Borderline Resectable Adrenal Cortical Carcinoma: A Potential Role for Preoperative Chemotherapy
Brian K. Bednarski . Mouhammed Amir Habra . Alexandria Phan . Denai R. Milton . Christopher Wood . Nicholas Vauthey . Douglas B. Evans .
Matthew H. Katz . Chaan S. Ng . Nancy D. Perrier . Jeffrey E. Lee . Elizabeth G. Grubbs
Published online: 11 March 2014 @ Société Internationale de Chirurgie 2014
Abstract
Background Adrenal cortical carcinoma (ACC) may have tumor or patient characteristics at presentation that argue against immediate surgery because of an unacceptable risk of morbidity/mortality, incomplete resection, or recurrence. This clinical stage can be characterized as borderline resectable ACC (BRACC). At present, systemic therapies in ACC can reduce tumor burden in some patients, creating an opportunity in BRACC for a strategy of preoperative chemotherapy (ctx) followed by surgery.
Materials and Methods A single-institution retrospective review was conducted of all patients considered for surgery for primary ACC. Patients with BRACC treated with pre- operative ctx were categorized as follows: group A, imaging suggesting a need for multiorgan/vascular resection; group B, imaging suggesting potentially resectable oligometasta- ses; and group C, patients having marginal performance status/comorbidities precluding immediate surgery. Both the disease-free survival (DFS) and the overall survival (OS) were compared in BRACC patients treated with preoperative ctx+surgery and those who had upfront surgery.
Results Fifty-three patients with primary ACC were con- sidered for surgery (median follow-up: 49.9 months). Thirty- eight patients (71.7 %) had initial surgery and 15 of them (28.3 %) were considered BRACC and received preoperative therapy. Of these 15 patients, 12 (80 %) received combination therapy with mitotane and etoposide/cisplatin-based ctx, 2 (13 %) received mitotane alone, and 1 (7 %) received ctx alone. Six patients were defined as group A, 5 as group B, and 4 as group C. Thirteen (87 %) BRACC patients underwent surgical resection. BRACC patients were younger but had more advanced disease than the patients having initial surgery (stage IV in 40 vs 2.6 % [p < 0.01]). By Response Evaluation Criteria In Solid Tumors criteria, 5 patients (38.5 %) had a partial response, 7 (53.8 %) had stable disease, and 1 (7.7 %) had disease that progressed. Postoperative mitotane use was similar between groups (p = . 15). Median DFS for resected BRACC patients was 28.0 months [95 % confidence interval (CI), 2.9-not attained] vs 13 months (95 % CI, 5.8-46.9) (p = 0.40) for initial surgery patients. Five-year OS rates were also similar: 65 % for resected BRACC vs 50 % for initial surgery (p = 0.72).
B. K. Bednarski · N. Vauthey · M. H. Katz ·
N. D. Perrier . J. E. Lee . E. G. Grubbs ☒
Department of Surgical Oncology, University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1484, Houston, TX 77030, USA
e-mail: eggrubbs@mdanderson.org
M. A. Habra
Department of Endocrine Neoplasia and Hormonal Disorders, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
A. Phan
Department of Medical Oncology, Methodist Cancer Center, Houston, TX 77030, USA
D. R. Milton
Department of Biostatistics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
C. Wood
Department of Urology, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
D. B. Evans
Department of Surgery, Medical College of Wisconsin, Milwaukee, WI 53226, USA
C. S. Ng Department of Radiology, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
Conclusions The favorable outcome of patients with BRACC, despite more advanced stage of disease compared to those treated with surgery first, together with uncommon disease progression, suggests a benefit of neoadjuvant treatment sequencing in patients with BRACC.
Introduction
Adrenal cortical carcinoma (ACC) is a rare but aggressive malignancy with a poor prognosis [1-4]. While there have been advances in pharmacologic and chemotherapeutic treatment of this disease, initial complete surgical resection remains the dominant therapeutic influence on disease recurrence and survival [5]. Unfortunately, these primary tumors often present with radiographic evidence of locally advanced disease in which technical resectability is not assured, despite the need for multi-organ resection. In addition, for patients with indeterminate or questionable synchronous metastatic disease at diagnosis or those with a poor performance status secondary to tumor hormone excess or other factors, an initial surgical approach is unattractive. Perhaps most important is the effort to avoid early postoperative recurrence after an operation that is of significant duration, complexity, and potential morbidity; in such cases, when early recurrence develops, the patient and clinical care team are forced to acknowledge that the treatment was likely worse than the disease. Historically, given the lack of available alternatives, such patients have been allocated either to undergo an immediate attempt at surgical treatment, with the high associated risks of incomplete resection, rapid disease progression, or post- operative morbidity, or, alternatively, receive systemic therapy with no consideration of future surgical interven- tion. The results of the European non-randomized multi- institutional comparison study of adjuvant mitotane estab- lished this treatment as an option for the management of patients following surgical resection of ACC [6]. In addi- tion, the recent report of results of the First International Randomized trial in locally advanced and Metastatic Adrenocortical Carcinoma Treatment (FIRM-ACT) ran- domized phase III trial have identified the combination of etoposide, doxorubicin, and cisplatin together with mito- tane as the standard of care in the treatment of patients with advanced ACC [7]. These reports taken together suggest that systemic therapy is effective in patients with ACC, and we suggest in the present report that preoperative (neoad- juvant) treatment sequencing may be reasonable in a subset of patients with ACC, particularly in those at highest risk for surgery, or early postoperative recurrence or metastasis.
