User:Emma.dickins/Sandbox

Please note there will be individual wiki page reference lists when the content becomes live.

Content updates by author Tracy Smith

What is the role of advance care planning and timing of referral for patients with lung cancer?

Introduction

Palliative care is appropriate for all people facing life threatening disease, though in practice in Australia, most services are directed toward people with life limiting/terminal disease. While most studies reviewed to create the palliative care section of this guideline are derived from studies relating to patients with NSCLC, it is likely that the themes and concepts are broadly applicable to those with SCLC.

Metastatic lung cancer is a leading cause of death. In 2009 2014, it was the leading cause of cancer deaths in Australian men and women and was responsible for the deaths of 7,786 8,251 Australians, making it the fourth leading cause of death. The prognosis after the diagnosis of stage IV (metastatic) lung cancer has been estimated to be less than one year. Due to the high mortality rate, the rapidity of disease progression which is sometimes seen, as well as late presentation and co-morbidities, questions regarding timing of referral to palliative care along with questions regarding advance care planning are highly relevant to this patient group.

Back to top

Timing of referral

There has been debate over the optimal time of referral to palliative care for patients diagnosed with advanced cancer, including patients with stage IV lung cancer. Temel et al conducted a randomised controlled trial (RCT) to answer this question comparing early referral to palliative care (within eight weeks of diagnosis of metastatic NSCLC) to usual care (including referral to palliative care when requested by patient, family or treating oncologist). Patients who were referred early to palliative care had better quality of life (QOL) assessed 12 weeks after referral, and the improvement in their QOL was both statistically and clinically significant. In addition, patients referred to palliative care early were more likely to have their wishes with respect to resuscitation documented, had less depression and were less likely to receive aggressive care at the end of life. Furthermore, those who were referred early lived an average of 2.7 months longer than those who received usual care. A survival benefit for patients with lung cancer referred to palliative care earlier was also found in a study which extracted data from a large US database and found longer survival for patients referred to hospice than those not referred.

In contrast a RCT conducted in Australia found a different result. This study included patients who had a range of different advanced malignancies (including lung cancer) and compared the effect on QOL and mortality of early referral to a Palliative Care nurse to usual care. In the early referral group, there was a trend toward decreased quality of life and a statistically significant decrease in mortality. Patients in this study received a relatively ‘low dose’ of palliative care involvement compared with the study by Temel et al quoted above. Additionally, despite randomisation, there were some important differences between the groups at baseline which may explain some of the results. The authors suggest that the change in QOL may partially be explained by patients in the intervention arm being more comfortable disclosing symptoms and/or those in the control arm maintaining denial as a coping mechanism for longer.

Overall, the preponderance of evidence remains in favour of early referral to palliative care, but other factors, including the ‘dose’ of palliative care may be important.

Back to top

Advance care planning

Advance care planning (ACP) is a patient centred process in which patients, in consultation with family members and health care providers, make decisions regarding their future health care known should they later become incapable of expressing such preferences. The process of advance care planning usually occurs over a series of conversations, rather than being a single 'one off' event. Reviewing patients priorities and preferences as their illness progresses is almost always appropriate. Given the poor prognosis of stage IV inoperable lung cancer, such discussions would appear to be highly relevant. This type of discussion requires effective communications skills, and guidance regarding these skills may be found in some excellent Australian clinical practice guidelines.

Back to top

Efficacy and acceptability of advance care planning

One randomised controlled trial has assessed the efficacy and acceptability of ACP in Australian populations. This study randomised elderly, hospitalised inpatients with non malignant disease to an ACP intervention facilitated by a specially trained health care professional. It found patients who completed ACP were more likely to have their preferences known and respected at end of life. Importantly, patients who were randomised to the ACP arm of this trial reported higher levels of satisfaction with care for the inpatient episode in which the ACP occurred. In addition, lower levels of distress and depression were reported by family members of patients who died after completing ACP than those who did not participate in ACP. Qualitative research where patients were interviewed regarding their experience of undertaking ACP discussions confirms that the process of ACP is acceptable to patients.

Back to top

Need for advance care planning

Good health care can only be achieved when clinicians and patients share a common understanding of the patients' illness, prognosis and preferences. Information regarding prognosis is highly valued by patients and their families, however, with respect to lung cancer, clinicians may not adequately communicate prognosis. Patients with other solid malignancies tend to overestimate their prognosis, however this issue has not been addressed in patients with lung cancer. In addition, patients often overestimate the probability of success of aggressive interventions like CPR and may not understand the role of such interventions in the context of their advanced cancer. Accurate prognostic understanding may be associated with a decreased desire for CPR and/or intensive care admission and an increased preference for hospice care.

A minority of patients with advanced lung cancer have had discussions regarding resuscitation status with their clinicians. A substantial proportion of those who have not discussed their preferences would like to do so, but cite lack of initiation of these discussions by health care practitioners as a major barrier, and may rely on the responsible physician and/or the health care system to initiate such discussions. When patients have not completed an advance care plan their next of kin may not fully understand their wishes. In contrast, when advance care directives have been completed, bereaved family members report it was helpful in guiding care prior to death. In addition, there is a mismatch between doctors' perceptions of patients' preferences with respect to CPR, as well as other life sustaining interventions, and their patients’ actual preferences. Finally, there is evidence to suggest that referral to palliative care at the time of diagnosis increases the likelihood of documenting patients' wishes with respect to resuscitation and avoiding aggressive care at end of life.

Back to top

Process of advance care planning

Initiating and facilitating discussions with patients and their families is a complex communication task for which some excellent Australian clinical practice guidelines exist. It is important to emphasise that ACP is a process rather than a single conversation, and the involvement of loved ones is almost always appropriate. Importantly, patients and their families may not always recognise when the topic of end of life or advance care planning has been raised, so it may be necessary to sensitively check the patients’ and their families understanding the issue in subsequent conversations.

A systematic review examining patient preferences with respect to end of life discussions found that patients and their care givers want honest information delivered with sensitivity and hope, but without jargon by a trusted health professional. They value compassion, care and empathy in the delivery of this type of information. However, the information needs of patients and their care givers may vary over time, with patients preferring less information as death approaches and care givers generally wanting more information as time proceeds desiring more, reinforcing the need to regularly ask patients and loved ones if they would like to re-explore these issues. Patients may change their mind about how much information they want about their illness over time, with some wanting more and some wanting less information as their illness unfolds. In addition, patients preferences regarding who should control decision making (ie doctor controlled, shared or patient controlled) may fluctuate over time. These changes may be unpredictable and thus repeatedly exploring these preferences may be appropriate.

Patients may need time to process the idea that end of life is approaching and doctors may delay discussions until there is definite evidence of medical deterioration, by which time, end of life may be quite close. Furthermore, In addition, there are cultural factors which may impact on information needs and some patients with lung cancer experience stigma which complicate communication and information needs.

