 57.6% in Arm B. The difference was not statistically significant between the Arms (p = 0.447) ( Table 1 ).

Among the study patients, squamous cell carcinoma was found 18 (54.5%) patients in Arm A and 16 (48.5%) patients in Arm B. Adenocarcinoma was found in 15 (45.5%) patients and 17 (51.5%) patients in Arm A and Arm B respectively. The difference was not statistically significant between the groups (p > 0.05) ( Table 2 ). The staging (TNM stage) of the study patients at the time of presentation in both arms was evaluated. In Arm A: 14 (42.4%) patients were Stage III A, 18 (54.5%) patients were Stage III B and 1 (3%) patient was stage IIIC; in Arm B: 16 (48.5%) patients were Stage III A, 17 (51.5%) patients were in Stage IIIB. The distribution of the sample was homogenous in both the Arms ( Table 2 ).

p-values obtained from Chi-square test, ns = not significant.

After completion of 4 cycles of chemotherapy: Patients showing partial responses (PR) were 27 and 29 in Arm A and Arm B respectively; while, stable disease (SD) was observed in 6 and 4 patients in the two arms respectively. The difference was not statistically significant (p = 0.492). No complete response (CR) and progressive disease (PD) were found in both arms ( Table 3 ).

After 6 weeks: In Arm A, 1 (3%) patient showed complete response (CR) whereas in Arm B, CR was observed in 3 (9.1%) patients. At that time, 27 (81.8%) patients in Arm A and 29 (87.9%) patients in Arm B showing partial response (PR); stable disease (SD) was found among 5 (15.2%) patients and 1 (3%) patient in the two Arms respectively. Although Arm B showed better response mathematically, the difference was not statistically significant (p = 0.154) ( Table 3 ).

After 12 weeks: 3 (9.1%) patients in Arm A and 5 (15.2%) patients in Arm B showed complete response (CR). At this time, most patients in Arm A [21 (63.6%)] and Arm B [23 (69.7%)] maintained partial response (PR); stable disease (SD) was observed among 6 (18.2%) patients and 3 (9.1%) patients in the two Arms respectively. While, 3 (9.1%) patients from Arm A and 2 (6.1%) patients from Arm B showed progressive disease (PD). The difference was not statistically significant (p = 0.617) ( Table 3 ).

p-values obtained from Chi-square test, ns = not significant.

After 24 weeks: In Arm A, 3 (9.1%) patients showed complete response (CR) and in Arm B, CR was observed in 5 (15.2%) patients. Partial response (PR) was found among 20 (60.6%) patients and 21 (63.6%) patients in Arm A and Arm B respectively. Stable disease (SD) was found among 4 (12.1%) patients in Arm A and 2 (6.1%) patients in Arm B. There were 6 (18.2%) patients having progressive disease (PD) in Arm A and 5 (15.2%) patients had PD in Arm B. At that time, treatment response was not statistically significant between the Arms (p = 0.733) ( Table 3 ).

In this study, according to ECOG performance status (PS); ECOG PS-0 was 13, ECOG PS-1 was 39 and ECOG PS-2 was 14. Complete response and partial response (CR + PR) were seen in 11 patients with ECOG PS-0 and 38 patients with ECOG PS-1; but only partial response (PR) was seen in 1 patient with ECOG PS-2. No treatment response was observed in 2 patients with ECOG PS-0, one (1) patient with ECOG PS-1 and 13 patients with ECOG PS-2. This result reflected that, patients of ECOG PS-0 and patients having ECOG PS-1 were associated with good response rate. The difference was statistically significant (p < 0.001) ( Table 4 ).

In terms of primary tumor response of TNM stages: stage IIIA had significant higher treatment response than other two sub-stages (p = 0.001) ( Table 4 ).

We found that there was significant association between histological type and treatment response of primary tumor (p = 0.015) ( Table 4 ).

It was observed that treatment response was better in non-smoker group but

p-values obtained from: a = Chi-square test, b = Fisher’s exact test.

no significant association between smoking habit and treatment response of primary tumor (p > 0.05) ( Table 4 ).

Most observed toxicity in both Arm A and Arm B (during and after 6 weeks and 12 weeks of completion of radiotherapy) were as follows:

Esophagitis was frequently observed during radiotherapy and after treatment in both arms. In Arm A, 15 patients developed grade 1 esophagitis and 10 patients developed grade 2 esophagitis respectively. In Arm B, 17 patients developed grade 1 esophagitis, 12 patients developed grade 2 esophagitis respectively. This difference was not statistically significant (p > 0.05) ( Table 5 ).

p-values obtained from: a = Chi-square test, b = Fisher’s exact test.

Radiation pneumonitis was observed mostly after treatment completion in both arms. In Arm A, 9 patients developed grade 1 and 9 patients developed grade 2 pneumonitis. In Arm B, 12 and 11 patients developed grade 1 and grade 2 pneumonitis respectively. This difference was not statistically significant (p > 0.05) ( Table 5 ).

In Arm A, 6 (18.1%) patients and 2 (6.1%) patients developed grade 1 and grade 2 dermatitis respectively. In Arm B, 8 (24.2%) patients and 2 (6.1%) patients developed grade 1 and grade 2 dermatitis respectively. The difference was not statistically significant (p > 0.05) ( Table 5 ).

Nausea and vomiting were observed during treatment. But there was no statistically significant difference was found between the groups (p > 0.05) ( Table 5 ).

In Arm A, 15 patients and in Arm B, 17 patients had grade 1 anemia. Regarding neutropenia and thrombocytopenia, no significant difference was found between the groups ( Table 5 ).

No patients from either group experienced grade 3 toxicity.

The multivariate logistic regression analysis assessed independent predictors of treatment response. ECOG performance status was not significantly associated with response (OR 0.600, 95% CI: 0.123 - 2.940, p = 0.529). Stage IIIA showed a lower odd of response (OR = 0.140, 95% CI: 0.026 - 0.756), but this did not reach statistical significance (p = 0.221). Patients treated with hypofractionated sequential chemoradiotherapy had significantly higher odds of response compared to those treated with conventional fractionated chemoradiotherapy (OR = 3.692, 95% CI: 1.027 - 13.268, p = 0.045). Smoking status and histology were not included in the multivariate model ( Table 6 ).