In the treatment of pancreatic adenocarcinoma at our and select other institutions, a subset of patients with similar clinical and radiographic characteristics has been
described as having borderline resectable disease [8]. In this patient population a multidisciplinary approach to include neoadjuvant systemic chemotherapy and often chemoradiation followed by surgical resection has been associated with relatively favorable outcomes. This treat- ment algorithm has since been adopted by most published guidelines, including the NCCN [9]. Noting the biologic and clinical similarities between patients with borderline resectable pancreatic adenocarcinoma and selected patients with ACC, our multidisciplinary team has applied a similar neoadjuvant treatment algorithm to ACC patients with borderline resectable disease. The objectives of this report were to present our classification system for patients with borderline resectable ACC and to evaluate the management and outcome of patients defined as falling into one or more of these borderline resectable categories compared to patients with resectable disease who underwent upfront surgery.
Materials and methods
After obtaining approval from our institutional review board, a retrospective review was performed of a pro- spectively maintained adrenal database within the Depart- ments of Surgical Oncology and Endocrine and Hormonal Disorders at the University of Texas MD Anderson Cancer Center. The patient cohort for this study consisted of all patients with an intact primary tumor who were considered surgical candidates and underwent initial treatment at our institution. Patients deemed to not be candidates for eventual surgical intervention, such as those with wide- spread metastases at presentation, were excluded. The database was reviewed to obtain patient and tumor char- acteristics, stage of disease, neoadjuvant and adjuvant therapy, response to therapy, surgical outcomes, and time to recurrence or death.
Clinical stage of the tumor was determined using the European Network for the Study of Adrenal Tumors (ENSAT) staging system [10]. All patients treated with neoadjuvant therapy were considered borderline resectable. These neoadjuvant patients were further categorized ret- rospectively into three groups based on the primary indi- cation for preoperative therapy. Group A patients were considered to be borderline resectable due to the perceived requirement for multi-organ or vascular resection and high risk for a margin-positive resection based on preoperative imaging review. Such findings included direct invasion of the tumor into the liver, the wall of the inferior vena cava or renal vein, or the presence of major tumor thrombus within the vein lumen. Additional anatomic criteria for being considered borderline resectable included the abut- ment or involvement of unresectable vascular structures,
such as the celiac axis or superior mesenteric artery, again suggesting a very high risk for a margin-positive resection. Group B patients were considered borderline resectable due to radiographic suspicion for indeterminate/question- able metastatic disease, or demonstrable evidence for potentially resectable oligometastases. Group B patients could have a resectable or a borderline resectable primary tumor (as defined in group A). Finally, patients were cat- egorized as group C if they had the presence of significant comorbidities that argued against immediate surgical intervention, such as the new diagnosis of a pulmonary embolus, or had poor performance status secondary to hormone overproduction. Group C patients were thought to have potentially reversible risk factors, to include hyper- cortisolism that could be treated with steroidogenesis inhibiting agents. Group C patients could have a resectable or borderline resectable primary tumor (as defined in group A) and/or the presence of potential or definite resectable metastatic disease (as defined in group B). In summary, a patient’s radiographic and clinical findings could warrant inclusion into more than one borderline group, but for the purposes of this analysis, they were assigned to a single group and were categorized by the most severe indication (group C > group B > group A) [8].
Borderline resectable patients underwent neoadjuvant treatment with mitotane and/or systemic chemotherapy, based on individual physician preference. Restaging stud- ies were obtained frequently during the course of treatment to monitor response to preoperative therapy. Typically, patients had imaging studies following every two cycles of therapy. Tumor response to preoperative therapy was evaluated on the basis of Response Evaluation Criteria in Solid Tumors (RECIST) criteria and categorized as partial response (PR), stable disease (SD), or progressive disease (PD) [11].
Disease-free survival (DFS) was calculated from the date of surgery until disease recurrence, death without documented recurrence, or the last follow-up date. Disease recurrence was defined as radiographic evidence of loco- regional or distant metastatic disease. Biopsy was not required to confirm recurrence. Two patients were exclu- ded from DFS analysis because they had metastatic disease at the time of initial surgical resection. Overall survival (OS) was measured from the date of diagnosis until death.
Statistics
Comparison of categorical variables between groups was performed using Fisher’s exact test. Continuous variables were assessed as a comparison of medians using the Wil- coxon rank sum test. Statistical evaluation of the stage of disease between the neoadjuvant group and the initial surgery group was conducted with the Mantel-Haenszel
Chi square test. Finally, DFS and OS were estimated using Kaplan-Meier survival curves and the log rank test. All statistical analyses were performed with SAS 9.3 for Windows (SAS Institute Inc., Cary, NC). A p value of 0.05 was used to determine significance.