There is marked variation between individuals in terms of their information needs and decision making preferences. These need to be respected for successful ACP. A Belgian study reported that patients with advanced lung cancer had diverse opinions regarding what they wanted to know, who should be involved in discussions and whom they thought should be involved in making decisions at end of life. Some patients prefer to almost unilaterally make their own decisions while others prefer medical staff to have ultimate decision making authority. In addition, a patients religious or spiritual beliefs may influence their desire to participate in ACP, with one study suggesting that those who were more reliant on spiritual coping were less likely to engage in ACP discussions. Taken together, this evidence suggests the importance of exploring patient understanding and preferences as well as the importance communication skills which use obtaining permission before delving further into these areas.

In terms of the tools to assist ACP, an Australian randomised controlled trial found that the provision of a prompt list to patients increased both the number of questions asked and the number of topics covered in prognostic and end of life discussions, suggesting a role for similar tools in ACP. As mentioned above, patients tend to overestimate the likelihood of success of life sustaining interventions like CPR. Relatively simple educational tools which describe the success rate of CPR via the use of a written scenario have been shown to be effective in reducing patient preference for this type of intervention. When decisions to limit life prolonging interventions are made, it is important to then emphasise that care will continue to be provided to ensure that the person is supported throughout the course of their illness the care that will be provided to the patient in their routine care as well as during the process of dying. As with all medical care, it is important to document the outcomes of these types of discussions.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
The evidence suggests that referral to palliative care at the time of diagnosis of metastatic lung cancer is associated with better outcomes in terms of quality of life, survival and aggressiveness of care at the end of life, however the amount of contact from the palliative care service may be important.	II	,

Evidence-based recommendation	Grade
It is recommended to refer patients with stage IV inoperable NSCLC to palliative care at the time of diagnosis of metastatic disease.	B

Evidence summary	Level	References
The evidence suggests that advance care planning is effective and acceptable to Australian populations. The benefits of completing ACP include higher rates of preferences known and respected at end of life, higher patient and family satisfaction and lower rates of family distress and depression.	II

Evidence-based recommendation	Grade
Advance care planning discussions should be initiated with patients, as there are multiple benefits.	B

Evidence summary	Level	References
The evidence suggests that there is a need for patient centred advance care planning to address the gaps between clinician and patient expectation and to better understand patients' preferences.	II, IV	, , , , , , ,

Evidence-based recommendation	Grade
Clinicians may explore patients’ understanding of their health situation and offer to provide further information about their prognosis and to explore the patients’ goals/priorities/fears and concerns about the future.	C

Evidence summary	Level	References
The evidence suggests that in the area of prognostic and end of life communication, the provision of a prompt list increases the number of questions asked and the number of topics discussed.	II

Evidence-based recommendation	Grade
It is recommended to offer the opportunity to discuss advance care plans with patients and their care givers using communication strategies tailored to individual needs, with the assistance of simple tools, like a prompt list, where appropriate.	B

Practice point
If appropriate or necessary to discuss specific treatment options (e.g. CPR), first check the person’s understanding of the likely outcomes then provide clarification as needed. Consider referral to palliative care when metastatic disease is diagnosed. Don’t wait until there is definite evidence of medical deterioration.

Practice point

If the decision is made to limit life prolonging treatment (e.g. CPR/ ICU level care etc), emphasise other care and support that will be given to patients, such as other appropriate medical care, symptom directed care and care during the dying process. It may take some time for patients and their families to comprehend and process advance care planning discussions. Discussing patients understanding of their disease and/ or prognosis along with their hopes and fears may enable important conversations. It is never ‘too early’ to explore these concerns.

Back to top

Content updates by author Michael Brown

What is the clinical benefit of neoadjuvant chemotherapy for patients with stage III operable NSCLC?

Introduction

Defining operable and inoperable disease in stage III

The management of Stage III NSCLC has been divided into sections dependent on whether the disease is considered operable or inoperable at the time of diagnosis.

Read full explanation

Stage III NSCLC encompasses a broad spectrum of disease extent from tumour involving a single nodal station identified only postoperatively despite extensive pre-operative staging to involvement of multiple contralateral mediastinal nodes and supraclavicular nodes appreciated on clinical examination. In patients with clinically equivocal involvement, pathological confirmation of nodal status should be made if it will influence management options.

The decision as to operability should be made in a multidisciplinary setting.

Patients with Stage III NSCLC may be deemed inoperable because of patient factors (the patient’s respiratory function or co-morbidities may preclude operative intervention or the patient may choose not to proceed with surgery) or tumour factors (the extent or location of gross disease might make surgical resection technically impossible, for example left sided tumours with mediastinal nodes to the right of the aorta, N3 nodal involvement and most T4 tumours).

In the absence of other factors precluding surgery, patients with N1 disease should be considered for surgery. Patients with confirmed N2 disease should not be treated by surgery as the sole modality, but resectable cases may be considered for a multimodality approach. There is no consensus on the distinction between resectable and unresectable N2 disease. Factors influencing assessment of resectability include nodal size, number of stations involved, extracapsular extension and involvement of the recurrent laryngeal nerve.

The clinical rationale for neoadjuvant chemotherapy includes:

(i) availability of clinical and/or pathological assessment of treatment response to indicate the likelihood of systemic disease control

(ii) tumour regression to improve the chances of successful tumour resection, and

(iii) increased chemotherapy delivery rate as chemotherapy would be better tolerated before rather than after major surgery.

Conversely, major disadvantages of neoadjuvant chemotherapy are:

(i) chemotherapy-induced accelerated tumour cell repopulation, and

(ii) delay in performing a potentially curative operation thus risking metastatic spread if the chemotherapy is ineffective.

Back to top

Neoadjuvant chemotherapy and surgery versus surgery alone

A recent high-quality, individual patient data-based meta-analysis of 2,385 patients in 15 trials compared chemotherapy and subsequent surgery with surgery alone. The median follow-up period was 6 years for all patients. The patients were mostly men (80%) with a median age of 62 years (IQR 55-68) and good performance status (88%). Patients had predominantly squamous (50%) rather than adenocarcinoma histology (29%). Most patients had clinical stage IB-IIIA NSCLC (93%), and 22.2% patients had stage III disease.

The primary objective of this analysis was overall survival. A clear survival benefit of neoadjuvant chemotherapy was observed (HR 0.87, 95% CI 0.78-0.96; p=0.007). This benefit represented a 13% reduction in the relative risk of death, and translated to a 5% absolute improvement in survival at 5 years from 40% to 45% for all patients. As listed below, the effect of neoadjuvant chemotherapy on survival occurred irrespective of a large number of clinical factors. Consequently, this overall HR of 0.87 was applied to the survival of control group patients, and thus survival at 5 years for stage III patients receiving neoadjuvant chemotherapy improved from 20% to 25%. However, 98% of the stage III patients were stage IIIA, and comprised 21.6% of the total number of patients. Therefore, study results for stage III NSCLC can only be confidently applied to stage IIIA patients.

