Surgery for Adrenocortical Carcinoma

Abstract

Surgery plays a pivotal role in the management of adrenocortical carcinoma. Unfortunately, there is no clear, widely adopted or shared definition of the technical steps of a correct oncologic resection. This chapter discusses the different technical aspects of an accurate adrenalectomy for cancer, based on the guidelines and the relevant literature. Similarly, it addresses the difficult topic of the surgical strategy to be adopted for recurrent disease, with special focus on the management of tumor-bed and peritoneal recurrences.

9.1 Introduction

A correct adrenalectomy represents the most important prognostic determinant of the clinical course of a patient with adrenocortical carcinoma (ACC) and it should be considered a prerequisite for cure. The definition of curative adrenalectomy is intriguing, as it has not yet been fully delineated. Multiple elements contribute to its achievement; among them, the integrity of the tumor during surgical manipulation, the extent of periadrenal soft-tissue clearance, the role of lymphadenectomy and the surgeon’s expertise. The quality indicator for this surgery is represented by the rate of local/peritoneal recurrence, which is impressive also for early-stage tumors. Attention to every single detail leads to optimization of the surgical treatment of the disease. In this chapter we will describe how a correct adrenalectomy for cancer should be performed, and discuss the surgical strategy for recurrent local or peritoneal ACC.

9.2 Upfront Adrenalectomy: The Guidelines

Among the different guidelines issued in recent years [1,2,3,4], those published in 2018 by the European Society of Endocrinology in collaboration with the European Network for the Study of Adrenal Tumors [1] stand out for their completeness and reasoned development.

Concerning the surgical management of localized ACC, they state [1]:

“R.3.1. We recommend that adrenal surgery for suspected/confirmed ACC should be performed only by surgeons experienced in adrenal and oncological surgery”;

“R.3.2. We recommend complete en bloc resection of all adrenal tumors suspected to be ACC including the peritumoral/periadrenal retroperitoneal fat. We recommend against enucleation and partial adrenal resection for suspected ACC. If adjacent organs are suspected to be invaded, we recommend en bloc resection. However, we suggest against the routine resection of the ipsilateral kidney in absence of direct renal invasion”;

“R.3.4. We suggest that routine locoregional lymphadenectomy should be performed with adrenalectomy for highly suspected or proven ACC. It should include (as a minimum) the periadrenal and renal hilum nodes. All suspicious or enlarged lymph nodes identified on preoperative imaging or intraoperatively should be removed”;

“R.3.7. We recommend perioperative hydrocortisone replacement in all patients with hypercortisolism that undergo surgery for ACC”.

9.2.1 The Literature

Because of the rarity of ACC no prospective randomized studies exist and the literature offers only retrospective, often multicentric studies.

9.2.1.1 R.3.1

Adrenal surgery has low mortality and morbidity rates. A review of 9820 procedures by the French Association of Endocrine Surgeons reported a 30- and 90-day mortality rate of 0.8% and 1.5%, respectively. Mortality was unevenly distributed according to the hospital case-load: 1% vs. 0.4% (30-day) and 1.8% vs. 0.9% (90-day) in low- and high-volume institutions, respectively. At multivariate analysis the hospital case-load was independently associated with operative mortality (OR: 1.8, p < 0.010), along with other risk factors such as age, comorbidity, malignancy, open surgery and reintervention, and this was highly appreciable in high-risk patients (p = 0.003); a case-load of 32 patients/year was indicated as the best cut-off to recognize high-volume hospitals [5]. Similarly, considering a range of short-term indicators, a volume-outcome effect was recognized in England [6]; the benefits of centralization became appreciable above 10 procedures per year, with greater advantages for patients at the threshold of 20 adrenalectomies/surgeon/year and 30 per institution. In the Netherlands, the centralization of adrenal surgery led to an impressive improvement of survival: from 78% to 93% and from 42% to 63% after 1 and 5 years, respectively, for patients receiving potentially curative adrenalectomies (stage I–III) [7]. Panelists of the ESE-ENSAT guidelines [1] suggest a minimum of >20 adrenalectomies/year for the surgical treatment of primary malignancies; these should be performed only in referral institutions with dedicated adrenal tumor boards, considering the complexity of this surgery. These requirements also apply to those cases in which the clinical diagnosis of malignancy is not certain but deemed possible. Dedicated mentorship should be pursued by those interested in adrenal surgery.

