Efficacy of neoadjuvant chemotherapy followed by radical surgery versus concurrent chemotherapy and radiotherapy in patients with stage IB2, IIA, or IIB cervical squamous carcinoma: a randomized controlled trial
objective
We compared the efficacy and toxicity of neoadjuvant chemotherapy followed by radical surgery with standard cisplatin chemoradiotherapy in people with locally advanced cervical squamous cell carcinoma.
Patients and methods
This is a single-center, phase III, randomized controlled trial (ClinicalTrials.gov identifier: NCT00193739). Eligible patients are between the ages of 18 and 65 years with stage IB2, IIA, or IIB squamous cervical cancer. Stratified by stage, patients were randomized to receive three rounds of neoadjuvant chemotherapy (paclitaxel and carboplatin every 3 weeks followed by radical hysterectomy) or standard radiotherapy with cisplatin once weekly for a total of 5 weeks. Patients in the neoadjuvant group received postoperative adjuvant radiotherapy or concomitant chemotherapy and radiotherapy (if indicated). The primary endpoint is disease-free survival (DFS), defined as survival without recurrence or cancer-related death, with secondary endpoints including overall survival and toxicity.
outcome
Between September 2003 and February 2015, a total of 635 patients were randomized, of whom 633 (316 patients in the neoadjuvant chemotherapy plus surgery group and 317 patients in the concomitant chemoradiotherapy group) entered the final analysis, with a median follow-up of 58.5 months. The 5-year DFS rate was 69.3% in the neoadjuvant plus surgery arm versus 76.7% in the chemoradiation arm (hazard ratio 1.38; 95% CI, 1.02 to 1.87; P = .038), compared with the corresponding 5-year OS rate of 75.4% and 74.7%, respectively (hazard ratio 1.025; 95% CI, 0.752 to 1.398; P = .87). Delayed toxicities at 24 months or later after completion of treatment in the neoadjuvant chemotherapy plus surgery versus concurrent chemoradiotherapy were rectal (2.2% v. 3.5%), bladder (1.6% v. 3.5%), and vaginal (12.0% v. 25.6%), respectively.
conclusion
For locally advanced cervical cancer, cisplatin-based concurrent chemoradiotherapy has superior DFS compared with neoadjuvant chemotherapy followed by radical surgery.
introduce
Cervical cancer is a major public health problem worldwide, with squamous cell carcinoma being the most common subtype. International Federation of Obstetrics and Gynecology (FIGO) patients with stage IB2 to IVA are considered to have locally advanced cancer. Radiation therapy is the mainstay of treatment, but definitive surgery can also be performed in patients with stage IB2 or IIA.
Multiple trials have demonstrated that concurrent chemotherapy and definitive radiotherapy improve survival compared with radiotherapy alone, resulting in a paradigm shift in the treatment of cervical cancer. Although one trial failed to demonstrate benefit, concurrent platinum-based chemotherapy and radiotherapy have been the standard treatment for locally advanced cervical cancer since 1999, based on cumulative evidence. However, 25% to 40% of patients still experience recurrence, and some patients develop distant metastases despite local control after chemoradiotherapy; In addition, this treatment is associated with early and long-term toxicity. Therefore, further improvements in treatment outcomes are needed, but there have been few reports of definitive studies on new treatment strategies over the past two decades.
Radical hysterectomy followed by neoadjuvant chemotherapy is considered an attractive approach to improve disease control and reduce toxicity. Cervical cancer has a high response rate to modern therapies, including taxanes and platinum-based drugs. Neoadjuvant chemotherapy has the potential to eradicate micrometastases and reduce systemic failure, in addition to promoting local control through surgical resection. Several studies and meta-analyses have shown that surgery followed by neoadjuvant chemotherapy improves survival outcomes compared with radiotherapy. However, radiotherapy without concomitant chemotherapy has become an outdated standard, and the chemotherapy regimens used in these studies are not contemporary. Neoadjuvant chemotherapy prior to definitive radiotherapy failed to show benefit, possibly because radiotherapy-resistant clones were chosen, but there may be no such interaction between chemotherapy and surgery. Although the neoadjuvant chemotherapy plus surgical approach is promising, it continues to be practiced in many parts of the world despite the lack of sufficient evidence.
Therefore, we conducted a randomised controlled trial comparing neoadjuvant chemotherapy (using a contemporary paclitaxel and carboplatin regimen) followed by radical surgery versus concurrent chemoradiotherapy in patients with stage IB2, IIA, or IIB cervical squamous carcinoma.
discuss
We report here the results of the first randomized study known to us that compared neoadjuvant chemotherapy followed by radical surgery with concurrent chemoradiotherapy in patients with locally advanced squamous squamous cervical cancer. Treatment is provided by gynaecologist, internal medicine and radiation oncology specialists at a tertiary cancer centre in India. Concurrent chemoradiotherapy improved the 5-year DFS rate compared with neoadjuvant chemotherapy and surgery, with an absolute difference of 7.4 percentage points between the two groups. Subgroup analyses suggest that the primary benefit of concurrent chemoradiotherapy lies in patients with stage IIB disease. There was no difference in OS between the two groups, but the study was unable to draw definitive conclusions about this endpoint.