List of clinical factors that did not influence the effect of neoadjuvant chemotherapy on survival:

(i) age;

(ii) age group (<60, 60-64, 65-69, ≥70);

(iii) histology (adenocarcinoma or squamous);

(iv) performance status (0, 1, 2+);

(v) whether the chemotherapy was given pre-operatively, or both pre-operatively and post-operatively;

(vi) the number of pre-operative chemotherapy cycles (2 or 3);

(vii) the type of chemotherapy regimen (platinum plus second generation chemotherapy, platinum plus third generation chemotherapy, or non-platinum chemotherapy);

(viii) the number of chemotherapy agents (non-platinum single agent regimen [i.e. docetaxel], doublet regimen, or triplet regimen);

(ix) the chemotherapy regimen and the number of chemotherapy agents (non-platinum single agent regimen, platinum second generation [doublet], or platinum second generation [triplet]);

(x) whether the regimen was cisplatin-based or carboplatin-based;

(xi) whether post-operative radiotherapy given or not.

Deferring surgery because of neoadjuvant chemotherapy did not appear to produce an excess of early mortality. This meta-analysis did not identify deleterious effects of neoadjuvant chemotherapy on mortality survival within either 30 days of surgery or within 6 months of randomisation (OR 0.88, 95% CI 0.67-1.14, p=0.33; heterogeneity p=0.60). Administering neoadjuvant chemotherapy may have a practical advantage over adjuvant chemotherapy. In 10 of 15 trials in this meta-analysis, the mean compliance rate for neoadjuvant chemotherapy was 85% (range 71-100%), which contrasts with the lower mean compliance rate of 62% (range 41-98%) for adjuvant chemotherapy.

No evidence was found, for or against, to indicate that neoadjuvant chemotherapy improved the likelihood of complete resection by making tumours more operable. Indeed, an effect of neoadjuvant chemotherapy on complete resection rates could not be reliably estimated either because of possible variations in the classification of extent of complete resection or heterogeneity of the effect between trials, especially as the complete resection rate for control patients varied considerably. Furthermore, there was no clear effect of neoadjuvant chemotherapy on time to locoregional recurrence (HR 0.88, 95% CI 0.73–1.07; p=0.20; heterogeneity p=0.89).

As secondary outcomes, neoadjuvant chemotherapy conferred a clear benefit both on recurrence-free survival (HR 0.85, 95% CI 0.76–0.94, p=0.002; heterogeneity p=0.41) and time to distant recurrence (HR 0.69, 95% CI 0.58–0.82; p<0.001; heterogeneity p=0.40). The recurrence-free survival at 5 years improved from 30% to 36%. The time to distant recurrence at 5 years improved from 60% to 70%. This 10% absolute benefit of neoadjuvant chemotherapy on distant recurrence rate at 5 years was greater than the 5% absolute benefit at 5 years for adjuvant chemotherapy, and suggests that neoadjuvant chemotherapy may have greater potential to eradicate micrometastases. Moreover, there was a difference in effect by chemotherapy scheduling (p=0.05) for time to distant recurrence. A substantially greater relative benefit existed for those responding patients who also received adjuvant chemotherapy (HR 0.53, 95% CI 0.39–0.73, p<0.001) than for those who received neoadjuvant chemotherapy alone (HR 0.78, 95% CI 0.63–0.96, p=0.02).

The value of adding neoadjuvant radiotherapy to neoadjuvant chemotherapy and surgery was investigated recently using a randomised trial design. In a study of 232 patients with pathologically proven stage IIIA/N2 NSCLC, Pless et al (2015) allocated 117 patients to receive chemoradiotherapy and 115 patients to receive chemotherapy. The primary endpoint was event-free survival. There was no significant difference in the median event-free survival between the two groups, and median overall survival also did not differ significantly. Hence, radiotherapy did not add any benefit to induction chemotherapy followed by surgery. The authors suggested that one definitive local treatment modality combined with neoadjuvant chemotherapy is adequate to treat resectable stage IIIA/N2 NSCLC.

In a recent systematic review and meta-analysis, Xu et al (2015) aimed to determine (i) the survival benefit of multimodality therapy including surgery to stage IIIA/N2 NSCLC patients and (ii) if neoadjuvant chemoradiotherapy was superior to neoadjuvant chemotherapy in stage IIIA/N2 NSCLC patients. Seven trials involving 1049 patients were included in this study. There was no significant difference in OS or PFS in stage IIIA/N2 NSCLC patients who received neoadjuvant chemotherapy or chemoradiotherapy before surgery compared to those who received neoadjuvant chemotherapy or chemoradiotherapy before radical radiotherapy. These data lend further support to the notion that one definitive local treatment modality combined with neoadjuvant chemotherapy is adequate to treat resectable stage IIIA/N2 NSCLC.

Until this the recent report of the NSCLC Meta-analysis Collaborative Group, which was based on individual patient data, reliable evidence had not been available to show a consistently beneficial survival effect of neoadjuvant chemotherapy in stage III NSCLC.

The earlier systematic review and literature-based meta-analysis had been based on a relatively small number of trials and patients: seven randomised controlled trials and 988 patients. Neoadjuvant chemotherapy was found to increase survival with a HR of 0.82 (95% CI, 0.69 - 0.97; p=0.022). There was no evidence of statistical heterogeneity, and when this HR was applied across all stages of disease, it gave an equivalent absolute survival benefit of 6%, increasing overall survival from 14% to 20% at 5 years. However, this analysis had been unable to establish if stage III NSCLC patients may benefit more or less from neoadjuvant chemotherapy. An indirect comparison meta-analysis had been performed to obtain the relative hazards on survival of postoperative to preoperative chemotherapy administration in patients with resectable NSCLC. Stage III patients were not analysed separately and in the final analysis, there were no evident differences in overall and disease-free survival in the timing of chemotherapy administration. A subgroup meta-analysis was performed within a literature-based meta-analysis to understand the possible survival benefits of neoadjuvant chemotherapy in stage III NSCLC patients. The quality of evidence in this study was low with a substantial risk of bias.

Finally, other recent but low-quality and biased studies do not weigh against the strength of evidence provided by the NSCLC Meta-analysis Collaborative Group.

