Surgical consolidation of initially unresectable urothelial carcinoma: an incremental opportunity to cure

Despite 20 years of research since methotrexate, vinblastine, doxorubicin and cisplatin (M-VAC) was first reported, the overall prognosis for patients with metastatic or surgically unresectable urothelial carcinoma remains poor, with a median survival of approximately 13–15 months [1,2]. Multiple factors, including a poor performance status and visceral metastases, have been associated with a worse prognosis for this disease [3]. Clinical trials have, by necessity of their design, focused on the median survival as the most important measure of outcome. However, another important measure of potential patient benefit lies in the ‘tail’ of the survival curve, wherein lie patients who may still have long-term survival outcomes. It has been our experience that a large number of these patients achieve this notable long-term survival through the use of surgical consolidation for their initially unresectable urothelial cancer.

Data supporting a role for surgical consolidation of initially unresectable urothelial cancer arose early in the era of cisplatin-based chemotherapy. Logothetis et al. noted that some patients with metastases relapse in the same site as they did initially, without rapid spread of disease elsewhere [4]. Of the 147 patients treated with either cisplatin, cyclophosphamide, and doxorubicin (CisCA) or M-VAC, the authors observed that 74% of patients with disease confined to lymph node groups relapsed (at their first progression) in nodal sites only, with only 26% relapsing in a non-nodal location. This contrasts with the setting of visceral metastases, where patients were more likely to relapse in another metastatic site at their first relapse. Even so, a small number of patients (38%) with visceral metastases, typically in the lung, relapsed in the site of their initial disease without progression in other sites.

A major limitation for this strategy, has been selecting those patients who are most likely to benefit from surgical consolidation following cytoreductive chemotherapy. Some institutions have advocated for the use of surgical consolidation, even in the setting of a partial response, as a method of rendering patients without evidence of disease. Implementing this strategy, Dodd et al. reported that, of 203 patients treated with M-VAC on clinical trials, 30 had surgical resection to remove residual disease [5]. Of those treated with surgery, 33% remain alive after 5 years. The group of patients with node-only disease did better than their counterparts who had resection of visceral metastases.

However, the overall prognosis in patients who have residual tumor at surgery remains quite poor. Neoadjuvant clinical trials, which typically enroll patients at a lower stage, have found only a 10% chance of survival beyond 5 years when cancer remains in the lymph nodes at surgery [6]. The median survival of these patients was 13 months, which is no different from the survival observed on clinical trials for initially metastatic disease.

In an attempt to select for a patient cohort more likely to benefit from surgical consolidation, we began offering this to patients who had a near complete response to chemotherapy. Of the 172 patients treated with cisplatin-based chemotherapy on a clinical trial, 30 had surgical consolidation [7]. In the 20 patients with node-only disease, 11 remained disease free at a median follow-up of 51.8 months, with a median survival of 63.8 months. Only one out of ten patients remained disease free after resection of solitary visceral metastases. In the same time interval, only four patients were alive and disease free from chemotherapy alone.

Given the strong association between pathologic downstaging and patient outcomes, we attempted to maximize response to chemotherapy prior to surgical consolidation. We developed a clinical trial design that incorporates response into the treatment algorithm [8]. At the first 6-week interval of chemotherapy, patients must have had a response of greater than 40% to continue on the same therapy, and at the second 6-week interval, a greater than 90% response, which we would categorize as a major response. When this threshold of response was not met, the patient was re-randomized to alternate chemotherapy. Surgical consolidation was performed in 35 out of the 117 patients treated on this trial. While the majority, 69%, had consolidation of node-positive disease in pelvic or retroperitoneal nodes, 17% had surgery for initially fixed, cT4b tumor and 14% for visceral metastatic sites. The median survival from surgery (not including the time spent on prior chemotherapy) was 20 months, with a 5-year survival of 29%.

The extent of response had a strong impact on patient outcomes. Patients who had surgical consolidation in the setting of a major response had a 40% 5-year survival, compared with a 10% 5-year survival when a major response did not occur (p = 0.04) [8]. Similar results were reported by Nieuwenhuijzen et al.[9]. Of the 52 patients with biopsy proven nodal metastases, 44 received surgical consolidation. In the setting of a complete response the 5-year survival was 42%, compared with a 19% 5-year survival when only a partial response was achieved.

Even in more distant nodal metastases, the extent of response continues to remain of benefit. Of 11 patients with biopsy-proven retroperitoneal node involvement undergoing surgical consolidation following cytotoxic chemotherapy, the 4-year disease specific survival was 36%. Viable tumor in no more than two nodes correlated with an improved survival (p = 0.006) [10].

Resection of visceral metastases remains largely investigational for a highly selected cohort of patients. Traditionally, we have offered it to patients with disease in one visceral site, who respond well to chemotherapy with no evidence of rapid progression elsewhere. The majority of cases have been for resection of lung metastases. Despite the clinical appearance of a solitary lesion, more cancer has typically been found at surgery with multiple wedge resections containing tumor. A 5-year survival of 33% has been reported in 31 patients treated at our institution [11]. Of the five who remain alive and without disease beyond 3 years, four had additional chemotherapy and metastatectomy, while only one has never relapsed with more than 5-years of follow-up.

Chemotherapy remains the standard therapy for the treatment of surgically unresectable or metastatic disease. Surgical consolidation has been reported most often in the setting of aggressive cisplatin-based chemotherapy [5,7,9,11], and more recently with ifosfamide-based combinations [8]. It has been my experience that the use of chemotherapy that does not contain either of these alkylating agents is unlikely to achieve the degree of response necessary to consider surgical consolidation.

Our current strategy is to offer surgical consolidation to patients who have initial node-positive disease as long as the lymph node groups are at or below the renal artery and vein. We will give patients sequential chemotherapy in a strategy that maximizes response by continuing a treatment that is working, and switching to alternate chemotherapy if certain thresholds of response are not achieved. Patients with a major response as defined by more than a 90% reduction of tumor have been our main target for surgical consolidation. If there is obvious residual tumor remaining we will perform a biopsy of the lymph node. If the biopsy confirms the presence of cancer then the surgery will not be pursued. The use of surgical consolidation in visceral metastases is even more highly investigational. Studies have favored this strategy for a highly selected cohort with disease in a solitary visceral site, most frequently reported in lung metastases, with no evidence of rapid progression elsewhere. We have typically been observing these patients with visceral metastases for at least 3 months following cytotoxic chemotherapy to exclude rapidly progressive metastases in other sites, before proceeding with surgical consolidation.

Currently, this strategy has the most impact on patients with node-only disease. In patients with visceral metastases, more research is necessary to determine who will ultimately benefit from this strategy. Further efforts in the treatment of metastatic urothelial cancer need not focus solely on enhancing the median of the survival curve, but should also follow the ‘tail’ of the curve, where one may find patients with long-term control and potential cure of metastatic urothelial cancer.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.