Radiotherapy and chemotherapy in treatment of oesophageal and gastric cancer

Preoperative radiotherapy alone

There have been six randomised controlled trials (RCTs) of preoperative radiotherapy. Three trials were restricted to squamous cell carcinoma (SCC). A Cochrane group quantitative meta-analysis of preoperative radiotherapy using data from 1147 patients in five randomised trials has been performed. With a median follow-up of 9 years, in a group of patients with mostly SCC, the hazard ratio (HR) of 0.89 (95% CI 0.78–1.01) suggests an overall reduction in the risk of death of 11% and an absolute survival benefit of 3% at 2 years and 4% at 5 years. This result is not statistically significant ( P = 0.062). Preoperative radiotherapy as sole modality is not considered a standard of care.

Postoperative radiotherapy

Given the morbidity of oesophagectomy, patients are often unfit for postoperative therapy to be given within a reasonable timeframe. Furthermore, attempts at radiation therapy are compromised by the need to irradiate the gastric pull-up.

The available literature largely considers patients with SCC alone. Six randomized trials and 13 retrospective studies that included a total of 8 198 patients were included in a meta-analysis of postoperative radiotherapy (PORT) in patients with resected SCC of the oesophagus. The authors noted that PORT provided significant survival benefit compared with surgery alone in retrospective studies but not in RCTs. The randomised trials were deemed of low quality. The conclusion was that there was an improvement in disease-free survival and decreased risk of locoregional recurrence with PORT.

The meta-analysis observed that several of the studies included did not report resection status. In particular whilst longitudinal resection margin involvement in most series would be considered an incomplete resection in others circumferential resection margin involvement is variably reported as such.

To complicate matters, the definition of an R1 resection has changed in international staging manuals. Previously any distance from the resection margin was considered R0 but there is now consensus that R1 resection, and consequent worse prognosis as initially shown in rectal cancer, encompasses any disease within 1 mm of any resection margin.

Postoperative radiotherapy is therefore not established in completely resected oesophageal cancer. There have been no contemporaneous RCTs addressing the role of PORT where there is an R1 resection, where the risk of locoregional failure is clearly higher than with surgery alone.

In the absence of randomised evidence, the knowledge that radiotherapy has a demonstrated role in reducing the risk of locoregional recurrence in surgically treated oesophageal cancer probably justifies considering patients with longitudinal resection margin and/or circumferential resection margin involvement for postoperative CRT on an individual patient basis. Pragmatically, common practice is to target patients where the risk of systemic disease relapse is lower, i.e. those with no, or a lower ratio of, involved lymph nodes. Such a selective policy was shown to have some benefit in an audit of two high-volume tertiary UK referral centres. This showed that postoperative CRT improved overall survival (median survival: 16 months [no postoperative CRT] vs 24 [postoperative CRT] months; HR = 0.46; 95% CI 0.24–0.89; P = 0.021) and relapse-free survival (median survival: 12 [no postoperative CRT] months vs 17 [postoperative CRT] months; HR = 0.5; 95% CI 0.27–0.92; P = 0.026) in the R1 subgroup.

What is the optimal preoperative treatment for operable oesophageal cancer?

This continues to be the subject of much expert debate, limited evidence base and thus widespread international variation in practice

Preoperative chemotherapy

Preoperative chemotherapy (NCT) in both squamous cell and adenocarcinoma appears to achieve consistently good clinical response rates, as judged by symptom improvement, ranging from 47% to 61%. The mainstay of therapy is a platinum, e.g. cisplatin or oxaliplatin, with a fluoropyrimidine, e.g. 5FU infusion or oral capecitabine. These combinations seem to be active in both squamous and adenocarcinoma, although the benefit of adding a third drug such as an anthracycline, e.g. epirubicin, or taxane in SCC in particular, is less certain and therefore these are often omitted.

Postoperative chemotherapy alone

Patients undergoing major resections for oesophageal carcinoma often have a prolonged postoperative phase. The start of chemotherapy is thus often significantly delayed due to poor performance status, and patients commonly choose not to continue. This is reflected in poor chemotherapy completion rates in studies that adopted a perioperative or ‘sandwich’ approach, e.g. Intergroup 0113 and MAGIC. A strategy that relied solely on postoperative treatment could have significant problems. As a consequence, there are few useful trials that address the question of adjuvant postoperative chemotherapy alone.

The literature is largely confined to patients with SCC. One meta-analysis of three randomized controlled trials and six retrospective studies comprised a total of 1 684 cases. Postoperative chemotherapy could improve OS (HR 0.78, 95% CI 0.66–0.91; P = 0.002) and disease-free survival (DFS) (HR 0.72, 95% CI 0.6–0.86; P < 0.001).

Can postoperative chemotherapy be withheld when neoadjuvant therapy has been given?