Katakami et al (2012) performed a randomised controlled trial of neoadjuvant chemotherapy versus neoadjuvant chemoradiotherapy in 60 patients with stage IIIA-pN2 NSCLC by. This study was too small because slow accrual had led to early closure. No statistically significant difference between treatment arms was found. Chen et al (2013) reported on a study of 356 NSCLC patients (including 122 patients with stage IIIA disease) who were randomised to surgery alone or neoadjuvant chemotherapy then surgery. No effect of neoadjuvant chemotherapy was detected in patients with stage IIIA disease. Unfortunately, critical data were missing in this study, preventing a reliable assessment of outcomes to be made. Horita et al (2013) performed a meta-analysis of aggregated patient data from 16 randomised controlled trials of a total of 2,385 patients, which included 1,447 stage IIIA patients. A meta-analysis was also performed for the seven studies that evaluated only patients with stage III NSCLC. For 1,447 patients and 1,068 deaths, the pooled HR for OS was 0.77 (95% CI, 0.68-0.87; p < 0.001). In this group of stage III disease studies, the heterogeneity was low and was not significant (I² = 17%; p for x² = 0.300 [<0.01]). However, the heterogeneity between the pooled HR for studies only with stage III patients and those without stage limitation was high and significant (I² = 69%; p for x² = 0.073 [<0.01]). In addition, in the remaining group of nine studies without stage limitation, moderate heterogeneity was found, and was significant (I² = 47%; p for x² = 0.059 [<0.01]). Furthermore, in this meta-analysis evaluating only patients with stage III NSCLC, three of the seven studies contributed the majority both of patients (1,217) and deaths (903), but these studies were of poor quality using the Chalmers score.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
In patients with clinical stage IIIA disease treated by surgery, neoadjuvant chemotherapy reduces the relative risk of death by 13%, and improves absolute 5 year survival rates from 20 to 25%.	I

Evidence-based recommendation	Grade
It is recommended to consider pre-operative administration of 2-3 cycles of platinum doublet-based, third-generation chemotherapy as a treatment option in good performance status patients with operable clinical stage IIIA non-small cell lung cancer.	A

Practice point
Patients whose tumours respond to preoperative chemotherapy may derive additional survival benefit from postoperative chemotherapy.

Practice point

Patients with resectable stage III non-small cell lung cancer, who are being considered for preoperative chemotherapy and surgery or surgery and postoperative chemotherapy, should have their treatment plan reviewed in a lung cancer-specific multidisciplinary meeting. The recommended treatment plan may need to be individualized to take account of such patient-specific factors as treatment preference, availability and timing of surgery, and geographically remote location.

Neoadjuvant and adjuvant chemotherapy and the MDT

Back to top

What is the clinical benefit of adjuvant chemotherapy for patients with stage III operable NSCLC?

Introduction

Defining operable and inoperable disease in stage III

The management of Stage III NSCLC has been divided into sections dependent on whether the disease is considered operable or inoperable at the time of diagnosis.

Read full explanation

Stage III NSCLC encompasses a broad spectrum of disease extent from tumour involving a single nodal station identified only postoperatively despite extensive pre-operative staging to involvement of multiple contralateral mediastinal nodes and supraclavicular nodes appreciated on clinical examination. In patients with clinically equivocal involvement, pathological confirmation of nodal status should be made if it will influence management options.

The decision as to operability should be made in a multidisciplinary setting.

Patients with Stage III NSCLC may be deemed inoperable because of patient factors (the patient’s respiratory function or co-morbidities may preclude operative intervention or the patient may choose not to proceed with surgery) or tumour factors (the extent or location of gross disease might make surgical resection technically impossible, for example left sided tumours with mediastinal nodes to the right of the aorta, N3 nodal involvement and most T4 tumours).

In the absence of other factors precluding surgery, patients with N1 disease should be considered for surgery. Patients with confirmed N2 disease should not be treated by surgery as the sole modality, but resectable cases may be considered for a multimodality approach. There is no consensus on the distinction between resectable and unresectable N2 disease. Factors influencing assessment of resectability include nodal size, number of stations involved, extracapsular extension and involvement of the recurrent laryngeal nerve.

Operable stage III non-small cell lung cancer (NSCLC) has a poor prognosis although considerable heterogeneity in the T and N classifications of stage III disease results in variable five year survival rates e.g. 15-35% and 5-10% for stages IIIA and IIIB disease, respectively. Operability may be determined on medical or surgical grounds or for reasons of patient preference. The true extent of heterogeneity of stage IIIA disease can further refined by sub-classifying N2 involvement. Stage IIIB NSCLC with N3 disease is considered inoperable although surgery may be indicated for carefully selected patients with T4N0-1M0 disease. The prognosis of stage III NSCLC is poor because the risk of death originates from either locoregional recurrence or distant recurrence. Therefore, treatment strategies in operable stage III disease have been designed to counter both types of relapse. In addition to radiotherapy, chemotherapy may be given preoperatively as neoadjuvant or primary chemotherapy or postoperatively as adjuvant chemotherapy, or as both.

As has been shown convincingly for breast and colorectal cancers, the overall rationale for post-operative chemotherapy for resectable or resected node-positive NSCLC is the delivery of a systemic modality of anti-cancer treatment that may reduce risk of death from micrometastatic disease. Nonetheless, cytotoxic chemotherapy is a significant cause of morbidity and mortality that may militate against its beneficial effects.

The rationale for adjuvant chemotherapy is accurate knowledge of pathological stage and prognosis, and its major disadvantage is the lower rate of delivery in the post-operative setting.

Back to top

Surgery and adjuvant chemotherapy versus surgery alone

The 1995 Non-Small Cell Lung Cancer Collaborative Group meta-analysis based on individual patient data indicated that cisplatin-based chemotherapy after surgery might increase survival (hazard ratio [HR] 0.87, 95% CI 0.74 - 1.02; p=0.08), and was equivalent to an absolute benefit of 5% at five years. This study sparked a number of randomised controlled clinical trials designed to provide a definitive answer to the question of the role of adjuvant chemotherapy in resected NSCLC. The results of these studies have been inconsistent leading to the conduct of two independent meta-analyses based on individual patient data.

In the meta-analyses from the NSCLC Collaborative Group, an absolute improvement in five-year survival of 5% (95% CI, 3-8) for stage III disease (from 30% to 35%) was observed. In all but one trial (CALGB 9633), cisplatin was the platinum agent. Although there was no evidence of difference in the effect of chemotherapy between stage III NSCLC patients with good and poor performance status, an increasing relative effect of chemotherapy with improving performance status (PS) (trend p=0.002) was noted and was consistent across trials.

In the LACE meta-analysis, a 17% reduction in risk of death for stage III NSCLC patients was identified with the HR being 0.83 (95% CI, 0.72 - 0.94) in this group. Chemotherapy was likely to be detrimental for patients with PS of 2. A small excess of non–lung cancer deaths was observed in the adjuvant chemotherapy group in the first six months, which corresponded to a 2% decrease in survival after chemotherapy from 98.6% to 96.6%. The non–lung cancer deaths were mainly related to chemotherapy toxicity and an excess of pulmonary/cardiovascular deaths. Significantly more elderly patients (≥ 70 years) died from non-lung cancer-related causes, and elderly patients received significantly lower first and total cisplatin doses, and fewer chemotherapy cycles.

