Pathological complete response after neoadjuvant chemotherapy for rectal cancer with synchronous multiple liver metastases: a report of an unusual case
Discussion
We reported herein a rare case of 13 synchronous liver metastatic lesions from rectal cancer with a complete pathological response after neoadjuvant bevacizumab-containing chemotherapy. Another unusual feature of the case was that the response evaluation on CT after 6 cycles of chemotherapy indicated a progressive disease because a new lesion (the 13th lesion) emerged in the liver.
For patients with initially resectable CRLM, the survival benefit of NAC is still unclear. Potential disadvantages of NAC are the inherent risk of disease progression preoperatively and the risk of liver toxicity. The routine use of NAC for patients with clearly resectable lesions limited to the liver is not recommended owing to a lack of benefit on survival [4].
However, both imaging and pathological responses of liver metastases to preoperative chemotherapy are good prognostic variables [5, 6]. NAC is most beneficial when patients with favorable tumor biology are identified and selected for treatment [6].
A complete radiological response was achieved in 3 to 14 % of patients in the neoadjuvant setting with cytotoxic chemotherapy (oxaliplatin-based or irinotecan-based chemotherapy) for CRLM [7–10]. Even in patients with no metastases observed on imaging, residual macroscopic disease was found in about 25 to 45 % of the patients at the time of operation [11–14]. In patients with no obvious disease at surgery, microscopic cancer was observed in the resected specimen from the site of initial liver metastases in 80 % of the patients. In addition, in patients with no tumors observed and in whom the site of complete response was left in place, in situ recurrence was observed in 74 % of the cases after 1 year [13]. Therefore, a complete response on imaging does not necessarily equate with a complete clinical or pathological response [13, 14]. A pathological complete response was achieved in 2 to 11 % of the patients with only cytotoxic chemotherapy [5, 8, 15].
Bevacizumab is the only anti-angiogenic agent that has been extensively studied in the setting of resectable (or potentially resectable) or unresectable CRLM. Although the complete radiological response rate in patients treated with bevacizumab-containing regimens ranges from 0 to 9 % [16–18], a pathological complete response has been achieved in 8.9 to 16 % of the patients [14, 15, 19]. The discrepancy between radiological and pathological complete response rates may be because of the cytostatic mechanism of bevacizumab. Response evaluation criteria for solid tumors are standard measures used to evaluate the tumor response to treatment; they were developed to assess tumor shrinkage after cytotoxic chemotherapy. However, the criteria may be limited in assessing the response to bevacizumab-containing chemotherapy [19]. Chun et al. demonstrated that CRLMs tend to decrease in size and display unique morphological changes in CT images after bevacizumab-containing chemotherapy. They defined an optimal morphologic response on contrast-enhanced CT images as a change in metastases from lesions with heterogeneous attenuation and thick, irregular borders to homogeneously hypodense masses with a sharp interface between the tumor and adjacent normal liver parenchyma, which in some cases can mimic a cyst [19]. Effective morphological changes correspond to the replacement of tumor cells by fibrosis, pathologically [19].
The optimal timing of primary tumor and liver metastases resection in patients with synchronous resectable CRLM is still controversial. Our strategy has been to perform resection of the primary colorectal tumor first followed by hepatectomy (i.e., staged resection). The patient was treated with NAC for synchronous liver metastases following the resection of primary cancer because the number and size of liver metastases had rapidly increased before the planned hepatectomy.
On contrast-enhanced CT performed after 6 cycles of bevacizumab-containing chemotherapy, all the 12 liver metastases detected before chemotherapy had decreased in size, and radiological findings of the metastases showed effective signs corresponding to the optimal morphologic response [19]. However, we evaluated it as a progressive disease because a new lesion emerged in the liver. The reason why the new lesion, which consequently showed a pathological complete response, was detectable after chemotherapy is unclear. The lesion that was in situ before chemotherapy may have become detectable in CT images owing to its modification to a homogeneous, hypoattenuating lesion with well-defined borders due to the effect of bevacizumab-containing chemotherapy. As a result, the time when the 6 cycles of chemotherapy are completed may be the optimal timing for hepatectomy. As in the present case, it is necessary to evaluate the effectiveness of the treatment carefully when the number of tumors increases during bevacizumab-containing chemotherapy and the optimal morphologic response is identified.
Complete pathologic clearance of all liver metastases after chemotherapy is associated with considerable overall survival, and it is a strong predictor of both prolonged survival and disease cure [5]. Several retrospective studies have demonstrated that in patients with a good pathological response to NAC, adding bevacizumab to the cytotoxic drugs was associated with a better outcome [20–22]. The patient reported herein with more than 10 synchronous liver metastases is alive without recurrence for 4 years since the last operation, and we believe that the disease is cured.