Radio(chemo)therapy in locally advanced nonsmall cell lung cancer

Abstract

Definitive radiochemotherapy is the standard treatment for many patients with locally advanced nonsmall cell lung cancer (NSCLC). Treatment outcomes have improved over the last decades. Several treatment regimens have been shown effective and safe. This review summarises the results of significant studies between 1996 and 2015 on concomitant and sequential radiochemotherapy regimens and radiation dose per fraction. Beside therapy regimens, optimised radiotherapy planning is indispensable to improve outcome and minimise radiation-induced toxicity. An insight into the rationale of radiotherapy planning for stage III NSCLC is also provided.

Short abstract

Concomitant radiochemotherapy is an established standard treatment for locally advanced nonsmall cell lung cancer http://ow.ly/TTkkc

Introduction

Radio(chemo)therapy has been an important part of lung cancer treatment for several decades and its recent advances have led to significant improvements in treatment outcomes. The current review will focus on the application of radiotherapy for locally advanced nonsmall cell lung cancer (NSCLC).

Lung cancer is one of the most common causes of cancer death, with many patients diagnosed in a locally advanced stage. Although the treatment outcome has improved over the last decades, it remains poor [1].

Locally advanced lung cancer is a heterogeneous group and the definition often overlaps with stage III lung cancer. Typically patients with stage I/II lung cancer would be primarily considered for curatively intended surgery or stereotactic body radiotherapy in case of inoperability. Stage IV patients would typically receive palliative treatment [2], often involving systemic agents while nowadays multimodal concepts are being tested in oligo-metastatic cases.

Staging

An essential part of lung cancer staging is proper mediastinal lymph node staging, which may be achieved by fine needle aspiration (e.g. endobronchial ultrasound guided biopsies, transoesophageal biopsies), mediastinoscopy, 18-fluorodeoxyglucose positron emission tomography or a combination of these. When mediastinal lymph nodes are limited to the ipsilateral mediastinum, it is considered N2 (stage IIIA: T1–3 N2 or T4 N1) disease, while contralateral involvement denotes N3 disease (stage IIIB: T1–3 N3 or T4 N2). While stage IIIB N3 is generally not amenable to surgery [3], certain subsets of stage IIIA N2 disease are approachable with surgery. To help differentiate, the American College of Chest Physicians has published a sub-classification of N2 disease [4]. Typically bulky and/or multi-level N2 disease is associated with a high rate of micro-metastatic spread and extensive mediastinal involvement. In this setting as well as in stage IIIB, definitive radio-chemotherapy is the standard treatment.

Radiotherapy alone

Radiotherapy delivers ionising radiation, which in turn causes physical and chemical interactions leading to DNA and other tumour damage. Repetitive exposure to appropriate doses of radiation causes increased damage to tumour cells when compared to healthy tissue. This is the basis of conventionally fractionated radiotherapy, where typically a dose of 2 Gy is delivered five times per week.

Currently a daily fraction of 1.8–2 Gy is considered standard for lung cancer. Alternative dose-fractionation schedules have been investigated, especially for patients who are not suitable for concurrent chemotherapy. Several clinical trials in the last two decades show a survival benefit of hyperfractionated radiotherapy (twice or three times a day with doses <1.8 Gy) over conventional fractionation [5, 6]. However one of the latest studies, the randomised phase III CHARTWEL (Continuous Hyperfractionated Accelerated Radiotherapy Weekend-less) trial (ARO 97-1) [7], showed no difference in outcome between 60 Gy in 40 fractions over 2.5 weeks (CHARTWEL) or 66 Gy in 33 fractions over 6.5 weeks. However, in comparison to previous trials the rate of squamous histology in the CHARTWEL trial was lower, which is the subgroup that seems to benefit most from accelerated RT [8].

For patients with early stage NSCLC who are medically inoperable a good option is hypofractionated/stereotactic radiation therapy [9–12]. There is a clear benefit for a biological effective dose >100 Gy in NSCLC [13–15]. What about high-dose/fraction treatments for stage III NSCLC? There are some studies with dose-accelerated schedules, using doses between 45 and 60 Gy in 3–15 fractions with promising local control rates and acceptable toxicity [16–18]. While the results of ongoing randomised trials on hypofractionated radiotherapy have to be awaited, a recent meta-analysis has already confirmed the role of accelerated hyperfractionation resulting in an absolute benefit of 2.5% in overall survival (8.3–10.8%) at 5 years [19].

Radiochemotherapy

Radiotherapy alone has demonstrated reasonable response rates for locally advanced NSCLC, however the outcomes were very poor [20, 21]. The introduction of sequential radiochemotherapy has led to an increase of overall survival from approximately 5% to 10% at 5 years with the addition of chemotherapy [22–24]. With the implementation of concurrent radiochemotherapy this rate rose to 15% with an absolute survival benefit of 4.5% at 5 years [25].