A subgroup analysis for the cisplatin-vinorelbine regimen had been pre-specified in the LACE statistical analysis plan. Patients who were randomised to cisplatin-vinorelbine or observation were the largest subgroup (41%) and were the most homogeneous in terms of drug doses and eligibility. There was a significant interaction of cisplatin-vinorelbine effect and stage III disease. The greatest benefit was seen in patients with stage III disease who had a 14.7% improvement in overall survival at five years. In comparison with untreated controls, the HR of overall survival for adjuvant cisplatin/vinorelbine chemotherapy in patients with stage III disease was 0.66 (95% CI, 0.53 - 0.83). There is a suggestion that higher exposure to cisplatin (cumulative dose >300mg/m²) may confer a greater benefit although in the group of patients receiving cisplatin-vinorelbine, this effect cannot be dissociated from any benefit that may be conferred by the companion drug(s).

Adjuvant chemotherapy is not without toxicity. Chemotherapy-related deaths were 0.9% in the LACE meta-analysis and 1.4% for patients receiving cisplatin-vinorelbine versus 0.4% for those patients receiving cisplatin in combination with other drugs.

Interestingly, longer term follow up of IALT, which was the largest study in the LACE meta-analysis, showed continued survival benefit of adjuvant chemotherapy but which was no longer statistically significant. Although the survival benefit continued to be manifest as reduced recurrence rates at local and distant sites (with the exception of brain), its extent may have been diminished by accumulating non-lung cancer deaths more than five years post-randomisation. The statistically significant causes of non-cancer deaths were infections and circulatory and respiratory diseases rather than second cancers.

In a more recent randomised controlled trial, Shen et al reported on 140 patients with stage IIIA-pN2 NSCLC who were randomised to receive adjuvant chemotherapy or chemoradiotherapy. However, the study closed early, fell short of its target sample size of 300, and had an inadequate follow-up period. There was no statistically significant difference between treatment groups for overall survival, which was the primary endpoint of the study.

Finally, in a subgroup of NSCLC patients at high risk of significant morbidity, the North American intergroup 0160/Southwest Oncology Group (SWOG) 9416 reported on 111 patients with T3-T4 N0-N1 superior sulcus tumours. Patients were treated with neoadjuvant cisplatin and etoposide and concurrent radiation to 45Gy in 25 fractions. Following restaging at two to four weeks, patients with stable or responsive disease underwent thoracotomy. All patients were to receive an additional two cycles of chemotherapy. With this approach 75% (83 of 111) patients completed the entire treatment regimen. A complete resection (R0) was possible in 75 patients (90%) and gross total resection (R0 or R1) in 76 patients (92%). A five-year survival of 44% was reported for the entire group and for cases in which a complete response was achieved, the five-year survival was 54%. The authors reported these results were achieved with acceptable morbidity and mortality. The mortality rate was 2.7%.There was no information on quality of life.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
In patients with completely resected stage III NSCLC, adjuvant cisplatin-based chemotherapy increases survival compared with observation. Further research is required to identify which stage III patients have the most favourable risk-benefit profile for adjuvant chemotherapy.	I	,
The extent of this survival benefit diminishes with time as the number of non-lung cancer deaths accumulates.	I

Evidence-based recommendation	Grade
Patients who have a good performance status (WHO 1, 2) and completely resected stage III non-small cell lung cancer should be offered adjuvant cisplatin-based chemotherapy.	A

Evidence summary	Level	References
Limited evidence from one well-conducted study supports using induction chemoradiotherapy for locoregional control for patients with superior sulcus NSCLC.	III-3

Evidence-based recommendation	Grade
Patients with superior sulcus NSCLC may be considered for induction chemoradiotherapy.	C

Practice point
Caution is advised in recommending adjuvant cisplatin-based chemotherapy to good performance status patients who are 70 years of age or older and/or who have clinically significant cardio-respiratory or renal co-morbidities.

Practice point

Patients with resectable stage III non-small cell lung cancer, who are being considered for preoperative chemotherapy and surgery or surgery and postoperative chemotherapy, should have their treatment plan reviewed in a lung cancer-specific multidisciplinary meeting. The recommended treatment plan may need to be individualized to take account of such patient-specific factors as treatment preference, availability and timing of surgery, and geographically remote location.

Figure 1. Neoadjuvant and adjuvant chemotherapy and the MDT

Back to top

Content updates by author Shalini Vinod

What is the clinical benefit of radiotherapy to the lung primary in stage IV NSCLC?

Palliative thoracic radiotherapy

Introduction

The aim of palliative radiotherapy in stage IV Non-Small Cell Lung Cancer (NSCLC) is to alleviate symptoms and improve quality of life. This has to be balanced against toxicity and costs of treatment and the need to attend for treatment for a number of days to weeks.

The cost-utility of different fractionation schemes has been analysed using data from the study by Kramer et al which randomised patients between 30Gy in 10 fractions and 16Gy in 2 fractions. The higher dose provided significantly greater Quality Adjusted Life Years (20 weeks versus 13 weeks) but at greater societal cost. However the cost utility ratio was estimated to be US$40900 per QALY which is within the range of acceptable cost effectiveness for medical interventions.

Back to top

Radiotherapy fractionation scheme for palliative radiotherapy to the lung primary in stage IV NSCLC

Clinical benefit of palliative radiotherapy

Stevens et al conducted a systematic review of 14 randomised controlled trials which randomised between different doses of palliative radiotherapy. These trials were heterogeneous in terms of the performance status of the study population, tumour pathology (NSCLC, SCLC, no pathological diagnosis) and tumour stage (stage III, stage IV or patients who were deemed unsuitable for curative treatment). In the majority of trials the endpoint was symptomatic response although this was measured using different instruments and by patients and/or clinicians at various time points. A meta-analysis was not performed due to heterogeneity of study population and outcomes measured.

All studies showed an improvement in symptoms from lung cancer with no significant difference in palliative benefit between the different fractionation regimens. There was no difference in radiological response. Toxicity was mild overall but was greater in arms with higher radiotherapy dose. The three studies which examined quality of life showed mixed findings and no clear advantage to a particular radiotherapy dose. Four studies showed an improvement in survival with higher radiation doses (20Gy in 5 fractions - 50Gy in 25 fractions). The meta-analysis showed a statistically significant improvement in one year survival with fractionated regimens in patients with good performance status (RR=0.95, CI: 0.90-0.99) but not in those with poor performance status (RR=0.96, CI 0.91-1.02).