This study is based on the hypothesis that neoadjuvant taxanes in combination with platinum-based chemotherapy will significantly reduce the risk of distant recurrence and promote local control through surgery compared to chemoradiotherapy. However, our results do not deliver on the promise of the neoadjuvant chemotherapy combined with surgical strategy compared to radiotherapy alone. Notably, the control arm in our study included concurrent cisplatin chemotherapy with a cumulative dose of up to 200 mg/m2, which may have provided a level of systemic control that cannot be further improved by neoadjuvant chemotherapy regimens. The results suggest that the addition of more chemotherapy in addition to concurrent chemoradiotherapy is unlikely to further improve survival in patients with locally advanced cervical cancer. In this context, the results of three ongoing studies evaluating the effects of chemotherapy will be of interest (ClinicalTrials.gov identifiers: NCT00039338, NCT01414608, and NCT01566240). A study by the European Organization for Research and Treatment of Cancer (ClinicalTrials.gov identifier: NCT00039338) is testing the efficacy of neoadjuvant chemotherapy followed by surgery versus concurrent chemoradiotherapy in stage IB and II patients treated with cisplatin-based neoadjuvant chemotherapy regimens. The second study (ClinicalTrials.gov identifier: NCT01414608) is testing four cycles of paclitaxel plus carboplatin as adjuvant therapy after concurrent chemoradiotherapy. Another previously reported randomized study reported that concomitant and adjuvant gemcitabine and cisplatin chemotherapy improved DFS and OS, but this did not become standard practice due to excessive toxicity.
In our study, the majority of first recurrences (102 out of 162 recurrences, 62.96%) were local recurrences (with or without distant metastases), demonstrating the importance of good local control. There has been a long interest in surgery for locally advanced cervical cancer. Despite adequate neoadjuvant chemotherapy, only 227 patients (72.15%) in the neoadjuvant chemotherapy plus surgery group were able to undergo surgery, which is similar to the ongoing European Organisation for Research and Treatment of Cancer randomized study, which reported a surgical resection rate of 76%. Our findings suggest that the surgical strategy did not improve the rate of local control compared with concurrent chemoradiotherapy. In addition, many surgical patients (32.2%) required adjuvant radiotherapy or chemoradiotherapy, resulting in a significant proportion of patients in this group requiring triple therapy.
There are a variety of reasons that could explain the results of our study. Post-hoc analysis showed that DFS was lower in patients who were unable to undergo surgery and switched to concurrent chemoradiotherapy than in patients who could have surgery (HR, 1.49; 95% CI, 0.94 to 2.38; P = .089; online only). Another post-hoc analysis showed that in the neoadjuvant chemotherapy plus surgery group, patients who could undergo surgery (227 of 316 patients) had worse DFS compared with those who did not receive adjuvant radiotherapy or concurrent chemoradiotherapy (HR, 1.69; 95% CI, 1.02 to 2.80; P = .04). Although the prognosis is worse by selection, delaying definitive chemoradiotherapy and/or cross-resistance between induction chemotherapy and radiotherapy may be detrimental to disease control in these subgroups of patients. It can be said that cisplatin may be more effective than carboplatin when used in combination with paclitaxel. In a Japanese study of patients with recurrent or metastatic disease, 24-hour paclitaxel plus carboplatin was non-inferior to paclitaxel plus cisplatin in the entire population, but the latter regimen had higher OS in the platinum-naïve subgroup. However, subgroup analyses could not be considered definitive. In addition, the two regimens differ not only in terms of platinum, but also in terms of paclitaxel dosage and dosing schedule. Therefore, the choice of platinum-based agents is unlikely to be a key factor in our findings. An ongoing randomized study (ClinicalTrials.gov identifier: NCT01414608) also used paclitaxel plus carboplatin as an adjuvant regimen.
The treatment was well tolerated in both groups, and adverse events were within acceptable limits. Of note, there was no significant difference in the incidence of severity of early non-hematologic toxicity between the two groups. Although late toxicity involving the bladder and rectum was higher after 3 months in the concurrent chemoradiotherapy group, there was no significant difference at 24 months or later, suggesting that toxicity had resolved in most patients. Only vaginal toxicity remained significantly higher at 24 months or later in the concomitant chemoradiotherapy group.
There are some limitations to this study. We included patients at different stages of the disease, and the results of the analysis did not definitively answer the question of patients with operable cervical cancer (stage IB2 and IIA). Quality of life was not measured prospectively and it was not possible to elucidate the relative effect of the two treatment strategies on this outcome.
Our study answers an important long-standing clinical question in the treatment of patients with locally advanced cervical cancer. In this patient population, weekly concurrent chemoradiotherapy with single-agent cisplatin significantly increased DFS rates compared with neoadjuvant chemotherapy followed by radical surgery.
Neoadjuvant Chemotherapy Followed by Radical Surgery Versus Concomitant Chemotherapy and Radiotherapy in Patients With Stage IB2, IIA, or IIB Squamous Cervical Cancer: A Randomized Controlled Trial - PubMed