Hypofractionated Radiotherapy for Locally or Systemically Advanced Non-small Cell Lung Cancer

Background: Palliative thoracic radiotherapy (RT) can improve local control and survival in patients with unresectable locally or systemically advanced non-small cell lung cancer (NSCLC), but the optimal RT dose has not been well-dened. We investigated the survival outcomes of patients with NSCLC who underwent hypofractionated radiotherapy (HFRT). Methods: We retrospectively investigated survival and adverse effects among 74 patients with locally or systemically advanced NSCLC who received HFRT (45 Gy/15 fractions) at our institution. Results: The median overall survival (OS) was 18.7 months, with 1- and 2-year OS rates of 65.9% and 33.9%, respectively. The median local progression-free survival (LPFS) was 7.2 months, with 1- and 2-year LPFS rates of 27.9% and 9.4%, respectively. Sixteen patients (21.6%) developed grade ≤ 2 pneumonitis and 14 (19%) developed grade ≤ 2 esophagitis; no grades ≥ 3 pneumonitis or esophagitis occurred. Conclusions: HFRT is safe, tolerable, and effective for patients with unresectable locally or systemically advanced NSCLC exhibiting poor prognostic factors. modulated

Background Non-small cell lung cancer (NSCLC) accounts for over 80% of all lung cancers; more than 50% of patients have locally or systemically advanced disease [1,2]. Concurrent chemoradiotherapy(CCRT) is the recommended treatment for unresectable locally advanced NSCLC [3]. The conventional dose fractionation for radiation administered with concurrent chemotherapy is 60 Gy over 6 weeks [4,5].
However, in practice, a considerable number of patients are ineligible for curative chemoradiotherapy (CRT) owing to poor performance status, inadequate pulmonary function, advanced age, multiple medical comorbidities, or large tumors. Previous studies have con rmed that palliative thoracic radiotherapy (RT) could improve local control and survival in patients with poor prognostic factors, recurrent tumors, and non-response to chemotherapy, as well as in those who are candidates for potentially toxic drugs [6,7].
However, currently available data do not support the routine use of chemotherapy with palliative RT for these patients. Although the American Society of Radiological Oncology (ASTRO) published a clinical practice guideline in 2011 and updated it in 2018 [8,9], the optimal RT dose has not been well-de ned. A study performed at the MD Anderson Cancer Center found no statistically signi cant differences in local control and overall survival (OS) among patients with locally advanced NSCLC treated with hypofractionated radiotherapy (HFRT) (45 Gy at 3 Gy/fraction) and conventional fractionation radiotherapy (CFRT) (60-66 Gy at 2 Gy/fraction) [10]. However, this study mainly focused on patients with stage III disease including a small number of those with stage II, and treatments administered post RT were also unknown. On the other hand, studies have shown that RT produces synergistic effects with targeted therapy [11], antiangiogenic therapy [12] and immunotherapy [13]; however, there remain too few studies on the use of these modalities as consolidation therapy after palliative RT for patients with locally or systemically advanced NSCLC. As such, we performed this study to investigate the effect of hypofractionated HFRT in such patients.

Patient characteristics
We reviewed the records of 4174 patients with lung cancer treated with RT at Tianjin Medical University Cancer Institute and Hospital between January 2014 and December 2019. We identi ed 74 patients with unresectable locally or systemically advanced NSCLC who received HFRT (45 Gy/15 fractions). All patients had histologically con rmed disease. Staging was based on the American Joint Committee Classi cation (AJCC) 7th edition criteria. Pretreatment evaluation consisted of computed tomography (CT) or 18-uorodeoxyglucose positron-emission tomography (PET/CT). All patients completed their prescribed treatments. This study was approved by the internal review board of our institution. Treatment CT or PET/CT was used for RT planning. The gross tumor volume (GTV) included the primary tumor and the clinically positive lymph nodes (short axis > 1cm and/or positive at PET/CT). RT was administered 5 days a week in 15 fractions of 3 Gy/each for 3 weeks. The calculated biologically effective dose (BED) was 58.5 Gy (α/β = 10). RT was delivered using a linear accelerator with 6 MV photon beams with intensity-modulated radiotherapy (IMRT) or volumetric modulated arc therapy (VMAT).