Results
Patient characteristics
Upon review of all cases of ACC managed in the depart- mental database from 1995-2012, we identified 53 patients who presented with intact primary tumors and were treated with curative intent to include surgical resection. Of those 53 patients, 15 were given neoadjuvant therapy, and 38 underwent upfront surgery. For the purposes of this report, the 15 patients who received neoadjuvant therapy will be referred to as the borderline resectable ACC (BRACC) group. The remaining patients will be referred to as the resectable group.
While there was no statistical difference between the treatment groups with regard to gender, laterality of pri- mary, or incidence of hormonally functional tumors, the BRACC group was significantly younger, with a median age of 40 (20-70.3) years, compared to those undergoing upfront surgery, with a median age of 55.5 (23.7-86.6) years (p = 0.023) (Table 1). Not surprisingly, the BRACC patients had a significantly higher incidence of advanced stage of disease at presentation when compared to the patients in the resectable group (40 vs 2.6 % stage IV;
| BRACC (received neoadjuvant chemotherapy) n = 15 (%) | Resectable (underwent upfront resection) n = 38 (%) | p value | |
|---|---|---|---|
| Female gender | 9 (60) | 22 (57.9) | 1.00 |
| Median age, years (range) | 40 (20-70.3) | 55.5 (23.7-86.6) | 0.023 |
| Right side primary | 9 (69.2) | 17 (44.7) | 0.20 |
| Stage at presentation | <0.001 | ||
| I | 0 (0.0) | 3 (7.9) | |
| II | 2 (13.3) | 17 (44.7) | |
| III | 7 (46.7) | 17 (44.7) | |
| IV | 6 (40.0) | 1 (2.6) | |
| Hormone function | 8 (53.3) | 18 (47.4) | 0.77 |
| Median tumor size, cm (range) | 14.0 (6.6-24.0) | 10.5 (4.0-23.0) | 0.003 |
p < 0.001) (Table 1). Additionally, the median size of the primary tumor as measured by preoperative imaging was also larger in the patients treated with neoadjuvant therapy (14.0 vs 10.5 cm; p = 0.003) (Table 1).
Neoadjuvant therapy
Based on the information in the medical charts and review of preoperative imaging results, patients who received neoadjuvant therapy were categorized as borderline resectable according to anatomic criteria (group A), pre- sence or suspicion of metastatic disease (group B), or diminished performance status or comorbidities that pre- cluded immediate surgical intervention (group C). Using these criteria, the 15 patients with BRACC were allocated as follows: 6 (40 %) patients were categorized as border- line resectable group A, 5 (33.3 %) as group B, and 4 (26.7 %) as group C (Table 2). Individual detailed clinical data for the BRACC patients are shown in Table 3. For the 6 patients in group A, 1 patient had an inferior vena cava (IVC) thrombus to the level of the atrium in addition to suspected liver invasion, 2 patients had tumors abutting the celiac axis, and the remaining 3 patients were considered borderline resectable secondary to suspected combined kidney and liver involvement in which a margin-negative resection would be difficult to achieve. Patients categorized as group B included one with documented liver metastases, two with documented lung metastases, and two with indeterminate liver or lung nodules. Finally, two of the four patients categorized as group C had newly diagnosed pul- monary emboli precluding immediate surgery, one had Budd-Chiari syndrome with documented liver dysfunction secondary to tumor thrombus, and one was malnourished and deconditioned resulting in a performance status argu- ing against upfront surgical therapy. One of the two patients diagnosed with acute pulmonary emboli also manifested inclusion criteria for groups A and B, while the second patient fit no other group. Both patients underwent 6 months of anticoagulation therapy prior to surgery while receiving chemotherapy and had documented Eastern Cooperative Oncology Group (ECOG) scores of 1 both before and after therapy. The patient with Budd-Chiari syndrome also fit group A criteria and exhibited ECOG score improvement from 2 to 1 after chemotherapy. The malnourished patient also fit group A criteria, and her initial estimated ECOG score of 3 worsened to 4 during treatment with chemotherapy. All four group C patients also had hypercortisolism requiring treatment with ster- oidogenesis-inhibiting agents.
