Adjuvant radiotherapy after surgical resection for adrenocortical carcinoma: A systematic review of observational studies and meta-analysis
ABSTRACT
Purpose: Historically, the role of adjuvant radiotherapy (RT) for patients with adrenocortical carcinoma (ACC) has been controversial. The objective of this research is to review systematically the literature evaluating the role of adjuvant RT in patients with ACC undergone a surgical resection.
Materials and Methods: The electronic databases were searched for articles published until July 2017 without language restriction: Lilacs, Medline, Embase, and the Cochrane. Two reviewers independently appraised the eligibility criteria and extracted data. When possible, a fixed-effect meta-analysis was done. The systematic review (SR) followed all the criteria of the MOOSE guideline.
Results: Overall, 382 citations were identified. After the screening of titles and abstracts, 12 articles (eight case series [48 patients] and 4 cohort studies [136 patients]) were included in the final analysis. For the local recurrence, the pooled relative risk (RR) was RR = 0.46 (95% confidence interval: 0.28-0.75), in favor of adjuvant RT when compared with surgery alone. Concerning overall mortality and disease recurrence, no significant difference between adjuvant RT and surgery was detected, RR = 0.77 (CI 95% 0.49-1.22, P = 0.27), and RR = 0.95 (IC 95% 0.74-1.24, P = 0.67). In all cohort studies, the acute toxicités were graduated as mild and self-limited with nausea and fatigue being the most common symptoms. Only one case (1/50) of impairment of kidney function was detected as late toxicity in these studies.
Conclusions: This SR and meta-analysis indicate that adjuvant RT dramatically reduces the local recurrence of ACC after surgery. Moreover, the treatment has a low acute and late toxicity, resulting in a high therapeutic index. Further, prospective studies are needed to confirm or refute the role of RT on survival and disease recurrence.
KEY WORDS: Adrenocortical carcinoma, local control, radiotherapy, systematic review
INTRODUCTION
Malignant tumors from adrenal are rare neoplasms with an incidence rate of 0.7/1000.000 in the USA.[1] Adrenocortical carcinoma (ACC) is the most common tumor subtype among adrenal tumors. ACC is considered an aggressive malignancy with higher rates of recurrence and poor survival.[2] Surgical resection is the only chance to achieve the cure or a long survival. However, even after a complete surgical resection with negative margins about 30% of patients may experience a local recurrence of their disease.[2] In patients with compromised margins, the rate of local recurrence can be as high as 60%, even offering adjuvant treatment.[2] Historically, the use of adjuvant treatment for ACC utilizing radiotherapy (RT) has
been controversial.[3-10] In fact, the controversial role of RT in ACC comes from the paucity of good quality evidence describing the v benefits. Moreover, in the older series, the sample size was small and the follow-up short, limiting the statistical power to identify differences.[3-10]
However, recently, several reports using multimodality treatment approaches with a larger sample have hinted that adjuvant radiation therapy can reduce local failures in high-risk patients.[11-14]
This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.
For reprints contact: reprints@medknow.com
Cite this article as: Viani GA, Viana BS. Adjuvant radiotherapy after surgical resection for adrenocortical carcinoma: A systematic review of observational studies and meta-analysis. J Can Res Ther 2019;15:S20-6.
Gustavo Arruda Viani, Bruno Silveira Viana
Department of Radiation Oncology, Faculty of Medicine of Marilia, São Paulo, Brazil
For correspondence: Prof. Gustavo Arruda Viani, Warner Gomes Fernandes, Marília, Brazil. E-mail: gusviani@ gmail.com
Access this article online
Website: www.cancerjournal.net
DOI: 10.4103/jcrt.JCRT_996_15
Quick Response Code:
In some retrospective series, adjuvant RT reduces the risk of local recurrence about 50% in patients with positive surgical margins (60% with no RT versus 30% with RT). Therefore, to investigate whether adjuvant radiation therapy improves the rates of local recurrence and disease-free survival in patients with ACC undergone surgical resection, we developed a systematic review (SR).
MATERIALS AND METHODS
A SR of the literature was performed on the electronic databases (Embase, Cochrane, PubMed, and lilacs) from 1960 to July 2017 to find all the published studies on adjuvant RT for ACCs. Manual searches were also done (completed independently and in duplicate) to identify all published observational studies that reported the results of adjuvant RT compared or not with only surgical resection. The SR followed the recommendations of MOOSE guidelines for SR and meta-analysis of observational studies [Supplement Table 1].
Inclusion criteria
The following study designs were allowed in this SR: (a) case series (3-10 patients) reporting local control with adjuvant RT after surgical resection and (b) retrospective studies showing the results of adjuvant RT compared with the surgical control group. Studies reporting the results of palliative RT for ACC were excluded from the study.
Search methods
We used the following MESH search headings: “adrenal carcinoma;” “adrenocortical carcinoma,” “adjuvant treatment,” and “radiotherapy.” All the abstracts, studies, and citations found by the searching were reviewed. The information from each study was extracted independently by two reviewers. Independent of the study design, the following data from each study were extracted: the first author, extension of surgery, tumor stage, RT dose, chemotherapy, local recurrence, any recurrence, mortality, and adverse effects.