The main symptoms palliated by thoracic radiotherapy are cough, dyspnoea, chest pain and haemoptysis. Cough is improved in 20-65% of patients, dyspnoea in 40-55%, chest pain in 39-80% and haemoptysis in 39-95%. The median duration of palliative benefit for cough is 56-78 days, chest pain 56-74 days and haemoptysis 64-146 days. This duration of palliation equates to 50% or more of the patients remaining survival time.

None of the trials showed any superiority of a particular radiotherapy dose and fractionation in achieving symptomatic response. A higher radiotherapy dose (30Gy in 10 fractions) has been associated with a longer duration of response and a slower progression of symptoms. There is greater improvement in quality of life with higher radiotherapy doses (20Gy in 5 fractions). Higher doses (20Gy in 5 fractions, 30-39Gy in 10-13 fractions) have also been associated with a two month improvement in median survival and a 5-9% increase in one year survival and 3% at two years. This survival benefit is largely in those patients with stage III NSCLC and good performance status but Kramer et al also demonstrated this for patients with stage IV NSCLC and good performance status.

Back to top

Toxicity of radiotherapy

Palliative thoracic radiotherapy may cause fatigue, oesophagitis and pneumonitis. Acute toxicity is generally mild and self-limiting. Higher radiotherapy doses have been associated with increased toxicity particularly oesophagitis which can occur in 40-56% of patients. More recent trials have shown no difference in acute toxicity between lower and higher doses of radiotherapy. Late toxicity is uncommon, however, radiation myelitis has been documented in up to 2.5% of patients receiving 17Gy/2 fractions and 39Gy/13 fractions. This can be reduced by limiting the spinal cord dose to LQED2 of 48Gy.

Back to top

Optimal timing of palliative radiotherapy to the primary lung cancer in stage IV NSCLC

Timing of radiotherapy

In patients with minimal thoracic symptoms there is no advantage to immediate radiotherapy. A randomised trial of immediate versus delayed radiotherapy in these patients demonstrated that the chance of being alive and without moderate symptoms at six months was the same regardless of whether initial radiotherapy was given or not. Only 42% of patients who did not receive initial radiotherapy, required it later for symptoms. There was no difference in psychological distress between the two patient groups.

Back to top

Endobronchial brachytherapy

Conventional external beam radiotherapy is delivered by a linear accelerator. An alternative way of delivering radiotherapy is via a catheter placed endobronchially at the site of the cancer. A radioactive seed travels through this catheter and releases radiation in close proximity to the cancer without the need to travel through healthy normal tissue. However the range of radiation delivered is small, in the order of 1-2cm, and this technique is not suitable for large extra-bronchial tumours.

Reveiz systematically reviewed 14 randomised controlled trials comparing endobronchial brachytherapy to external beam radiotherapy and other interventions such as chemotherapy and laser. These trials had small study populations and were heterogeneous with regard to the range of radiotherapy doses and other treatment modalities. They found that external beam radiotherapy had greater palliative efficacy compared with brachytherapy alone. There was no evidence to support a combination of external beam radiotherapy and brachytherapy over external beam radiotherapy alone. Endobronchial brachytherapy should be reserved for select patients who have previously been treated with external beam radiotherapy and have symptomatic recurrent central endobronchial tumour.

There has been one randomised trial evaluating the dose of endobronchial brachytherapy. 142 patients with centrally located malignant tumours were randomised between 4 fractions of 3.8Gy and 2 fractions of 7.2Gy. Local tumour response as assessed at bronchoscopy was significantly greater for the 2 fraction course. There was no significant difference in fatal haemoptysis or survival.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
Palliative thoracic radiotherapy can relieve symptoms due to primary lung cancer.	I
Lower doses of radiotherapy (10Gy in 1 fraction, 17Gy in 2 fractions) are equivalent to higher doses (20Gy in 5 fractions, 30-39Gy in 10-13 fractions and higher) in terms of symptom palliation.	I
In patients with good performance status, higher doses of radiotherapy (20Gy in 5 fractions, 30-39Gy in 10-13 fractions) give a modest survival benefit of approximately 5% at one year and 3% at two years and are associated with longer duration of symptom palliation.	I, II	,
Acute toxicity of palliative thoracic radiotherapy is generally mild. Higher doses of radiotherapy are associated with greater acute toxicity particularly oesophagitis.	I
Patients with minimal thoracic symptoms do not benefit from immediate thoracic radiotherapy.	II
External beam radiotherapy is more effective for palliation of thoracic symptoms than endobronchial brachytherapy. There is no therapeutic advantage in giving both these treatment modalities over external beam radiotherapy alone.	I

Evidence-based recommendation	Grade
Patients who have thoracic symptoms of moderate severity from their primary lung cancer should be offered a course of palliative external beam thoracic radiotherapy.	A

Evidence-based recommendation	Grade
Patients who are of poor performance status should be treated with lower doses of palliative thoracic radiotherapy (8-10Gy in 1 fraction, 16-17Gy in 2 fractions) as this provides equivalent symptomatic response to higher doses of radiotherapy (20Gy in 5 fractions, 30-39Gy in 10-13 fractions).	A

Evidence-based recommendation	Grade
Patients who are of good performance status should be treated with higher doses (20Gy in 5 fractions, 30-39Gy in 10-13 fractions) of palliative thoracic radiotherapy in order to maximise duration of palliation and survival.	B

Practice point
Patients with a centrally located lung cancer who are at risk of major airway obstruction should be considered for palliative thoracic radiotherapy, even in the absence of symptoms.

Back to top

Content updates by author Stephen Barnett

What is the clinical benefit of resection of primary disease after complete resection of metastatic disease?

Introduction

Improvements in both structural and metabolic imaging in recent years mean that once undetectable metastases may now be identified. This is a ‘double edged sword’. Whilst clinicians may be encouraged to pursue an aggressive approach on the basis that widespread metastasis have not been identified (despite PET, MRI and high resolution CT scan), a more pessimistic view is that it is only the high sensitivity of the imaging that has detected a metastasis whilst it is still solitary and that this in turn represents the ‘tip of an oncological iceberg’ that would in past years have been detected only when protruding from the water and at multiple sites.

Back to top

Case series: Resection of primary disease after complete resection of metastatic disease

Multiple case reports and retrospective case series have reported long term survival in highly selected patients with brain, adrenal, small bowel, spleen, lymph node, skeletal muscle, and bone metastases with 5 year survival ranging from approximately 5% – 30% after resection of both primary and metastatic sites.

A recent systemic review of publications reporting patients with isolated metastasis to sites other than brain or adrenal accumulated 62 patients undergoing complete resection of metastatic site after definitive treatment of primary. The study found a clinically and statistically significant difference on multivariate analysis in survival with a hazard ratio of 8.2 (95%CI:2.1–32.5),p=0.003, for involvement of mediastinal lymph nodes.