Within the BRACC group, 12 patients received combi- nation systemic therapy that included mitotane as well as chemotherapy. The chemotherapy regimens included eto- poside/adriamycin/cisplatin in 8 patients, etoposide/
| Neoadjuvant therapy (n = 15) | |
|---|---|
| Neoadjuvant mitotane | 14 (93.3) |
| Neoadjuvant chemotherapy | 13 (86.7) |
| Borderline category | |
| A: Primary tumor anatomy | 6 (40.0) |
| B: Concern for metastases | 5 (33.3) |
| C: PS/comorbidity | 4 (26.7) |
| Response (RECIST) | |
| Partial response | 5 (33.3) |
| Stable disease | 7 (46.7) |
| Progression | 1 (6.7) |
| Unknown | 2 (13.3) |
| Surgical resection | 13 (86.7) |
PS performance status; RECIST Response Evaluation Criteria in Solid Tumors
cisplatin in 3 patients, and taxol/carboplatin/etoposide in 1 patient. Two patients received mitotane alone, and a single patient received etoposide and cisplatin without mitotane (Table 2). The median duration of neoadjuvant systemic therapy was 4.2 months (range: 0.9-8.1 months), with a median of 5 cycles administered (range: 2-10 cycles). The estimated ECOG performance status remained stable dur- ing chemotherapy in 9 patients, improved in one patient, and worsened in two others. While expected grade 1 adverse effects occurred with the administration of che- motherapy, only 2 of the 13 patients had to stop systemic chemotherapy because of toxicity; one for grade 2 neu- ropathy and the other for intractable nausea and failure to thrive. The median duration of mitotane administration was 5.3 months (range: 2.0-10.1 months) with median serum mitotane levels achieved of 7.1 mcg/ml (range: 2.0- 20.0 mcg/ml) in 12 of the 14 patients for whom these data were available.
Eight patients in the BRACC group experienced hy- percortisolemia secondary to tumor hypersecretion of cor- tisol, and seven of them received mitotane therapy, in part to control excess hormone production. Three of these seven patients required additional steroidogenesis inhibition with metyrapone. All eight patients had biochemical and/or clinical improvement from their hypercortisolism at the completion of neoadjuvant therapy.
The response to neoadjuvant therapy was evaluated on the basis of the RECIST criteria. Two patients were excluded secondary to the absence of appropriate quality radiographic images for review. Of the 13 patients with imaging data permitting RECIST assessment, only one patient had pro- gression (7.7 %), while 5 (38.5 %) had a partial response (Figs. 1, 2) and 7 (53.8 %) had stable disease. In 6 of 7 patients with vena caval tumor thrombus, a partial response
| Patient | BRACC category | ENSAT stage | Preop ctx | Length of preoperative ctx (months) | Preoperative mitotane | RECIST response | Multi-visceral resection | Margin status | Adj. Mitotane | Follow- up status | Survival (months) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | A, B, C | 4 | T/C/E | 6 | Yes | PD | Yes | R0 | No | Dead | 38 |
| 2 | A, B | 4 | E/A/P | 4 | Yes | SD | Yes | R0 | No | ANED | 78 |
| 3 | B | 4 | Yes | – | Yes | R0 | No | Dead | 29 | ||
| 4 | A | 3 | E/P | 8 | YES | SD | Yes | R0 | Yes | ANED | 77 |
| 5 | A | 2ª | E/P | 1 | No | SD | No | R0 | No | ANED | 114 |
| 6 | A | 3 | Yes | SD | Yes | R0 | Yes | ANED | 41 | ||
| 7 | A | 3 | E/A/P | 8 | Yes | – | Yes | R0 | No | ANED | 85 |
| 8 | B | 4 | E/C | 4 | Yes | PR | No | R0 | No | AWD | 11 |
| 9 | A, B | 4 | E/A/P | 4 | Yes | PR | Yes | R1 | Yesb | AWD | 40 |
| 10 | C | 3 | E/A/P | 4 | Yes | PR | Yes | R0 | Yes | Dead | 26 |
| 11 | A,C | 3 | E/A/P | 5 | Yes | PR | Yes | R0 | No | Dead | 7 |
| 12 | A | 3 | E/A/P | 4 | Yes | PR | Yes | R0 | Yes | ANED | 73 |
| 13 | A, B | 4 | E/A/P | 5 | Yes | SD | Yes | R0 | Yes | ANED | 19 |
| 14 | A,C | 3 | E/P | 1 | Yes | SD | – | – | – | Dead | 5 |
| 15 | A | 2ª | E/A/P | 2 | Yes | SD | – | – | – | Dead | 9 |
T/CIE taxol/carboplatin/etoposide; E/A/P etoposide/Adriamycin/cisplatin; E/P etoposide/cisplatin; PD progressive disease; SD stable disease; PR partial response; ANED alive with no evidence of disease; AWD alive with disease
a Celiac axis abutment
b Mitotane + E/A/P
A
SPR
B
(at least a 30 % decrease) of the venal caval tumor thrombus was documented during neoadjuvant treatment (Fig. 2). Importantly, the one patient who met RECIST criteria for progression was in group B secondary to documented liver and lung metastases. In this patient, restaging imaging, fol- lowing extensive neoadjuvant therapy, documented signifi- cant necrosis within the primary tumor, decrease in the extent of the vena cava thrombus, and stability of the known liver metastases. Unfortunately, however, the patient’s lung lesions had progressed. As a result of the increased size of the
lung lesions, the patient met RECIST criteria for progression of disease. This patient underwent simultaneous resection of her primary tumor and liver metastases. Postoperatively, she received additional therapy, both on and off clinical trials, but eventually she succumbed to disease 28.3 months after surgical resection.
