The optimal time of starting adjuvant chemotherapy after curative surgery in patients with colorectal cancer

Colorectal cancer (CRC) is one of the most aggressive gastrointestinal tract cancers and the incidence and mortality rank third and second, respectively [1]. For patients with high-risk stage II and stage III CRC, 5-flurouracil-based adjuvant chemotherapy (AC), which has been verified to decrease the risk of recurrence and cancer associated death via eliminating tumor micro-metastasis [2, 3], was recommended to applied following radical surgery by the National Comprehensive Cancer Network (NCCN).

Recently, the result of the International Duration Evaluation of Adjuvant Therapy (IDEA) study fueled interests in specifying the timing and duration of AC after surgery [4, 5]. With no recommendations provided in NCCN guidelines, AC was generally initiated 6–8 weeks after surgery in most clinical trials. However, delays frequently occurred in daily practice for various reasons [6]. Previously, two meta-analyses demonstrated that postponing the postoperative AC was associated with poor survival in CRC patients. Result from Biagi et al. showed that each 4 weeks delay result in a 14% decrease of overall survival (OS) [7]. Similarly, Guetz’s study indicated that delaying the initiation of AC for > 8 weeks after operation significantly decreased OS [8]. However, owing to limited research data and small sample size, no consensus has been reached on the actual time to start AC after surgery. Recently, studies with larger sample size has been conducted to investigate the impact of the timing of AC on prognosis in CRC, however the results were inconsistent [9‐12]. Therefore, further analysis was needed to instruct clinical practice.

Here in this study, we performed updated meta-analysis, comparing the survival benefits of different time intervals between surgery and postoperative AC (including < 8 weeks vs. > 8 weeks and ≤ 4 weeks vs. 5–8 weeks), to explore the optimal time to initiate AC after surgery. In addition, based on the existing data, the time point at which postoperative AC was no longer beneficial to CRC patients was discussed.

According to this updated meta-analysis, delaying initiation of AC after surgery for > 8 weeks was significantly associated with poorer OS. The negative prognostic indication value of delaying initiation of AC for > 8 weeks was not undermined by subgroup analysis based on region, tumor site, sample size and study quality. In addition, publication bias absence verified by Begg’s and Egger’s tests together with sensitivity analysis confirmed the stability and reliability of results from this study.

AC is nowadays well acknowledged as inhibiting tumor growth and prolonging survival after surgery in CRC patients. Some studies reported that trauma by surgery could promote residual tumor growth and tumor metastasis by releasing growth-stimulating factors and triggering immunosuppression [41, 42]. Studies also have shown that surgical trauma can cause an increase in transforming growth factor α (TGFα), which plays an important role in colorectal tumor invasion and metastasis [43, 44]. In vivo experiments on mouse model confirmed that growth factors in healing wounds may promote tumor growth, affecting therapeutic effect of immunotherapy with interleukin-2 (IL-2) and lympokine activated killer (LAK) cells [45, 46]. Therefore, a prolonged interval between operation and AC may lead to the tumor growth and micro-metastases [9]. Moreover, in the blood of some CRC patients after surgery, circulating tumor cells (CTC) or circulating tumor DNA (ctDNA) were detected, and RFS of these patients was significantly reduced [47‐50]. About 25% of minimal residual disease (MRD) positive patients can achieve ctDNA clearance through adjuvant infusional fluorouracil, leucovorin, and oxaliplatin chemotherapy, making contributions to improve survival [51, 52].

Important gap remains in our knowledge on mechanical association between the timing of AC and prognosis. There are several possible reasons. Firstly, as mentioned above, residual tumor-promoting cytokines and MRD after surgery are possible drivers for tumor recurrence and metastasis. Given that growth rate is higher in early stage, early intervention with AC may suppress the tumor growth and dissemination [53]. Secondly, with classic mathematical model, Goldie et al. demonstrated that aside from mutation rate of tumor cells and sizes, time is also one major contributor to mutation associated drug resistance [54]. Third, Farzaneh et al. found that patients in delayed AC group had a higher proportion of positive surgical margin, one important factor for poor prognosis in locally advanced CRC [38]. Harless and colleagues reported the effectiveness of AC was inversely proportional to the interval between surgery and AC initiation [55]. Finally, therapy delay is usually induced due to poor nutritional and performance statuses of patients, which may also contribute to poor clinical outcomes. Whether AC delay is a cause or a consequence of poor prognosis for these patients remains unknown.

Combined with previous studies, our study verified the inverse association of prolonged interval between surgery and AC and survival of patients. While in clinical practice, situation is more complex as therapeutic toxicity may be maximized owning to patient’s poor immune and performance status after operation. Early postoperative intervention may lead to severe chemotherapy-related adverse events and even death [56]. Therefore, the survival benefit of AC may be time-dependent [10]. In parallel with results from another meta-analysis by Guetz et al. [8], our study also suggested that AC should be started within 8 weeks after surgery. However, little evidence is available on optimal timepoint to start AC after surgery. Our pooled result showed that 5–8 weeks’ interval between AC and surgery did not increase the risk of mortality compared with that less than 4 weeks, and the 5-year OS was similar between two groups. On the other hand, Bos et al. [29] compared the survival of two cohorts, one with 2,950 patients receiving AC 5–6 weeks after surgery and the other with 1,562 patients receiving AC 7–8 weeks after operation, and found that the 5-year OS was slightly higher in the former one (76% vs. 73%). Generally, patients with colorectal surgery need a recovery period of at least 2–4 weeks for wound healing, physical recovery and treatment of postoperative complication. In sum, we hypothesized that 5–6 weeks’ interval may be more reasonable as it may be the optimal choice taking these factors into consideration. Deeper investigations are warranted for further validation.

Until now, the postoperative time point at which AC is no longer beneficial to CRC patients is unclear. In the study performed by Biagi and colleagues, they found that patients starting AC > 12 weeks exhibited better survival compared with patients not receiving AC, and posited a prognostic benefit at 4 or 5 months, which was longer than that commonly recommended in clinical practice [7]. Based on a total of 18,491 patients, Gao’s group showed that the survival benefit of AC was still statistically significant when initiated more than 20 weeks after operation compared with the non-chemotherapy group, thus, AC might be still beneficial with a delay of 5 months [10]. While Turner reported that 489 patients who initiated AC > 24 weeks had a significantly better OS than 20,807 patients who omitted chemotherapy [11]. Therefore, the above mentioned results indicated that AC might still be useful even with a delay of 5–6 months. These results should be confirmed by more studies in the future.

Limitations exist in our study. First, as randomized trial design of postponing AC is neither ethical nor feasible, most of the included studies in this study were retrospective cohort studies, and one was a secondary analysis from a randomized trial. Second, owning to the detailed information availability and limitations of the size and number of studies, we could not perform other subgroup analyses (such as stage III vs. stage II with high-risk features). Third, the cut-offs of time interval among selected studies were different. Besides, with great breakthroughs on CRC postoperative chemotherapy in the last decades, the adjuvant therapy regimen has changed. As 5-fluorouracil-based chemotherapy regimen was mostly used in previous studies while two-drug combination chemotherapy regimen, such as oxaliplatin plus capecitabine, is more commonly used currently. Whether different AC regimens make differences on the choice of initiation time point remains to be determined.

In conclusion, delaying the initiation of AC for > 8 weeks after surgery was significantly associated with poor OS. AC started within 8 weeks after surgery brought more benefits to CRC patients. There were no obvious differences in survival benefits between AC within 5–8 weeks and ≤ 4 weeks. Compared to patients not receiving AC after surgery, a delay of approximately 5–6 months was still useful to improve prognosis.