There is no randomised data available. The UK MAGIC study was designed to give both preoperative and postoperative treatment. In practice only a minority of patients received postoperative treatment. Some believe that postoperative chemotherapy should be offered to patients with a heavy tumour burden after neoadjuvant therapy to try to improve what is likely to be poor outcome, while others believe that postoperative chemotherapy should only be administered to patients with a good response to neoadjuvant therapy. This author favours the latter approach given that the observed pathology reflects the responsiveness or otherwise to systemic chemotherapy.

Preoperative chemoradiotherapy

Radiotherapy is commonly combined with chemotherapy for a number of theoretical reasons and is an attractive preoperative strategy. First, early administration of systemic therapy could deal with occult micrometastases that would lead to overt systemic metastases, and second, the optimum time to treat micrometastases is probably when their volume is minimal, supporting the application of chemotherapy as early as possible. Third, hypoxic cells exist in solid tumours in regions of poor vascularity and are likely to be more resistant to conventional radiotherapy treatment as sole therapy. These cells, however, may be more sensitive to the combined effects of cytotoxic agents and radiotherapy. Finally, certain chemotherapy drugs, such as 5-fluorouracil, cisplatin, gemcitabine and paclitaxel are known to be ‘radiosensitisers’, which result in normal and malignant tissues becoming more sensitive to ionising radiation.

Several strategies that combine systemic chemotherapy with radiation (as definitive locoregional treatment) have been explored:

• concurrent chemoradiation, in which chemotherapy and radiation are delivered synchronously in order to maximise radiosensitisation;

• neoadjuvant (pre-emptive) chemotherapy, in which chemotherapy is administered first, usually followed by assessment of response and implementation of definitive therapy;

• adjuvant therapy, in which systemic treatment is administered after completion of locoregional therapy in an attempt to control micrometastatic disease;

• combination of concurrent chemoradiation with either neoadjuvant or adjuvant chemotherapy.

There is good evidence that pathological complete response (pCR) rates are significantly higher with CRT than with radiotherapy or chemotherapy given alone. pCR and R0 resections are possible surrogate proxies of benefit to patients and do seem to translate into a survival benefit, which remains the gold standard assessment. A review of the MAGIC pathology data showed on univariate analysis, high Mandard TRG (3,4, or 5) and lymph node metastases were negatively related to survival, but on multivariate analysis, only lymph node status was independently predictive of OS (HR = 3.36; 95% CI 1.70–6.63; P < 0.001).

Both radiotherapy and chemotherapy rely on achieving an acceptable balance between increased response rates in the tumour on one hand and normal tissue morbidity coupled with patient tolerance on the other. Whilst many side-effects of chemotherapy occur relatively early in presentation, e.g. hair loss, emesis, and myelosuppression, radiotherapy side-effects can present late, from 6 months to years after treatment. If radical surgery is added in combined modality therapy then the potential for higher levels of morbidity and mortality becomes significant.

Non-randomised studies of CRT have appeared in the literature since the late 1980s (reviewed by Geh et al.). Pooled data from these studies is notable for the heterogeneity of approach, in terms of scheduling of chemotherapy and radiotherapy, and eligibility criteria. Geh et al. showed that, of 2 704 patients (squamous 68% and adenocarcinoma 32%), 79% were operated on with a pCR rate of 24% of those treated and 32% of those resected. The controversy and delay in implementing what is clearly an active therapeutic approach has revolved around the associated increased morbidity and postoperative mortality. Reported CRT-related deaths in the non-randomised series ranged from 0–15% (mean 3%). Postoperative deaths ranged from 0% to 29% (mean 9%). Adult respiratory distress syndrome, anastomotic leak and breakdown, pneumonia and sepsis were the commonest causes of death following oesophageal resection. Treatment-related deaths ranged from 3–25% (mean 9%) of all patients treated. It seems clear that the risk of chemotherapy-related toxicity, particularly myelosuppression, rises with the number of drugs used and the radiation dose-intensity of the CRT regimen. An increased risk of tracheobronchial fistula has also been reported. However, most of these reported series did not use contemporaneous radiotherapy planning and treatment techniques that allow greater precision and sparing of organs and tissues to within normal tissue tolerance.

In the ‘CROSS’ trial, 363 patients with operable oesophageal or oesophagogastric junction tumours were randomised to surgery alone or to a preoperative CRT regimen (NCRT) of weekly carboplatin (AUC2) and paclitaxel (50 mg/m ² ) concurrent with radiotherapy (41.4 Gy in 23 fractions). Of the 175 patients assigned to the CRT arm, 163 completed protocol treatment and the study reported a low incidence of grade 3/4 CRT toxicity (haematological, 6.8%; non-haematological, 16%). The R0 resection rates in the surgery and CRT + surgery arms were 67% and 92.3%, respectively ( P = 0.002). The results of this study, would suggest that where preoperative CRT is delivered safely, there is a significant improvement in outcome compared to surgery alone.