Following neoadjuvant therapy, 13 (86.7 %) patients categorized as borderline resectable were able to undergo surgical resection. Of the two patients who did not have surgery following neoadjuvant therapy (4.1 and 6.3 months
A
C
B
D
1
duration) one was categorized as group A and one as group C. The patient in group C had a progressive decline in performance status, from an ECOG score of 3-4, with worsening cachexia and malnutrition, which precluded safe operative intervention. The other patient presented with hormone excess and a large tumor abutting critical mes- enteric vascular structures and a large caval tumor throm- bus. While that patient’s hypercortisolism was effectively managed medically and her ECOG score remained stable at 1, the primary tumor did not respond to neoadjuvant che- motherapy and mitotane, preventing a margin negative surgical resection with curative intent.
Among the BRACC group, 6 patients (46.2 %) also received postoperative adjuvant systemic therapy consist- ing of mitotane plus chemotherapy in one patient and mitotane alone in the other five. In the resectable group, 11 patients (28.9 %) received postoperative adjuvant systemic therapy; one patient received mitotane plus chemotherapy; and the other 10 received mitotane alone.
Surgical outcomes
In comparing the operative outcomes of the BRACC patients and the resectable patients, there was a trend toward a higher incidence of multivisceral resection in the neoadjuvant group (84.6 %) when compared to the sur- gery-first group (52.6 %) (p = 0.053) (Table 4). Vena cava
metastasis as well as d improvement in the burden of tumor thrombus with delineation of renal vein allowing for en bloc resection of the left kidney together with IVC tumor thrombectomy and interval segment VIII liver resection
resections were performed in 3 patients in the BRACC group, but none of the patients in the resectable group underwent vena cava resection. Despite the greater number of multivisceral resections, the frequency of a microscop- ically involved margin on final pathology (R1) was similar between the BRACC and resectable groups (7.7 vs 10.5 %, respectively; p = 1.00) (Table 4). No grossly incomplete (R2) resections occurred in either group.
Oncologic outcomes
Disease-free survival was calculated from the time of surgical resection; two patients were excluded from DFS analysis in the BRACC group secondary to the presence of metastatic disease at the time of resection of the primary tumor. The median DFS for the BRACC group was 27.6 months (range: 2.9 months to not attained) and for the surgery-first group it was 12.6 months (range: 6.6-46.9 months), which was not statistically different (p = 0.48) (Table 4; Fig. 3). The actu- arial 5-year DFS was 45 % for patients treated with neoad- juvant therapy and surgical resection compared to 30 % for patients who underwent upfront surgery. For the BRACC patients, four (30.8 %) failed at distant sites and one (7.7 %) had both local and distant disease at the time of recurrence. In the resectable group, 16 patients (42.1 %) suffered distant recurrences, whereas 5 (13.2 %) had isolated local
| Neoadjuvant + surgery n = 13 (%) | Surgery n = 38 (%) | p value | |
|---|---|---|---|
| Multivisceral resection | 11 (84.6) | 20 (52.6) | 0.05 |
| Margin status | 1.00 | ||
| Negative | 12 (92.3) | 33 (86.8) | |
| Positive | 1 (7.7) | 4 (10.5) | |
| Unknown | 0 (0.0) | 1 (2.6) | |
| Median size, cm (range)ª | 11.5 (4.4-22.5) | 12.3 (4.0-30.0) | 0.67 |
| Median DFS, months (95 % CI) | 27.6 (2.9, NA) | 12.6 (6.6, 46.9) | 0.48 |
| Actuarial 5-year DFS | 45 % | 30 % | 0.32 |
| Median OS, months (95 % CI) | Not attained | 110 (34.7, 182) | 0.75 |
| Actuarial 5-year OS | 65 % | 50 % | 0.35 |
Size based on final pathologic measurement of tumor; DFS disease- free survival; OS overall survival
a Size of tumor measured at pathology evaluation
1.0
Probability of Disease-Free Survival
- -. Neoadj+surgery -Surgery alone
0.8
p=0.48
0.6
0.4
m
0.2
0.0
0
4
8
12
16
Time from Date of Surgery (years)
recurrences and 3 (7.9 %) failed both locally and distantly. These patterns were not statistically different (p = 0.83).
Similarly, possibly because of small sample sizes, the median OS for the BRACC group and the resectable group was not statistically different (not attained vs 110 months; p = 0.75) (Table 4; Fig. 4). The actuarial 5-year OS was 65 versus 57 % for the neoadjuvant and surgery groups,
1.0
—. Neoadj+surgery
-Surgery alone
Probability of Overall Survival
0.8
p= 0.75
0.6
0.4
0.2
0.0
0
4
8
12
16
Time from Date of Surgery (years)
respectively. The frequency of adjuvant mitotane use was not statistically different between the two groups (p = 0.18, data not shown). Interestingly, the pathologic size of the primary tumor was also similar in both the BRACC group and the resectable group, despite the fact that the primary tumors in the BRACC group were larger on initial imaging measurement, presumably reflecting the influence of neoadjuvant treatment on primary tumor size.