Follow-up and statistics
All patients returned for a follow-up (which included CT and PET/CT as necessary) 1-3 months after the completion of RT, every 3-6 months thereafter. The therapy response was assessed as complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) according to the Response Evaluation Criteria in Solid Tumors (RECIST) guideline version 1.1 [14].Toxicities were evaluated within 3 months from RT using the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Local progression-free survival (LPFS) was de ned as the time from RT to relapse inside or within a 1 cm margin around the PGTV on follow-up imaging. OS was de ned as the time from RT to death from any cause or to the last follow-up.
Survival was estimated using the Kaplan-Meier method, and used the log-rank test for univariate analysis of LPFS and OS. Multivariate analysis was carried out using the Cox regression method. Variables found to be signi cant on univariate analysis (p ≤ 0.1) were subjected to multivariate analysis; a p-value < 0.05 was considered statistically signi cant. SPSS version 24 (IBM Corp., Armonk, NY, USA) and RStudio version 1.2 (RStudio Inc., Boston, MA, USA) were used for Statistical analyses.

Discussion
We found HFRT to be effective; the median OS was 18.7 months (with 1-and 2-year OS rates of 65.9% and 33.9%, respectively) while the median LPFS was 7.2 months (with 1-and 2-year LPFS rates of 27.9% and 9.4%, respectively). The adverse effects were tolerated given that no grade ≥ 3 toxicities occurred; this was also the case in the study by Amini et al. [15].
Studies at the MD Anderson Cancer Center used HFRT (45 Gy/15 fractions) to treat patients with unresectable NSCLC who could not tolerate the CFRT; no signi cant differences in local recurrence and OS were observed between the 2 regimens. The 1-and 2-year OS rates of patients who received HFRT were 53% and 12%, respectively [10,15,16]. In a study by Abratt et al., the median OS of patients with stage III NSCLC who received palliative RT (45 Gy/15 fractions) was 8.5 months [17]. The patients in our study had better survival than did those in previous studies, which might be due in part to the innovations in cancer therapeutics (including targeted therapy, antiangiogenic therapy, immunotherapy, combination therapy, and technological improvements in other modalities).
A previous analysis demonstrated that an absolute OS bene t was observed that ranged from 0.36-0.7% for every 1 Gy increase in BED [16]. Machtay et al. also found that an increase of 1 Gy BED could improve survival by approximately 4%, along with greater acute toxicity [18]. A meta-analysis by Fairchild et al.
found that patients received higher dose schedules (BED ≥ 35 Gy, α/β = 10) achieved a signi cantly better survival than those received lower dose schedules, with 2-year OS rates of 26.5% vs. 21.7% for wellperforming patients, albeit at the cost of greater esophageal toxicity. While there was no advantage in administering high dose schedules for poorly performing patients [19]. However, a higher total dose appeared not to bene t palliative patients with lung cancer (50 Gy/20 fractions vs. 25 Gy/5 fractions) [20].
The traditional approach to increase BED is to increase the number of fractions. CFRT is usually not welltolerated among patients with poor performance or advanced disease. Hypofractionation, on the other hand, enables the delivery of higher BED, as HFRT involves signi cantly fewer fractions with higher doses per fraction compared to standard modalities. Large fractions are an effective way of for ablating tumors, likely owing in part to the reduced rate of tumor cell repopulation; and may be more convenient and cheaper than conventional fractionation [21,22]. Compared with other studies of palliative RT regimens with lower BED, patients in our study achieved higher survival bene ts by increasing the dose per fraction to obtain higher a BED (58.5 Gy, α/β = 10) while adverse events remained tolerable.
Many studies comparing the e cacy of HFRT and stereotactic body radiation therapy (SBRT) for the palliative treatment of patients with advanced lung cancer showed that HFRT appeared to provide a greater survival bene t. Lewis et al. investigated the effects of fraction schedules (30 Gy/10 fractions, 20 Gy/5 fractions, and 10 Gy/1 fraction) on the survival of lung cancers treated with palliative RT and found that increasing the total dose was associated with better survival regardless of performance status [23]. Moreover, SBRT is generally not suitable for tumors that are central (i.e., close to the trachea or heart), those that are accompanied by necrosis or mediastinal lymph node metastases, or those that are large in size.
In the past decade, the use of molecularly targeted therapy has led to signi cantly longer survival in patients with EGFR or ALK sensitizing mutations [24]. However, previous studies focused mainly on concurrent administration of these therapies with RT as rst-line treatment for patients with advanced disease, which maybe place them at a higher risk of adverse effects. In our study, 9 patients who were treated with TKIs after RT achieved signi cantly longer LPFS and OS with no obvious adverse effects observed. Okamoto et al. found that 2 of 9 patients with unresectable stage II NSCLC exhibiting EGFR mutations who were treated with ge tinib and concurrent RT survived for more than 5 years, although the study was stopped owing to a high incidence of adverse events [25]. In the RECEL trial that compared erlotinib versus etoposide/cisplatin with concurrent RT for patients with unresectable stage III NSCLC carrying EGFR mutations, the erlotinib/RT arm had signi cantly improved PFS compared to the CRT with the same incidence of adverse effects [26]. The SINDAS study showed that rst-line TKI with upfront SBRT to all sites signi cantly improved both PFS and OS compared with TKI alone in previously untreated patients with oligometastatic NSCLC (≤ 5 lesions) carrying EGFR mutations [27].