The study also gives clues as to future directions. CRT delivered a major reduction in R1 resections that translated into a reduction in local failure from 20.5–7%. However, its effect on haematogenous spread remains limited (35.4–28.5%). Distant disease control clearly remains suboptimal with this approach.

Meta-analysis of 21 RCTs involving 3 138 patients were included in a comparison of NCRT with surgery. Compared with surgery alone, NCRT was associated with increased OS (HR 0.74, 95% CI 0.66–0.82, P < 0.01; high quality), 5-year OS (RR 1.51, 95% CI 1.28–1.78, P < 0.01; (high quality) and R0 resection rate (RR 1.16, 95% CI 1.07–1.25, PP < 0.01; moderate quality). Improvement in outcome was seen in both adenocarcinoma and SCC histology

Which preoperative strategy: neoadjuvant chemoradiotherapy or chemotherapy?

To date there have been eight reported randomised phase III trials that include the comparison of NCRT with NCT. The largest trial to date, however, includes fewer than 240 patients. Not surprisingly therefore these studies have been underpowered to answer the question individually but have been subject to a number of meta-analyses the most recent published in 2020. Compared with NCT, NCRT was associated with increased OS (HR 0.78, 95% CI 0.62–0.99, P = 0.04; high quality), 5-year OS rate (RR 1.48, 95% CI 1.06–2.07, P = 0.02; moderate quality), R0 resection rate (RR 1.13, 95% CI 1.07–1.20, P < 0.01; high quality), and pCR (RR 3.74, 95% CI 2.03.6.88, P < 0.01; moderate quality). The 30-day postoperative or in-hospital mortality rate was 7.7% with NCT and surgery versus 8.0% with surgery alone.

Among the SCC patients, meta-analysis provided high-quality evidence of a strikingly improved OS associated with NCRT compared with surgery alone, not observed with the use of NCT. In contrast, in patients with oesophageal adenocarcinoma there was longer OS following NCRT than following NCT, and the HR of NCRT versus surgery alone (HR 0.73, 95% CI 0.62–0.86; high quality) was better than that of NCT versus surgery alone (0.83, 0.72–0.96; moderate quality). The current data is moving towards NCRT.

Whether current trials will finally answer this question is a moot point. The current international study neo-AEGIS trial initially aimed to recruit 594 patients randomised between the ‘CROSS’ CRT regimen and perioperative chemotherapy and was initially powered to show a 10% improvement in 3-year survival. Due to poor accrual it has reduced its sample size to 540 and is now a non-inferiority design.

Earlier identification of non-responding patients to chemotherapy (e.g. by PET-CT scanning) could allow a change in neoadjuvant therapy direction to reduce the R1 resection rate, the number of involved lymph nodes and possibly translate into improved outcomes.

The good outcomes from surgery alone in stage I disease make neoadjuvant therapy difficult to justify. Given the frequent upstaging of clinical stage II disease when resected fit and young patients will often be offered preoperative therapy.

Future direction

There is rightly a clear separation in current and future trials for adenocarcinoma and squamous carcinoma. The majority of adenocarcinoma patients will present with stage III disease (at least T3 with lymph node metastases). There are two issues to consider:

• 1.

  Local control.

  • Tumours frequently threaten the CRM, although a clear plane for surgical excision does not exist as it does for other anatomical sites such as the rectum. Disease present at or within 1 mm of the circumferential margin (R1) occurs in around one-third of cases and is a poor prognostic factor. Preoperative CRT has become a standard management strategy in rectal cancer for patients who have a threatened CRM on preoperative staging. MRI scanning revolutionised the identification of patients with rectal cancer at risk of a positive margin and preoperative CRT is mandatory for cases where MRI predicts a positive margin. Such a paradigm shift is required for oesophageal cancer. Currently there is no agreed staging tool that accurately predicts R1 resection.

• 2.

  Systemic control

  • Whatever improvements in locoregional treatments are proposed, the strategy being addressed with new trials for stage III adenocarcinoma is systemic relapse, in other words overall survival. In particular larger RCTs using immunotherapy in addition to chemotherapy will become available, combination chemo-immunotherapy having demonstrated benefit in incurable disease as will be described later.

  • The provisional results of CheckMate 577 have been made available. 794 patients, 70% of whom had adenocarcinoma, who had received NCRT and had residual pathological disease were randomised 2 to 1 to receive 1 year of nivolumab or placebo. The primary endpoint was disease-free survival, which was 11 months in the placebo group and 22.4 months in the investigational arm (HR.0.69 (0.56–0.86; P = 0.0003)). Whilst mature survival results remain not available this study potentially raises a step change in outcome.

  • Hopefully the addition of immunotherapy to NCT will improve systemic treatment for the disease. At the same time these studies will re-ignite the long-standing debate of the role of radiotherapy in the preoperative management of operable oesophageal cancer.