Discussion
In this study, a total of 53 patients determined to be sur- gical candidates for treatment of their primary ACC at a single institution were reviewed. Fifteen of those patients were treated with neoadjuvant mitotane ± chemotherapy secondary to one or more factors arguing against initial surgical therapy: tumor anatomy suggesting the need for multivisceral or vascular resection (group A), the presence of or concern for possible oligometastases (group B), or poor performance status or medical comorbidities (group C). We define these patients as having borderline resectable ACC; we suggest that this is a clinically meaningful des- ignation, and we submit that initial systemic therapy rather than surgery is a reasonable treatment strategy in such patients. Despite significantly more advanced disease at presentation, including 40 % of patients with stage IV disease, as well as larger primary tumors, 86.7 % of the patients treated with neoadjuvant therapy were able to proceed to surgical resection. Objective response rates were seen in more than one third of patients; importantly, most patients with vena cava tumor thrombus (6 of 7)
exhibited significant shrinkage of their tumor thrombus with neoadjuvant therapy. Only one patient had docu- mented progression of disease while on preoperative ther- apy, as evidenced by increased size of lung metastases. Interestingly, in all BRACC patients treated with initial systemic therapy, the primary tumor, vena cava thrombus, and liver metastases either responded or remained stable, permitting simultaneous surgical resection. Importantly, this group of advanced-stage patients treated with neoad- juvant mitotane ± chemotherapy followed by surgery had high rates of margin-negative resection, as well as rates of disease-free and OS at least equal to those of patients with less advanced disease who underwent upfront surgery.
Adrenal cortical carcinoma is a relatively unique cancer in several respects, most notably its potential for hormone production [12, 13]; patients with very high levels of cor- tisol production in particular may tolerate major surgery poorly and are at high risk for postoperative complications, including infections, poor wound healing, thromboembo- lism, and adrenal insufficiency. Aggressive medical man- agement with steroid-inhibiting agents, such as mitotane and metyrapone, is often the only option in patients with advanced ACC. ACC also has a propensity for vascular invasion, direct local organ extension, and distant metas- tasis. These common characteristics, taken together, argue for a multi-disciplinary approach to the treatment of patients with ACC, including consideration for adminis- tration of mitotane and/or chemotherapy in the preopera- tive setting for patients with borderline resectable disease. This subset of patients is at very high risk for an unfa- vorable outcome following surgery, and includes those with relatively advanced ACC with local-regional exten- sion of their tumors to adjacent organs, such as the ipsi- lateral kidney or vena cava; patients with known or suspected oligometastases; and those with a poor but potentially correctable performance status. Early tumor recurrence after a large operation accompanied by a lengthy period of recovery will be viewed as a treatment failure-the treatment being worse than the disease. Treatment sequencing to place surgery after induction therapy is one effort to prevent this clinical scenario.
The concept of neoadjuvant therapy for patients with ACC was previously introduced in a study of patients with recurrent ACC in which the prognostic significance of a preoperative response to mitotane was examined; in that investigation, we found that patients who had at least stable disease on mitotane prior to surgical resection had a sig- nificantly improved median OS compared to those who progressed (18 vs 9 months) [14]. The arguments in favor of preoperative therapy for patients presenting at the time of diagnosis with BRACC are similar to those for patients with recurrent but potentially resectable ACC: (1) starting systemic therapy for presumed or documented metastatic
disease at the earliest possible time; (2) downstaging of disease to maximize the probability of achieving a com- plete resection and minimize the magnitude of the opera- tion required; (3) delivering any nephrotoxic systemic therapy prior to ipsilateral nephrectomy rather than after- wards; (4) biologic selection of patients for major surgery to include those who have responded to induction therapy; (5) determination of the effectiveness of systemic therapy by having an in vivo assessment of response, which is not possible after the primary tumor is removed; and (6) pro- vision of an opportunity to address correctable comorbid- ities. The need to define “borderline resectable” ACC as a clinical stage is critical to selection of patients for whom this treatment strategy is appropriate. A standard definition allows for a tailored, patient-centered approach, and is essential for the construction of reproducible eligibility criteria for future clinical trials.
In this report we provide a working definition for BRACC by employing criteria similar to those successfully applied to patients with pancreatic adenocarcinoma, which have been validated by cooperative groups and incorporated into cur- rent standard treatment guidelines [9, 15-17]. For patients with ACC, the anatomic criteria employed to place them in the borderline resectable group A included the anticipated need for a multivisceral resection or major vascular resec- tion/reconstruction in which an R0 resection was considered unlikely. In the setting of an aggressive tumor such as ACC, with an anticipated recurrence rate following a surgical resection requiring multiorgan or vascular resection signifi- cantly in excess of 50 % [6, 18], we submit that a neoadju- vant treatment approach using agents with at least moderate efficacy (mitotane+combination systemic chemotherapy) is an attractive management strategy. We considered patients to be in group B if they had documented, potentially resectable oligometastases, or if they had findings on pre- operative imaging suggesting the possible presence of met- astatic disease (ex: indeterminate pulmonary nodules). This category arose from physician uncertainty about which clinical stage to assign patients with questionable distant metastasis (stage I/II/III vs stage IV), thereby resulting in uncertainty regarding the appropriate treatment strategy (surgery vs systemic therapy). It also allowed a mechanism for describing patients with resectable oligometastasis for whom definitive surgical treatment was potentially feasible. Finally, given the unique potential for ACC to be hormonally active and to result in significant debilitation or predispose to serious comorbidities, including pulmonary embolism, we included in group C any patient who was unfit for immediate surgery due to hormone excess, poor performance status, or other significant comorbidities.
