Rationale and impact
- Diagnosis and staging
- Brain imaging for people having treatment with curative intent
- Surgery and radiotherapy with curative intent for non-small-cell lung cancer
- Management of operable stage IIIA–N2 non-small-cell lung cancer
- First-line treatment for limited-stage disease small-cell lung cancer
- Thoracic radiotherapy and prophylactic cranial irradiation in small-cell lung cancer
These sections briefly explain why the committee made the recommendations and how they might affect services. They link to details of the evidence and a full description of the committee's discussion.
Clinical audit is an important tool for maintaining high standards in the use of endobronchial ultrasound‑guided transbronchial needle aspiration (EBUS-TBNA) and endoscopic ultrasound‑guided fine‑needle aspiration (EUS-FNA). This is consistent with the British Thoracic Society guideline and quality standards (which are accredited by NICE).
The recommendations cover:
initial invasive investigations for people with an intermediate probability of mediastinal malignancy
subsequent investigations for people with a high probability of mediastinal malignancy, when neck ultrasound and biopsy are negative.
In these circumstances, when compared with alternative investigations, EBUS-TBNA and EUS-FNA:
produce a diagnosis and stage faster than bronchoscopy or CT-guided biopsy
are more acceptable to patients than surgery
reduce the need for further investigations and hospital visits compared with bronchoscopy.
There is evidence that surgical staging is useful when EBUS-TBNA and/or EUS-FNA are negative but clinical suspicion of mediastinal malignancy is high. While there are potential harms from the invasive nature of surgical staging, there is no evidence that these outweigh the benefits in this population.
Transthoracic needle biopsy, bronchoscopy and non-ultrasound-guided TBNA are no longer recommended for staging lung cancer in intrathoracic lymph nodes because:
bronchoscopy and non-ultrasound-guided TBNA are unlikely to reach the minimum sensitivity required by the British Thoracic quality standards and
they may discourage people from having more effective procedures (such as EBUS-TBNA) and subsequent investigations.
The word 'fibreoptic' has been removed because bronchoscopy can be fibreoptic, video or hybrid.
The recommendations on PET-CT reflect current practice, so will not incur an extra cost.
The recommendations on EBUS-TBNA and EUS-FNA will reinforce best practice and result in a more streamlined diagnostic service with more timely diagnosis and staging.
The surgical mediastinal staging recommendation will also reinforce best practice and restrict this procedure to people most likely to benefit.
Brain imaging is helpful before starting treatment with curative intent, because if brain metastases are detected then the treatment plan is likely to change. However, routine brain imaging is expensive, and the evidence showed that it does not always offer a good balance of benefits and costs.
In people with stage II and IIIA disease, the benefits of brain imaging outweigh the costs because:
brain metastases are more common than in stage I disease
people can start early treatment for metastases if they are identified, which improves prognosis
some people with brain metastases will not have radical treatment (depending on factors such as the number of metastases, prognosis and patient preference), and this reduces costs.
In people with clinical stage I NSCLC and no neurological symptoms, the prevalence of detectable brain metastases is fairly low (around 4%) compared with people with stage II or IIIA disease. People with stage I NSCLC who do have brain metastases often still have radical lung treatment, which is much more rarely the case for people with stage IIIA NSCLC. Overall, the lower prevalence of metastases and smaller reduction in numbers of people having radical treatment mean that the benefits of brain imaging in this population are too low to justify the costs.
The 2018 review only examined the clinical and cost effectiveness of imaging after the treatment plan has been decided, but the committee noted that it could be more efficient to conduct CT brain imaging alongside initial staging CT. With this in mind, the committee made a research recommendation on routine brain imaging with CT at initial diagnosis and/or staging.
Practice in this area is variable. The committee estimated that the recommendations will increase the number of people who have brain imaging. In turn, they thought this should prevent the use of treatment options (such as lobectomy and sublobar resection) in some patients for whom it is not expected to be beneficial. The recommendations may also lead to an increase in radical radiotherapy, stereotactic radiosurgery and brain surgery. These treatments would be expected to improve the person's prognosis, although each treatment would carry its own risks and side effects.
For people with non-small-cell lung cancer (NSCLC) who are well enough and for whom treatment with curative intent is suitable, the evidence showed that lobectomy provides better survival outcomes than stereotactic ablative radiotherapy (SABR). Lobectomy is a good compromise between preserving pulmonary function and being more likely to remove cancerous cells compared with sublobar resection.
For people with stage I–IIA (T1a–T2b, N0, M0) NSCLC, the evidence showed that:
if they decline lobectomy or it is contraindicated, sublobar resection and SABR both provide better survival outcomes than conventionally fractionated radiotherapy, although it is not clear which of these 2 is better
if they decline any surgery or it is contraindicated, SABR provides better survival outcomes than conventionally fractionated radiotherapy, and people often prefer it because it involves fewer hospital visits
if surgery and SABR are contraindicated, conventionally fractionated radiotherapy provides better survival outcomes than no radiotherapy.