The only published prospective phase II study was reported by Downey et al and details the treatment of a heterogenous group of 23 patients between 1992 and 1997 at a single US centre with a solitary, synchronous, resectable metastasis (including brain, adrenal, bone, lung, spleen and colon), a T 1-3, N 0-2 NSCLC and good performance status and adequate cardio-respiratory reserve to allow lung resection. Treatment included induction chemotherapy with mitomycin, vinblastine and cisplatin, followed by restaging, resection of all sites of disease and adjuvant vinblastine and cisplatin. In the case of a brain metastasis it was resected prior to induction chemotherapy and whole brain irradiation administered at the discretion of the treating neurosurgeon.

The median survival for all patients entered into the study was 11 months. Actual five-year survival from time of thoracotomy was 8%.

The author concludes that induction chemotherapy; surgical resection of primary and metastatic sites; and adjuvant chemotherapy is so poorly tolerated and commonly associated with disease progression as to preclude its recommendation. Further, that retrospective series reporting superior survival epitomise selection bias with only those selected for and completing resection of both sites of disease finding their way into institutional databases subsequently searched to report retrospective experience. This fact is imperative to appreciate when counseling an individual patient with an isolated metastasis considering embarking on an aggressive approach with curative intent, as even the fittest patients screened and accepted onto a phase II protocol have a 4 – 8% chance of long term disease free survival.

Back to top

Evidence summary and recommendations

Practice point
In highly selected patients with T1-3 N0-1 lung cancers with good performance status, adequate pulmonary reserve and solitary site of metastasis, it may be reasonable to consider resection of primary and metastatic sites.

Practice point
It is advisable to consider only those patients who would require less than pneumonectomy and with T 1-3, N0-1 NSCLC for resection of primary and metastatic sites.

Back to top

Content updates by author Shawgi Sukumaran

What is the role of chemotherapy after surgery in the treatment of operable stage II NSCLC?

Introduction

This clinical practice guideline addresses the question of the role of chemotherapy after complete resection of stage II NSCLC. The primary outcome assessed while preparing these guidelines is overall survival. The studies examined have used the 6th edition of TNM staging. Curative treatment for early stage NSCLC is surgery. However 30-60% of patients treated with surgery develops recurrence. Many of the recurrences are systemic, indicating that adjuvant treatment might be beneficial. The question of adjuvant treatment in patients with positive margins after surgery has not been addressed here. Adjuvant use of Tegafur- Uracil is excluded as these drugs have not been well studied in Caucasian population.

Back to top

Chemotherapy after surgery in stage II NSCLC

Several studies have confirmed the benefit of adjuvant chemotherapy in Stage II NSCLC. A large meta-analysis of individual patient data involving 8447 patients from 34 trials (1291 stage II patients) reported a 5% (from 40 to 45%) absolute benefit in overall survival, at five years, in stage II patients, using platinum based chemotherapy. These results are confirmed by results of a pooled analysis (LACE meta-analysis) of individual patient data from five well designed, large, randomised controlled trials which included a total of 4584 patients. Out of this 1616 patients had stage II disease. At a median follow up of 5.1 years the absolute overall survival benefit was 5.4% for the whole population and 10% for stage II patients (from 39% to 49%) with HR of death for stage II patients at 0.83 ( 95% CI 0.73-0.95). This analysis included only patients treated with Cisplatin based chemotherapy as this was shown to be the combination most effective in a previous large meta-analysis. The same study also revealed combination of alkylating agents with Cisplatin to be detrimental and this combination is no longer recommended.

The major characteristics of the five major randomised controlled trials, included in the LACE meta-analysis mentioned above, are shown in Table - Summary of five randomised trials included in the LACE meta-analysis. These constitute the major evidence on which these guidelines are based upon. The first three trials showed a survival benefit with adjuvant Cisplatin based chemotherapy while the last two did not. The Big lung trial was underpowered to demonstrate survival benefit. The IALT is the largest trial and was the first RCT to demonstrate a statistically significant benefit in overall survival. However, in an updated publication, with a median follow up of 7.5 years, the survival benefit in favour of chemotherapy became non-significant. In contrast, results of long term follow up of the JBR 10 trial (median 9.3 years), showed that the survival benefits continued to be significant for the overall population as well as for patients with stage II disease. The median age was similar in all the trials. The majority of patients had good performance status (ECOG 0, 1) with only 4.5-7.2% of patients having an ECOG performance status of 2, in trials including them. At least 50% of enrolled patients completed the planned number of chemotherapy cycles across the studies. The dose of Cisplatin varied from 80-120mg/m2 per cycle. Cisplatin was combined with Vinorelbine in two of the trials while in the rest, it was combined with a range of other drugs. These include Etoposide, other Vinca alkaloids, Mitomycin C, Ifosfamide and Prednisolone.

Pre-specified subset analysis of the LACE meta-analysis looking at the combination of Cisplatin and Vinorelbine found this combination to be superior in terms of overall survival when compared to the other combinations. HR 0.80, 95% CI: 0.70–0.91, p<0.001 versus HR 0.95, 95% CI: 0.86– 1.05, p = 0.33. Chemotherapy was reasonably well tolerated with 50% or more patients completing all planned cycles. Toxic deaths reported ranged from 0.8 % (JBR 10) to 2% (IALT). The absolute survival benefit at five years in the three positive trials ranged from 3.9 % (IALT), 8.6% (ANITA) and 15% (JBR10). Quality of life (QOL) of the patients were reported from the JBR10 trial. It was seen that the QOL was impaired in the immediate period following chemotherapy. However, scores improved to match the control group within a period of nine months. QOL scores for peripheral neuropathy and hearing impairment persisted in the chemotherapy group for up to 30 months. These results have been further confirmed by an updated individual patient data meta-analysis. An absolute improvement in 5-year survival of 5%, from 40-45%, for stage II disease was again demonstrated. The benefit for chemotherapy was also evident in patients who received adjuvant radiotherapy.

The Cancer and Leukemia Group B trial (CALGB 9633) looked at benefit of four cycles of Paclitaxel/Carboplatin as adjuvant treatment in patients with stage IB NSCLC. The regime was well tolerated with no toxic deaths. The trial did not demonstrate any overall survival benefit. However, in an exploratory subset analysis of the trial, patients with tumour diameter >4cm demonstrated a significant survival advantage. These tumours would be classified under IIA in the 7th edition of TNM classification. However, no firm recommendation can be made regarding adjuvant chemotherapy in this subgroup.

These groups of patients may benefit from adjuvant chemotherapy as evidenced by updated meta-analysis data.