In the present study, the patients were fairly equally distributed (group A = 6, group B = 5, and group C = 4). Interestingly, for the two patients who received preoperative
therapy but did not undergo surgical resection, one was categorized in group A and one in group C. These two patients were deemed unable to tolerate surgery secondary to progressive deterioration in performance status, as well as (in one patient) no documented tumor response, pre- cluding a margin-negative resection.
Nearly all patients in this study met the anatomic criteria for classification as group A, as 84.6 % of patients required multivisceral resection. Importantly, six of the seven patients with tumor thrombus on imaging had a docu- mented decrease in the extent of tumor thrombus following preoperative therapy. Even the single patient with radio- graphic progression following neoadjuvant therapy was able to undergo a complete resection of the primary tumor, and the rates of margin-negative resection in the neoadju- vant patients compare favorably with those in the surgery- first group. These data suggest that the neoadjuvant strat- egy is effective in many patients with BRACC, and the patients who may be at greatest risk not to proceed to surgery are those patients in group C. While additional attention should be paid to medical optimization of patients in the group C subset to maximize the probability that they will make it to the operating room, a neoadjuvant strategy seems particularly attractive for patients with relatively advanced ACC, including those with significant vena caval tumor extension and/or potentially resectable oligometas- tasis. In such patients, the downstaging afforded by neo- adjuvant treatment can make the operation technically more straightforward (Fig. 1) and can confirm respon- siveness to systemic therapy, helping to justify an extended operation in a patient with established stage IV disease (Fig. 2) [5, 19-21]. Certainly, future studies analyzing these variables in a prospective fashion will be required to determine whether there is a discrepancy in the incidence of failure to reach operative resection.
In our cohort, 80 % of the BRACC patients received a combination of mitotane and etoposide/cisplatin-based chemotherapy. This regimen has recently been shown to be the first-line therapy of choice for patients with advanced and metastatic ACC in a multi-institutional, multi-national phase III randomized trial [7]. The treatment regimen was tolerated well, with only 2 patients requiring cessation of treatment related to toxicity, including one patient with grade 2 neurotoxicity that did not preclude surgery. The other patient, who presented with an initial ECOG of 3 and therefore was a poor surgical candidate at first assessment, continued to decline. That patient exemplifies the use of neoadjuvant therapy as a predictor of whether a patient would likely tolerate an extensive operation.
The median duration of preoperative mitotane use was 5.3 months. While the median serum mitotane level of 7.1 mcg/ml may be considered below the standard thera- peutic level when used as monotherapy, its use in this
setting was generally in combination with systemic therapy as well as for its anti-steroidogenesis effect in patients with hypercortisolemia.
As delivery of postoperative adjuvant cytotoxic therapy was rare in either cohort, the success of the preoperative approach may be due in part to aggressive treatment of radiographically apparent or microscopic distant disease with combination therapy. Furthermore, it remains critical to emphasize the role for complete surgical resection in a cohort of patients with locoregionally advanced tumors. As described above, nearly 85 % of the patients treated pre- operatively required a multivisceral resection to completely extirpate their tumors, and the frequency of a margin- negative resection in these patients was 92.3 %. Given the documented importance of complete resection and the necessity to resect these tumors intact and avoid frag- mentation and tumor spillage, the contribution of com- prehensive, en bloc surgical resection to a satisfactory patient outcome cannot be overstated [5, 22, 23]. Down- staging by reducing the size of these often formidably large primary tumors and reducing the extent of the venal caval tumor thrombus have significant implications for any sur- geon, regardless of level of expertise or experience, to achieve a satisfactory oncologic result.
Our study has limitations, including the relatively small patient population due in large part to the rare nature of the malignancy. The results reported here must therefore be considered preliminary, ideally to be evaluated in a larger, multi-institutional phase II trial of a population of patients defined as having borderline resectable ACC. While we attempted to include all patients considered for surgical resection and exclude those not considered potential can- didates for surgery at initial evaluation, the retrospective nature of the analysis makes it possible that some patients whom other surgeons would have considered candidates for operation were not included, thus potentially overesti- mating the success of preoperative therapy. This potential limitation also is best evaluated through a prospective phase II trial. Despite these limitations, we submit that it is important to note that the majority of patients with bor- derline resectable ACC were able to complete preoperative therapy and subsequently undergo surgical resection. In addition, they went on to have measured oncologic out- comes, including DFS and OS, at least comparable to those of patients treated with upfront surgery. These favorable outcomes were achieved despite the fact that the neoad- juvant patients had larger tumors and more advanced dis- ease (40 % stage IV).