9.2.1.2 R.3.2 and R.3.4

Curative surgery is a prerequisite for cure, and for this reason the surgeon should guarantee the best procedure, considering all the different factors that measure the quality of adrenal surgery.

It has long been appreciated that tumor effraction and positive margins negatively affect oncologic outcomes, favoring local/peritoneal recurrence, an expression of tumor seeding and tumor persistence. Tumor integrity and negative resection margins concur in the definition of curative adrenalectomy. Case series of patients operated on during the first decade of this century reported 5-year recurrence-free survival rates around 30% and 15%, and 5-year overall survival rates of 60–95% vs. 15–65% in the case of negative or positive margins, respectively [8,9,10]. Removal of the adrenal gland en bloc with the renal capsule and retroperitoneal connective tissue reduces the risk of tumor rupture and increases the rate of margin-negative resections. Margin status has such a relevant impact on survival that the panelists of the ESE-ENSAT guidelines recommend: “R.3.6. If the first surgery was suboptimal and macroscopically incomplete (R2 resection), we suggest to discuss repeat surgery in a multidisciplinary expert team”. Surgery should be indicated if the residual tumor is detectable at cross-sectional imaging, and a curative R0 surgery is deemed possible.

Lymph node involvement affects local recurrence and survival. In recent series nodal metastases were reported in about one-third of cases subjected to lymphadenectomy [9, 11,12,13]. Despite this, lymphadenectomy is performed in 20–30% of cases and almost exclusively in academic referral centers. The reason for this is the lack of a clear, widely adopted and shared anatomically based surgical strategy dictating the dissection rules to which surgeons should adhere. In fact, a thorough evaluation of different series shows that the extent of lymphadenectomy is widely inhomogeneous, ranging from systematic removal of locoregional nodes to nodal sampling. Experts agree on a minimum number of 4–5 nodes to define a real lymphadenectomy [11, 12]. There is also consensus on considering the right paracaval and left para-aortic nodes, together with those at the ipsilateral renal hilum as first-level lymphatic stations. A clear description of the attitude toward lymphadenectomy in ACC was provided by the German ACC study Group: 47/283 stage I–III patients (16.6%) received lymphadenectomy, which was more likely to be performed in the case of larger tumors, stage III tumors and multivisceral resections. Median follow-up was 59 months for the nodal dissection group and 39 months for the non-dissection group; at these time points recurrence and disease-specific death rates were similar (68.1 vs. 60.6 and 29.8 vs. 30.5, respectively) between the groups, but time to recurrence was longer after lymphadenectomy (20.1 vs.12.8 months); the presence of lymphatic metastases had a negative prognostic role as disease-free and disease-specific survival were both shorter (12.5 vs. 31.3 months, p = 0.058, and 86.4 vs. 135 months, p = 0.002, respectively). Interestingly, considering the node-positive patients, after adjustment for age, tumor stage, multivisceral resection, adjuvant treatment and nodal status on preoperative imaging, lymphadenectomy resulted in a significant reduction of the risk of recurrence (HR 0.65; 95% CI 0.43–0.98, p = 0.42) and of disease-specific death (HR 0.54; 95% CI 0.29–0.99, p = 0.049), and this was particularly evident for stage III ACC. Gerry et al. confirmed the German Group’s oncologic outcomes and detailed how lymphadenectomy has a limited impact on the postoperative course [14]. A third study based on 386 patients from the National Cancer Database [15] reached substantially similar results but also showed that the strongest prognostic factor for poor survival was the number of positive nodes, which showed a progressive effect: the hazard ratio was 2.3 (CI 1.5–3.6) for 1 node, 3.0 (CI 1.1–8.0) for 3 nodes, and 4 (CI 2.5–6.2) for ≥4 nodes. A different analysis of the same database [13] added some more information: lymph nodes were retrieved more frequently in open than in minimally invasive surgery, and the percentage of nodal metastases was proportional to the number of nodes examined. On these bases it is now clear that lymphadenectomy should complete any adrenalectomy for ACC. Interestingly, however, the panelists of the ESE-ENSAT guidelines [1] do not support surgical radicalization if a margin-negative adrenalectomy has been performed, as in the case of misdiagnosis at preoperative work-up; they consider the harm/benefit balance to be uncertain and prefer to support the early start of adjuvant therapy.