For people with stage IIIA or IIIB NSCLC who cannot tolerate chemoradiotherapy or who decline it, the evidence was not strong enough to recommend conventional radiotherapy over hyper-fractionated regimens or vice versa. However, people who cannot tolerate chemoradiotherapy may also be unable to tolerate radical radiotherapy, so this will not be an option for everyone with stage IIIA or IIIB NSCLC. For an explanation of the recommendations covering surgery in this group, see the rationale on management of stage IIIA-N2 NSCLC.
55 Gy in 20 fractions is the most common conventional radical radiotherapy regimen in the UK. If conventionally fractionated radiotherapy is used, a total radiation dose of 60 Gy provides better survival outcomes and fewer adverse events than 74 Gy. A total dose of 60 to 66 Gy is also normal NHS practice.
There are not many randomised controlled trials comparing SABR with surgery (lobectomy or sublobar resection). SABR is non-invasive, so if it is as effective as surgery then it may be a preferable option for many people with lung cancer. There are also various factors that may make SABR less costly than surgery. For example, it is usually delivered as outpatient treatment. There might also be subgroups for whom different forms of surgery or SABR might be the most cost-effective options. The committee made a research recommendation on SABR compared with surgery to investigate these uncertainties.
The new recommendations on SABR are a change from the 2011 guideline and improve choice for people with NSCLC. However, practice has also changed since 2011, and SABR is now widely used, so implementing the recommendations may not involve a significant change in practice. The remaining changes to the recommendations reflect current practice.
The available evidence showed that chemoradiotherapy and surgery are more effective than chemoradiotherapy alone in people who are well enough for surgery and when the disease is operable. For chemotherapy and surgery, there was no evidence that survival outcomes were better than for chemoradiotherapy, so the additional costs of including surgery outweighed the benefits.
The key benefit associated with chemoradiotherapy and surgery is the longer progression‑free survival time. An analysis of multiple trials showed improved progression-free survival and cost effectiveness for chemoradiotherapy with surgery, compared with chemoradiotherapy alone. There was an 89% probability that chemoradiotherapy and surgery improved average overall survival time compared with chemoradiotherapy. However, the evidence in favour of chemoradiotherapy and surgery involved indirect comparisons, and no head-to-head trials showed meaningful differences in overall survival for any of the interventions. And as with any major surgery, there is a perioperative mortality risk for people who have chemoradiotherapy and surgery.
The 3 to 5 week wait for surgery is recommended to give people time to recover from the chemoradiotherapy.
Chemoradiotherapy with surgery is not often offered in current practice. In addition, there are specific factors to take into account when offering all these treatments together. Therefore, multidisciplinary teams providing it should have expertise both in the combined therapy, and in all the individual components.
Immunotherapy has been shown to be effective in a variety of NSCLC indications but there is currently no evidence on whether it is clinically or cost effective for people with stage IIIA-N2 NSCLC following surgery. The committee made a research recommendation on immunotherapy after multimodality treatment to address this.
The committee felt that chemoradiotherapy and surgery is offered far less often than chemoradiotherapy alone or chemotherapy and surgery for people with NSCLC stage IIIA-N2. Therefore, these recommendations could lead to a change in current practice.
The evidence showed a survival benefit from twice-daily radiotherapy compared with once-daily. However, the committee agreed that some people with small-cell lung cancer will not be well enough to tolerate twice-daily radiotherapy, so they recommended giving people the option of once-daily radiotherapy.
The committee noted that, in practice, radiotherapy is not started in chemotherapy cycle 1, because this is when planning for the radiotherapy often takes place (see the recommendation on twice-daily radiotherapy with concurrent chemotherapy in the section on first-line treatment for limited-stage disease small-cell lung cancer). However, there was no new evidence on when to start radiotherapy, so the 2019 recommendation on this is the same as the original 2011 recommendation.
There were limited data available on whether continuous radiotherapy with concurrent chemotherapy was more effective than alternating radiotherapy and chemotherapy. Because of the limited data, and the committee's experience that people prefer to complete treatment as quickly as possible, the 2019 recommendation on concurrent therapy (see the recommendation on twice-daily radiotherapy with concurrent chemotherapy) is the same as the 2011 recommendation.
There was some uncertainty in the evidence. However, the study most relevant to UK practice showed that thoracic radiotherapy improves long-term survival for people who have had a partial or complete response to chemotherapy, if they live longer than 1 year after the radiotherapy. The committee specified that thoracic radiotherapy should be given alongside prophylactic cranial irradiation. This is to match recommendation 1.4.62. In addition, the reviewed clinical trials gave thoracic radiotherapy alongside prophylactic cranial irradiation.
The evidence showed that prophylactic cranial irradiation improves survival versus best supportive care.
Prophylactic cranial irradiation can adversely affect quality of life, and the survival benefits are limited. There is also some evidence from a study outside the UK that routine MRI follow-up may be more cost effective. The committee made a research recommendation on prophylactic cranial irradiation compared with routine MRI follow-up in extensive-stage SCLC, to provide evidence more relevant to the UK and to see if MRI could identify people who need whole-brain radiotherapy and so reduce the number of people having unnecessary treatment.
The 2011 recommendation only recommended thoracic radiotherapy for people with a complete response to chemotherapy at distant sites. Therefore, this recommendation could increase the number of people who are given thoracic radiotherapy.
It is likely that the recommendation reflects current clinical practice.