There is limited evidence on the long term toxicities of adjuvant chemotherapy. An updated analysis of the IALT study with a follow up of 7.5 years confirmed a benefit of adjuvant chemotherapy within the first 5 years. However after that period there was an increased risk of non-cancer mortality in the group who received chemotherapy( HR 3.6; 95% CI 2.2-5.9, p< 0.001). These results need to confirmed from results of similar studies.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
In patients with operable stage II NSCLC, the evidence supports the use of 3-4 cycles of adjuvant cisplatin-based chemotherapy after surgery.	I, II	, , , , , , , , , ,
The benefit of chemotherapy after surgery is present in patients who received radiotherapy as part of the loco-regional treatment.	I

Evidence-based recommendation	Grade
Patients with completely resected stage II NSCLC should be offered 3-4 cycles of adjuvant cisplatin based chemotherapy.	A

Practice point

The chemotherapy combination of cisplatin and vinorelbine was the most widely studied regimen which showed benefit. There is insufficient evidence to support adjuvant chemotherapy for patients with ECOG performance status of ≥ 2. No recommendation can be made for patients who have had less than a lobectomy. Based on the 7th edition of TNM classification tumour size of >5cm would fall under stage IIA. These patients may be considered for adjuvant chemotherapy. Chemotherapy benefit is seen even in patients who have received radiotherapy as part of loco-regional therapy in addition to surgery. Potential long term side effects need to be considered while deciding on chemotherapy after surgery.

Back to top

What is the role of chemotherapy before surgery in the treatment of operable stage II NSCLC?

Introduction

This clinical practice guideline addresses the question of the role of chemotherapy before surgery or neoadjuvant chemotherapy in operable stage II lung cancer. This does not address treatment of tumours involving superior sulcus. There are theoretical advantages in using chemotherapy in this setting, although available evidence is non-conclusive. Randomised controlled trials (RCTs) and meta-analysis have been confounded by low number of patients with stage II disease, poor accrual, early closure and significant heterogeneity. Sub group analysis of these trials cannot be considered as conclusive evidence regarding use of neoadjuvant chemotherapy.

Back to top

Neoadjuvant chemotherapy

The major evidence on which these guidelines are based comes from five RCTs and two meta-analyses.The salient features of the RCTs are depicted in the Table - Summary of five randomised trials comparing neoadjuvant chemotherapy with surgery alone.

The recently published CHEST trial had progression free survival (PFS) as its primary end point. The HR for PFS and overall survival (OS) was significant in favour of the neoadjuvant chemotherapy arm. Patients with stage IIB/IIIA where grouped together and showed significant benefit for both PFS and OS compared to control group while IB/IIA patients did not have any significant benefit. This contrasts with results of the S9900 trial, which did not show any significant OS difference in the overall population, while subgroup IB/IIA demonstrated a significant OS benefit. The NATCH trial had an adjuvant arm in addition to the neoadjuvant chemotherapy arm. The primary endpoint of this study was disease free survival (DFS). There was no difference in DFS and OS amongst the three groups. Stage II patients receiving neoadjuvant chemotherapy demonstrated a trend towards improved DFS, which failed to reach statistical significance. Interestingly 90% of subjects in the neoadjuvant chemotherapy arm completed all planned chemotherapy compared with 60% in the postoperative arm. Surgical outcomes and postoperative mortality were similar. The European intergroup trial MRC-LU22 EORTC NVALT trial did not demonstrate any overall survival benefit with neoadjuvant chemotherapy. Neoadjuvant chemotherapy did not have any impact on the quality of life. The FTCG study did not demonstrate any survival benefit in the population studied. However ,subset analysis revealed survival benefit in stage I/II disease. The compliance rates were very good across the trials, ranging from 75-90%. The response rates ranged from 34% to 64%. A systematic review and meta-analysis including 988 patients across seven RCTs demonstrated an overall survival benefit (HR 0.82, 95% CI 0.69-0.97;P=0.02). This equated to an absolute improvement in overall survival of 7% at five years in patients with stage II disease. Updated analysis including the LU22 trial demonstrated a shift in HR to 0.88 (0.76-1.01) with the benefits not maintaining statistical significance. However, these meta-analyses were not based in individual patient data and meaningful subgroup analysis could not be undertaken for early stage disease due to significant heterogeneity amongst the trials. The benefits of neoadjuvant chemotherapy are similar to that of adjuvant chemotherapy with absolute benefit of 6% at five years from meta-analysis. However, the overall evidence is insufficient to recommend neoadjuvant chemotherapy as standard therapy for operable stage II NSCLC. Individual patient meta-analysis of all the available trials may help to clarify this question further across all the sub-groups. A more recent individual patient meta-analysis provides a clearer picture across these studies. This individual patient meta-analysis included 15 randomised controlled trials involving 2385 patients majority, across stages 1B- IIIA. Out of 1194 patients for whom staging information was available 330 (28%) were stage II patients. A clear overall survival benefit was demonstrated with a HR of 0.87, 95 % confidence interval 0.78-0.96,P=0.007. There was a 13% decrease in relative risk of death with an absolute survival improvement of 5% at 5 years from 40% to 45% overall and 30% to 35% in stage II. This benefit was independent of other variables tested including chemotherapy regimen, patient demographics and tumour characteristics. There was no demonstrable effect on the operability rate or on the likelihood of achieving a complete resection with administration of .chemotherapy before surgery.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
Individual patient meta-analysis shows benefit in overall survival for chemotherapy given before surgery.	I
Individual studies looking at chemotherapy use before surgeryThe evidence of benefit for neoadjuvant chemotherapy in stage II disease is unclear with trials having have inconsistent end points and lacking power due to poor accrual and early closure.	II	, ,
Majority of individual trials do not show statistically significant benefit in stage II disease.	II	, , ,
Neoadjuvant Chemotherapy given before surgery does not adversely affect the quality of life.	II
Compliance to neoadjuvant chemotherapy is better compared to adjuvant chemotherapy.	II
Meta-analysis showed overall survival benefit, however subset analysis for stage II disease was not undertaken due to heterogeneity of individual trials included.	I
The benefits of neoadjuvant chemotherapy are similar to that of adjuvant chemotherapy	I	,

Evidence-based recommendation	Grade
Chemotherapy before surgery is not recommended as standard of care for patients with operable stage II NSCLC.	B

Evidence-based recommendation	Grade
Chemotherapy before surgery may be considered as an option for patients with operable stage II NSCLC.	A

Evidence-based recommendation	Grade
Chemotherapy before surgery in operable stage II disease, with at least three 3-4 cycles of platinum-based regimes, may be considered in select patients, who are unlikely to receive it as adjuvant therapy.	CA

"CA" is not in the list (A, B, C, D, GPP, PP, C/GPP, A (ESPEN grade), N/A, Strong, ...) of allowed values for the "Recommendation grade" property.

Practice point
No benefit has been demonstrated in using chemotherapy before surgery to improve resection rates. Benefit of chemotherapy seem to be similar when given either before or after surgery. Chemotherapy before surgery may be considered for those patient who are expected to have prolonged delay in surgery.

References

Back to top