It is important to bear in mind that nephrectomy to facilitate the lymphadenectomy or to reach a “safer and easier margin” is abusive [16]. Nephrectomy is only admitted in the case of direct infiltration of the renal parenchyma, and, if this is minimal, nephron-sparing strategies should be considered. In fact, a preserved renal function may have an important role during the subsequent course of the disease, facilitating systemic treatments. There is no demonstration of a positive impact from the resection of neighboring organs (spleen, pancreas, colon, stomach, liver) if these are not infiltrated; indeed, these procedures increase the risk of complications and mortality.

9.2.1.3 R.3.7

Intra- and postoperative glucocorticoid replacement must be provided to all patients with adrenal autonomous cortisol secretion at the doses suggested for major stress by the guidelines [17]. The surgeon must consider that glucocorticoid deficiency also affects the coagulation cascade and may cause delayed postoperative hemorrhages. Since a critical phase for such deficiency is typically observed at the switch from parenteral to oral administration, it is safe to overlap the two routes for some days.

9.2.2 Reasoning to Establish a Sound Surgical Strategy

As for other organs, a curative resection of ACC should remove the entire area of embryonal development, deploying all efforts to maintain intact its envelope, and sectioning all vascular, lymphatic and nervous structures at their origin/confluence. This is the rationale at the basis of a curative D2 gastrectomy, of a total mesorectal excision for rectal cancer, of a complete mesocolic excision for colonic cancer or radical antegrade modular pancreatosplenectomy for pancreatic cancer. However, a major difference exists: the absence of a complete mesothelial envelope to define the surgical target. The complexity of the adrenal vascular and lymphatic system reflects the gland’s peculiar embryogenesis, both ectodermal (medulla) and mesodermal (cortical). Accordingly, embryonal development follows two different axes: horizontal, from the celiac ganglia, and ascendant, from the renal hilum. Since the two embryological components share the vascular and lymphatic systems, the extension of the area to be dissected is relatively large. The lymphatic system is mainly posterior, toward the celiac trunk and renal hilum and finds its medial limit along the right or left side of the aorta with prolongation up to the renocaval confluence for left-sided tumors. In the case of large tumors causing caudal compression of the kidney or stage III ACC with nodal involvement blocking the main lymphatic flow, an anterior lymphatic route to the interaortocaval nodes can also be activated as well as a cranial one, along the diaphragmatic vessels with possible direct flow to the thoracic duct and the mediastinum. The cranial limits of this “embryonal envelope” are represented by the diaphragmatic crux, while the caudal limits are represented by a transverse line from the caudal limit of the renal hilum to the aorta.

Some studies tracked the frequency of nodal metastatic involvement also in relation to different pathologic details such as disease stage [18, 19] and recognized that the right paracaval and left para-aortic nodes along with those at the renal hilum represent the first level of lymphatic drainage. In turn, these converge to the interaortocaval nodes which represent the second lymphatic level. Nodal involvement beyond the second level should be considered metastasis and the therapeutic strategy should be decided accordingly at multidisciplinary discussion. Interestingly, lymphatic metastases do not cross the aortic midline unless in case of very advanced stages or of previous excision of the interaortocaval nodes (recurrent disease). The involvement of anterior nodes such as those at the hepatic hilum is anecdotal. Bearing in mind these theoretical bases, it is possible to modulate the surgical strategy according to the clinical stage.

It seems wise to perform a “cavity” adrenalectomy for stage I ACC. This implies the en bloc removal of the adrenal gland with the cranial part of the renal capsule (anterior and posterior) and the retroperitoneal connective tissues. The caudal limit of this dissection is represented by the renal vein, the cranial limit by the diaphragm and the posterior one by the muscular plane. Lymphadenectomy includes the right posterior paracaval or the left para-aortic nodes along with those at the renal hilum, including the nodes positioned caudally and posterior to the renal vein and artery. When the tumor is small, the anterior dissection plane is generally easy to achieve maintaining a safety margin, and dissection from the liver, spleen and pancreas is not demanding. It is possible to perform this procedure with minimally invasive techniques, but the nodal dissection is easier with the robotic approach. Considering the confluence of the cited lymphatic structures into the main longitudinal lymphatic axis, attention should be paid to ligate or clip all longitudinal lymphatic structures.