The results reported here are promising, and demon- strate for the first time a successful multi-disciplinary, multi-modality neoadjuvant approach to the treatment of patients with borderline resectable ACC. Future prospec- tive randomized trials are necessary to help further
delineate which patients are most likely to benefit from such an approach in an effort to maximize oncologic out- come and minimize therapeutic morbidity and mortality.
Disclosures The authors have no disclosures.
References
1. Fassnacht M, Johanssen S, Quinkler M et al (2009) Limited prognostic value of the 2004 international union against cancer staging classification for adrenocortical carcinoma: proposal for a revised TNM classification. Cancer 115:243-250
2. Maluf DF, de Oliveira BH, Lalli E (2011) Therapy of adreno- cortical cancer: present and future. Am J Cancer Res 1:222-232
3. Kerkhofs TM, Verhoeven RH, Van der Zwan JM et al (2013) Adrenocortical carcinoma: a population-based study on incidence and survival in the Netherlands since 1993. Eur J Cancer 49:2579-2586
4. Stigliano A, Cerquetti L, Sampaoli C et al (2012) Current and emerging therapeutic options in adrenocortical cancer treatment. J Oncol 2012:408131
5. Grubbs EG, Callender GG, Xing Y et al (2010) Recurrence of adrenal cortical carcinoma following resection: surgery alone can achieve results equal to surgery plus mitotane. Ann Surg Oncol 17:263-270
6. Terzolo M, Angeli A, Fassnacht M et al (2007) Adjuvant mito- tane treatment for adrenocortical carcinoma. N Engl J Med 356:2372-2380
7. Fassnacht M, Terzolo M, Allolio B et al (2012) Combination chemotherapy in advanced adrenocortical carcinoma. N Engl J Med 366:2189-2197
8. Katz MH, Pisters PW, Evans DB et al (2008) Borderline resectable pancreatic cancer: the importance of this emerging stage of disease. J Am Coll Surg 206:833-846 discussion 846-848
9. Tempero MA, Arnoletti JP, Behrman S et al (2010) Pancreatic adenocarcinoma. J Natl Compr Cancer Netw 8:972-1017
10. Lughezzani G, Sun M, Perrotte P et al (2010) The European net- work for the study of adrenal tumors staging system is prognosti- cally superior to the international union against cancer-staging system: a North American validation. Eur J Cancer 46:713-719
11. Therasse P, Arbuck SG, Eisenhauer EA et al (2000) New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Compr Cancer Netw 92:205-216
12. Bourdeau I, Mackenzie-Feder J, Lacroix A (2013) Recent advances in adrenocortical carcinoma in adults. Curr Opin Endocrinol Dia- betes Obes 20:192-197
13. Zini L, Porpiglia F, Fassnacht M (2011) Contemporary man- agement of adrenocortical carcinoma. Eur Urol 60:1055-1065
14. Gonzalez RJ, Tamm EP, Ng C et al (2007) Response to mitotane predicts outcome in patients with recurrent adrenal cortical car- cinoma. Surgery 142:867-875 discussion 875
15. Abrams RA, Lowy AM, O’Reilly EM et al (2009) Combined modality treatment of resectable and borderline resectable pan- creas cancer: expert consensus statement. Ann Surg Oncol 16:1751-1756
16. Evans DB, Farnell MB, Lillemoe KD et al (2009) Surgical treatment of resectable and borderline resectable pancreas cancer: expert consensus statement. Ann Surg Oncol 16:1736-1744
17. Vauthey JN, Dixon E (2009) AHPBA/SSO/SSAT consensus con- ference on resectable and borderline resectable pancreatic cancer: rationale and overview of the conference. Ann Surg Oncol 16:1725-1726
18. Berruti A, Fassnacht M, Baudin E et al (2010) Adjuvant therapy in patients with adrenocortical carcinoma: a position of an international panel. J Clin Oncol 28:e401-e402 author reply e3
19. Datrice NM, Langan RC, Ripley RT et al (2012) Operative management for recurrent and metastatic adrenocortical carci- noma. J Surg Oncol 105:709-713
20. Gaujoux S, Al-Ahmadie H, Allen PJ et al (2012) Resection of adrenocortical carcinoma liver metastasis: is it justified? Ann Surg Oncol 19:2643-2651
21. Kemp CD, Ripley RT, Mathur A et al (2011) Pulmonary resec- tion for metastatic adrenocortical carcinoma: the National Cancer Institute experience. Ann Thorac Surg 92:1195-1200
22. Hermsen IG, Kerkhofs TM, den Butter G et al (2012) Surgery in adrenocortical carcinoma: importance of national cooperation and centralized surgery. Surgery 152:50-56
23. Miller BS, Gauger PG, Hammer GD et al (2012) Resection of adrenocortical carcinoma is less complete and local recurrence occurs sooner and more often after laparoscopic adrenalectomy than after open adrenalectomy. Surgery 152:1150-1157