Although concurrent radiochemotherapy is associated with improved overall survival when compared with sequential treatment [1, 26–30] (table 1), the latter is associated with lower toxicity (less oesophagitis/pneumonitis). Thus, concomitant treatment is the preferred strategy for fit patients but sequential chemo-radiotherapy may be applied in selected cases (e.g. the elderly or those with poor performance status) for whom a concomitant radiochemotherapy is not deemed feasible.

TABLE 1.

An overview of selected trials investigating sequential or concurrent radio-chemotherapy

First author [ref.]	Year	Trial details	Chemotherapy regimen	Radiotherapy total dose	Radiotherapy single dose	5-year OS
Bradley [1]	2015	Phase III	Carboplatin/paclitaxel (con) +/- cetuximab (cons)	60 Gy	2 Gy	57.6% (2-year OS)
			Carboplatin/paclitaxel (con) +/- cetuximab (cons)	74 Gy	2 Gy	44.6% (2-year OS)
Oh [31]	2013	Phase III	Cisplatin/paclitaxel (con/cons)	60–66 Gy	2–2.4 Gy	Median OS 27.3 months
			Cisplatin/docetaxel (con/cons)	60–66 Gy	2–2.4 Gy	Median OS 27.6 months
			Cisplatin/gemzitabine (con/cons)	60–66 Gy	2–2.4 Gy	Median OS 16.5 months
Curran [30]	2011	Phase III	Cisplatin/vinblastin (con)	60 Gy	2 Gy	16.0%
			Cisplatin/vinblastin (seq)	60 Gy	2 Gy	10.0%
			Cisplatin/etoposide (con)	69.6 Gy	1,2 Gy twice a day	13.0%
Segawa [32]	2010	Phase III	MVP (con)	60 Gy	2 Gy	16.6%
			Docetaxel/cisplatin (con)	60 Gy	2 Gy	17.5%
Yamamoto [33]	2010	Phase III	MVP (con)	60 Gy	2 Gy	17.5%
			Carboplatin/irinotecan (con)	60 Gy	2 Gy	17.5%
			Carboplatin/paclitaxel (con)	60 Gy	2 Gy	19.5%
Hanna [29]	2008	Phase III	Cisplatin/etoposide (con)	59.4 Gy	1.8 Gy	26.1% (3-year OS)
			Cisplatin/etoposide (con) + docetaxel (cons)	59.4 Gy	1.8 Gy	27.1% (3-year OS)
Gouda [28]	2006	Randomised	Paclitaxel/carboplatin (con)	60 Gy	2 Gy	45% (2-year OS)
			Paclitaxel/carboplatin (ind/con)	60 Gy	2 Gy	40% (2-year OS)
			No chemotherapy	60 Gy	2 Gy	10% (2-year OS)
Fournel [34]	2005	Phase III	Cisplatin/etoposide (con)	66 Gy	2 Gy	21% (4-year OS)
			Cisplatin/vinorelbine (seq)	66 Gy	2 Gy	14% (4-year OS)
Zatloukal [35]	2004	Randomised	Cisplatin/vinorelbine (con)	60 Gy	2 Gy	18.6% (3-year OS)
			Cisplatin/vinorelbine (seq)	60 Gy	2 Gy	9.5% (3-year OS)
Albain [36]	2002	Phase II	Cisplatin/etoposide (con)	60 Gy	2 Gy	15.0%
Furuse [37]	1999	Phase III	MVP (con)	56 Gy	2 Gy	15.8%
			MVP (seq)	56 Gy	2 Gy	9.0%
Lee [38]	1996	Phase II	Cisplatin/etoposide (con)	69.6 Gy	1,2 Gy twice a day	35% (2-year OS)
Jeremic [39]	1996	Randomised	Carboplatin/etoposide (con)	69.6 Gy	1,2 Gy twice a day	23% (4-year OS)
			No chemotherapy	69.6 Gy	1,2 Gy twice a day	9% (4-year OS)
Dillman [40]	1996	Phase III	Cisplatine/vinblastine (con)	60 Gy	2 Gy	17%
			No chemotherapy	60 Gy	2 Gy	7%

OS: overall survival; Con: concurrent; cons: consolidation; seq: sequential; MVP: mitomycin, vindesine, and cisplatin; ind: induction.

Several regimens have been established in combination with radiotherapy. Doublet chemotherapy results in better progression-free survival rates than concomitant single-agent chemotherapy [25]. Platinum-based chemotherapy regimens are a standard of treatment [27]. These include: cisplatin (Cis)/etoposide (Eto), Cis/vinorelbine, Cis/vindesine, Cis/mitomycine/vindesine, Cis/docetaxel, Cis/gemcitabine, carboplatin (Carbo)/paclitaxel, Carbo/irinotecan, Carbo/Eto [1, 28–40]. The 5-year overall survival for the mentioned regimens is around 15–20%. Patient comorbidities may help guide the choice of agent. While overall survival seems to be similar with these, less haematological toxicity but a higher risk of radiation pneumonitis was observed with Carbo/paclitaxel when compared to Cis/Eto [41]. Pemetrexed (Pem)/Cis seems to have a survival benefit for patients with metastatic non-squamous NSCLC [42]; however, no clear advantages are reported in locally advanced disease. In the PROCLAIM study presented at the American Society of Clinical Oncology Annual Meeting in 2015, concurrent Pem+Cis arm did not improve OS versus Cis/Eto, but did have a better safety profile [43].