For stage II–IV ACC we suggest a “regional” adrenalectomy. This consists of the removal of the adrenal gland en bloc with the entire renal capsule, of the lymph nodes at the renal hilum and of all retroperitoneal soft tissues surrounding the vena cava, including the aortocaval space which is dissected up to the aortic midline, from the diaphragm down to the inferior mesenteric artery, with exposure of the right aspect of the celiac trunk and superior mesenteric artery (right adrenalectomy, Figs. 9.1, 9.2, and 9.3). In the management of left-sided ACC, the medial limit of soft tissue clearance is the aortic midline, with exposure of the left side of visceral arteries; the clearance of interaortocaval lymphatic tissues is limited to those surrounding the left renal vein (Fig. 9.4).

Fig. 9.1

View of surgical field after right regional adrenalectomy

Fig. 9.2

Same patient of Fig. 9.1, detail of renal hilum after regional adrenalectomy

Fig. 9.3

Same patient of Figs. 9.1 and 9.2, detail of renal vessels after regional adrenalectomy

Fig. 9.4

View of surgical field after left regional adrenalectomy

A large medialization of the liver or of the spleen-pancreas block is always the preliminary part of the procedure. Generally, it is advisable to perform the posterior dissection before dissection from the liver, spleen or pancreas, which is more easily performed once the entire block is better exposed. In the case of right-sided tumors we discourage dissection of the adrenal mass from the liver in favor of a resection of the liver, pursuing at least a sub-glissonian plane in the quest for radicality. The same is impossible for left-sided tumors, in which case dissection of the spleen and pancreas along with the splenic vessels should be meticulous to preserve tumor and margin integrity. Owing to their complexity, we perform these procedures in open surgery, through a subcostal incision. However, medialization of the spleen and pancreas may be extremely difficult in the case of large left-sided tumors as the spleen is often in a posterior position due to the tumor growth subtending the pancreas. In these conditions, it is wise to start with a laparoscopic approach with the patient in a right lateral decubitus: in this position the phrenosplenic ligament lays in an anterior position and is easily and completely dissected; the patient is then turned into a supine position and the procedure converted to open: the spleen will be free to follow all subsequent manipulations without any risk of capsule laceration. Resection of the surrounding organs such as the kidney, spleen or pancreas is only performed in the presence of direct infiltration.

9.3 Surgery for Recurrent Disease: The Guidelines

Unfortunately, recurrence is a common finding after surgery for ACC. It can present in different patterns: in the form of metastatic hematogenous disease (see also Chaps. 15 and 16) and in the form of tumor-bed or peritoneal recurrence.

On this subject the ESE-ENSAT guidelines [1] state the following:

R.8.5. In patients with recurrent disease and a disease-free interval of at least 12 months, in whom a complete resection/ablation seems feasible, we recommend surgery or alternatively other local therapies.

9.3.1 The Literature

Recurrence is frequent after surgery for ACC: a recent paper from the University of Michigan [20] reported a 70% recurrence rate in a series of 354 patients operated on for stage I–III ACC; recurrence was observed after a median disease-free survival (DFS) of 11 months. Surgery plays a pivotal role in the management of recurrence. In 1999, Schulick and Brennan, reporting the experience of the Memorial Sloan Kettering Cancer Center, focused on a group of 47 patients re-operated on (some of them multiple times, for a total of 83 re-resections). They showed that curative surgical treatment of recurrent ACC, both in the form of metastasectomy (lung, liver etc.) and/or ablation of tumor-bed recurrence and peritoneal metastases had a dramatic impact on survival: median survival was 74 months for those undergoing a R0 resection versus 16 months after incomplete resection; they also reported a 30-day mortality rate of 3.6% [21].