Many studies have shown that antiangiogenic agents could produce signi cant survival bene ts for patients with advanced NSCLC. However, previous efforts have focused more on concurrent antiangiogenic agents with RT, and few investigators have used such agents as maintenance therapy post-RT. The ECOG 4599 study showed that the addition of bevacizumab to paclitaxel/carboplatin for treating recurrent or advanced NSCLC produced a signi cant survival bene t [28]. The ALTER0303 trial showed that anlotinib administered as the third-or-greater-line treatment prolonged PFS and OS in patients with advanced NSCLC [29]. Moreover, patients in the HELPER study who received continuous infusion of endostar combined with CCRT for unresectable stage III NSCLC achieved favorable PFS and OS with tolerable toxicities [30]. In our current study, no survival bene t was found among 13 patients (17.6%) who were treated with antiangiogenetic therapy after RT; this may partly be due to the most of these patients (10, or 80%) having stage IV disease while 5 patients (38%) were still alive as of the last follow-up.
The recent development of immune checkpoint inhibitors (ICIs) has essentially changed the treatment pro les of patients with advanced NSCLC [31]. A network meta-analysis by Almutairi et al. that compared the e cacy and safety of PD-1/PD-L1 inhibitors in patients who were previously treated for advanced NSCLC found signi cant bene ts of pembrolizumab and nivolumab in terms of PFS, OS, and in pairwise comparisons to docetaxel [32]. Another meta-analysis by Schulz et al. found that PD-L1/PD-1 inhibitors as second-or-later-line treatments for advanced NSCLC demonstrated greater survival bene ts, producing the highest expected 5-year OS rates compared to other treatments [33]. PACIFIC study showed that, compared to placebo, durvalumab demonstrated signi cant PFS improvements for patients with unresectable stage III NSCLC without progression after CRT, the 3-year OS was 66.3% versus 43.5% [34,35]. In our study, 5 patients (17.6%) received immunotherapy after RT among whom 1 was lost to followup and 1 was still alive after 10 months (with 5.1 months of LPFS) as of the last follow-up; the LPFS times for the other 3 patients were 6.1, 21.1, and 6.7 months, respectively, while their OS times were 23.1, 29.6 and 12.8 months, respectively. Although the sample size was small, these data preliminarily suggested that ICIs after HFRT can provide a promising survival bene t.
Previous studies have shown that the combination of RT and either TKIs, antiangiogenetic therapy, or ICIs could improve survival in patients with locally or systemically advanced NSCLC, usually at the cost of a high incidence of toxicity, although the optimal RT dose remains uncertain. In contrast, we found that patients who received TKIs, antiangiogenetic therapy, or ICIs as maintenance therapy after HFRT achieved improved survival with no obvious adverse effects.
Recent studies also found that the NLR, well-known as a marker of host in ammation, may be a useful predictor of poor prognosis in patients with NSCLC. Scilla et al. found that baseline NLR was a signi cant prognostic indicator in patients with locally advanced NSCLC who received de nitive CRT with or without surgery [36]. A retrospective multicenter study indicated that patients with NSCLC who have high pretreatment NLR (i.e., > 5) may experience inferior PFS and OS treated with nivolumab [37]. A metaanalysis by Zhang et al. also showed that a high NLR was associated with poor PFS and OS in patients with NSCLC treated with ICIs [38]. Similar results were obtained in our study, in which patients with low NLRs had better prognoses.
Our study's nding that patients with larger tumor sizes and greater tumor-to-lung volume ratios had poorer prognoses were similar to the conclusions of Werner-Wasik et al. [39] and Bradley et al. [40]. Strøm et al. found that patients with NSCLC tumors > 7 cm appeared to bene t notably from palliative CRT, but with an increased risk of radiation-induced esophagitis and pneumonia [41]. As such, it is possible that a larger tumor volume might not be a limitation for RT, as the incidence of adverse effects could be reduced by administering HFRT alone or sequential CRT.
Moreover, we found that patients who underwent ≥ 2 previous chemotherapy lines may experience inferior survival; as such, we posit that administering RT earlier may improve the survival of these patients.
In conclusion, our ndings suggest that HFRT is safe and effective for treating patients with unresectable locally or systemically advanced who have poor prognostic factors. However, this study was limited by its retrospective design as well as its relatively short follow-up time. Further investigations using novel hypofractionated regimens combined with TKIs, antiangiogenic therapy, or immunotherapy are warranted.