Effective novel chemotherapy agents or targeted therapies, such as cetuximab, bevacizumab, gefitinib or anti-PD-1 inhibitors have so far led to rather disappointing results in combination with radiotherapy and may even bear the risk of unexpected toxicities like fistula, oesophagitis and bleeding [44–47]. So far, there is not enough evidence to support their routine use in radiochemotherapy protocols.

Planning

After obtaining imaging a planning computed tomography (CT), the typical procedure is to define the gross tumour volume (GTV), which represents the macroscopic tumour. Based on several investigations [48, 49], including histological evaluation of microscopic extensions the clinical target volume (CTV) is typically formed by adding 6–10 mm to the GTV in all directions (except where natural barriers would be expected, e.g. bone, pleura). Due to positioning and calculation inaccuracies, a further margin is added to create the planning target volume.

While imaging has generally improved and staging has become more precise, technological advancements have also helped in dealing with an obvious problem in radiotherapy for lung cancer: tumour movement. This has led the International Commission on Radiation Units to implement a new volume, internal target volume (ITV), which is used to account for tumour motion, primarily through breathing. Modern radiation oncology dedicated CT-scanners are capable of four-dimensional CTs, this means that imaging is acquired over different phases of the breathing cycle and the collected images are binned to create a “breathing” CT, which serves as a basis for the ITV. An alternative approach is to irradiate the CTV in a specific breathing phase (e.g. only inhalation or exhalation) using phased-gating or tracking, where small target volumes are irradiated and therefore an ITV is omitted.

Initially, mediastinal lymph nodes were irradiated electively, meaning they were irradiated beyond the involved lymph nodes. One of the first trials investigating this was performed by Yuan et al. [50]. This trial compared 74 Gy to only involved nodes (involved field irradiation) and compared it with 60 Gy to include elective nodal stations (elective nodal irradiation). Involved field irradiation was associated with lower rates of radiation-induced pneumonitis and improved local control when compared with prophylactic lymph node irradiation. Several other trials have confirmed this approach also demonstrating the rate of recurrence in elective nodal regions at <5% [51–53]. However, nearly all evidence on this topic derives from the three-dimensional conformal radiotherapy era with a relevant proportion of unaffected mediastinum being treated by incidental therapeutic doses. Except the single-centre prospective evidence from Yuan et al. [50], all data derive from retrospective series or secondary analyses. Despite this, the concept of involved node radiotherapy has become standard for locally advanced NSCLC radiotherapy to date. Recently, new data on a higher rate of out-of-field recurrences after more conformal planning and intensity-modulated radiotherapy [54] have arisen, so that further analyses and the result of running prospective trials may again change our perception here.

Dose escalation has been discussed and investigated for years. While several phase I–II trials have demonstrated improvement of local control, hard evidence for dose escalation beyond 60 Gy was lacking. The Radiation Therapy Oncology Group (RTOG) 0617 trial tested a dose escalation of 74 Gy versus 60 Gy. Surprising to many, the higher-dose arm was not associated with improved survival at 1 year but rather showed a contrary trend and the dose escalation arms were closed based on interim analysis [55]. Despite several doubts [56] about the causes involved in this result, the current standard of care is a radiation dose of 60–66 Gy and further dose escalation is still regarded as being experimental. This is also supported by the recent guidelines of the American Society for Therapeutic Radiation Oncology [57]. In many aspects, they are similar to the recommendations of the European Organisation for Research and Treatment of Cancer [58].

Thanks to modern treatment planning systems, it is possible to calculate the healthy tissue (organs at risk) doses in a respective radiotherapy plan. For decades, a consensus-based paper by Emami et al. [59] served as a pivot to predict tolerability of treatment. Due to an evidence-based pooled analysis of RTOG, the QUANTEC [60] project produced data that is currently used to predict the side effects of radiotherapy and the tolerability of evaluated treatment plans. Toxicities of radiotherapy typically include the organs that are irradiated due to their adjacency to the tumour. These include primarily organs of the thorax, specifically: the spinal cord [61], the lungs [62], the heart [63] and the oesophagus [64]. The typical constraints that are usually aimed for are: volume of organ receiving at least 20 Gy less than 30–35% and a mean lung dose of under 20–23 Gy. The volumes and doses in most analyses have considered the total lung volume, however individual (ipsi- and contralateral) lung doses have been also demonstrated to be predictive of radiation pneumonitis [5, 65]. When the appropriate criteria are taken into account the risk of side effects can be reduced and treatments become better tolerated.

Conclusion

Radiochemotherapy has an established role in the treatment of stage III lung cancer. Modern radiotherapy techniques may contribute to reducing toxicities. Concurrent chemotherapy is typically a platin-based doublet therapy. Whenever possible, concurrent radiochemotherapy is preferred, whereas for individual patients a sequential approach might be feasible.