In the study of Glenn et al. [20], tumor-bed and peritoneal recurrence was detected in 28% and 20% of cases, as single-site recurrence or in the context of multiple-site recurrence in 11% and 17% of cases and in 6% and 14%, respectively; a metastatic pattern to parenchymal filters was observed in 52% of cases. Laparoscopic adrenalectomy was a risk factor for peritoneal metastases, and postoperative radiotherapy showed a protective effect on tumor-bed recurrence. A non-surgical treatment was chosen for 142 patients; they more likely had multiple metastases in the same organ, multiple-site recurrence and a short DFS (median 8 months). One hundred patients were operated on for recurrence; in this case after a median DFS of 17 months. The second operation was deemed R0 in 80% of cases but 79 patients recurred again after a median DFS of 6 months; tumor-bed and peritoneal metastases were exposed to a higher risk of further recurrence which, in turn, was observed in the same site in 67% of cases. Despite this, surgical management of recurrence had a positive impact on survival, which was appreciable up to the third operation. The German ACC Study Group studied 154 recurrent patients and reached similar conclusions [22]. They identified DFS after adrenal surgery >12 months and curative resection of recurrence as the most important predictors of survival. Colleagues from the University of Turin (Italy) stratified 106 patients with ACC recurrence, as having a unique (35% of cases), multiple within a single organ (20.8%) or multiple-site (43%) recurrence. Locoregional treatments were used in 100%, 68% and in 26% of patients in the three groups, respectively; these included surgery (86%), radiotherapy or radiofrequency ablation (9%) or multiple treatments. They reported a disease-free status in 60% of treated cases and a subsequent DFS of 15 months. The best survival results were achieved in the case of single lesion subjected to locoregional treatments [23].

9.3.2 Reasoning to Establish a Sound Surgical Strategy

Surgery for recurrent disease requires an eclectic approach due to the different possible scenarios. Resection of hepatic and lung metastases follows the specific general rules of liver and lung surgery for metastatic disease and their integration with ablative treatments such as radiofrequency ablation and radiotherapy; although these treatments will not be discussed in this chapter, we consider that the adrenal surgeon should be proficient in hepatic surgery. If this is not the case, the surgical team should include a hepatic surgeon. In the same way, lung metastases are normally treated by thoracic surgeons, in a two-step procedure if multiple-site metastases (thoracic and abdominal) are deemed resectable with curative intent. Peritoneal metastases are managed following the rules of peritonectomy for peritoneal diffusion of other primaries such as ovary, colonic or gastric cancer. Sacrifice of solid organs (liver, kidney, spleen, left pancreas, uterus and adnexa) and bowel is performed as required; in our experience, the extent of peritonectomy varies according to the clinical presentation and the technical approach to adrenalectomy. In fact, peritoneal metastases after open surgery require less extensive peritonectomies than those observed after minimally invasive surgery. The Peritoneal Cancer Index (PCI) should be always assessed at the beginning of the procedure and the Completeness of Cytoreduction (CC) should be measured at the end. It is our practice to extend these measures to tumor-bed recurrences even though, in strict terms, these are not peritoneal metastases.

Concerning the management of tumor-bed recurrence, two possible scenarios exist: the recurrence is expression of tumor seeding following a correct oncological resection, or, alternatively, it is expression of tumor persistency due to incomplete primary resection, with or without tumor seeding. In the first case the surgical target is represented by all tumoral nodules detected at cross-sectional imaging; resection of adjacent infiltrated organs is mandatory. Dissection of recurrent tumor from adjacent organs carries an extremely high risk of recurrence and should be contemplated only if the required resection exposes the patient to a disproportionate risk (i.e., pancreatic resection in a high-risk, fragile patient). In any case the resection should achieve tumor-free margins. In the second case, the re-operation should clear all nodules detected at imaging and achieve the required regional soft-tissue and lymphatic clearance not performed at first instance. In any case, these procedures require the complete mobilization of the liver and of the spleen-pancreas block; this latter may be facilitated by the need to resect the spleen and/or the left pancreas due to tumor invasion. The lymphadenectomy may also be facilitated by resection of other involved organs such as the kidney (Fig. 9.5). In all referral institutions these procedures are conducted in open surgery, with preference for a median laparotomy if peritoneal metastases are suspected.

Fig. 9.5

View of surgical field after enlarged regional demolition including nephrectomy, left pancreatectomy, splenectomy, and left colectomy for recurrent adrenocortical carcinoma