Statistical analysis
When possible, the data analyzes were made with Review Manager (RevMan) [Computer program]. Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014. For categorical variables, weighted risk ratios and their 95% confidence interval (CI) were calculated using Mantel-Haenszel method.
Results were examined for heterogeneity at a significance level of P < 0.05 according to the methods outlined by DerSimonian and Laird.[15] A fixed-effects model was used if there was no evidence of heterogeneity between studies if there was evidence of heterogeneity random effects model was used for meta-analysis. The risk ratio and 95% CI were calculated for each study and presented in a Forrest plot. Sensitivity analyzes were performed by excluding the trials that were the potential heterogeneity source (sample size imbalance, follow-up time,
and different sample characteristics). The “funnel plot” was used to evaluate the publication bias.
RESULTS
The search found 382 citations. After the evaluation, 370 articles were excluded, resulting in 4 cohort studies (136 patients)[12-14,16] and 8 case series (48 patients).[3-10] A total of 184 patients with ACCs were included in this SR. Figure 1 describes the flowchart of study selection.
Case series
In our search, we found eight studies (48 patients with ACC) classified as case series including between three and ten patients [Table 1]. In the majority of these studies, no details on the margins of resection (R0, R1, or R2), the RT dose or target volume definitions were given.
In 1976, Percarpio and Knowlton published the first case series to report the results of adjuvant RT for ACC. In their study, the authors delivered adjuvant RT with a dose ranging from 28 to 41 Gy in four patients after incomplete resection. The authors reported a local control of 25% with one patient surviving for more than 11 years without a local failure. All others patients from this case series experienced a recurrence between 2 and 34 months.[3]
In 1979, King and Lack reported their results using adjuvant RT in four patients who survived for >5 years. Although this
Total number of studies retrieved on search (382)
Excluded duplicate studies (12)
Articles remaining after excluding duplicate studies (370)
Exclude articles not relates to the topic (340)
Articles selected for full review (20)
Exclude articles (8)
Review article (4)
Reporting treatment outcome on Metastatic disease (3)
With no comparison group (1) Insufficient data (1)
Articles selected for full review (12)
4 cohort studies
8 case series
Viani and Viana: Adjuvant radiotherapy for adrenocortical carcinoma
| Authors | Year | Patients | RT dose (Gy) | Local control (%) | Follow up |
|---|---|---|---|---|---|
| (reference) | |||||
| Percarpio and | 1976 | 4 | 28-41 | 25 | - |
| Knowlton[3] | |||||
| King and Lack[4] | 1979 | 4 | 42-55 | NR | - |
| Didolkar et al.[5] | 1981 | 10 | NR | 40 | - |
| Henley et al.[6] | 1983 | 10 | NR | 10 | - |
| Magee et al.[7] | 1987 | 9 | 20-30 | 56 | 11 years |
| Markoe et al.[8] | 1991 | 5 | 42-60 | 60 | - |
| Pommier and | 1992 | 3 | 39-45 | 0 | - |
| Brennan[9] | |||||
| Hermsen et al.[10] | 2010 | 10 | NR | 100 | - |
RT=Radiotherapy, NR=No reported
study gave information on four patients treated with adjuvant RT and with a long follow-up, the authors did not give any details on disease-free survival of their patients.[4]
In 1981, Didolkar et al. reported a local control rate of 40% from 10 patients treated with ACC and R2 resection followed by adjuvant RT. However, the authors did not provide any information on radiation technique, radiation dose, and follow-up time.[5] In 1983, Henley et al. described the treatment results of patients with ACC Stage I-III treated with aggressive surgery. In their case series, ten patients were submitted to adjuvant RT after R1-2 surgical resection. The authors reported a local control of 10% (1/10), but with no information on follow-up and disease-free survival.[6] In 1987, Magee et al. delivered adjuvant RT (20-30 Gy) to nine patients with ACC operated with R0-2 surgical resection. The local control delivering adjuvant RT was considered low (5/9 = 56%), and the authors concluded that adjuvant RT was ineffective to control the disease, mainly, in residual disease.[7] In 1991, the researchers from Philadelphia described their experience treating five patients with adjuvant RT after surgical resection. They reported a local control of 60%, with no severe late effects with doses between 42 and 60 Gy.[8] In 1993, Pommier and Brennan reported a negative experience with adjuvant RT. In their case series, three ACC patients received adjuvant RT with doses between 39 and 45 Gy after R0-1 surgery. With a median follow-up 28 months, all patients developed a local failure.[9] Thus, the authors concluded that adjuvant RT should not recommend for ACC after surgical resection. In opposite to Pommier and Brennan results, Hermsen et al. reported a good experience delivering adjuvant RT to three patients submitted to a surgical resection due to ACC. The authors reported no failure with adjuvant RT. However, they did not give any information on RT technique, dose, or follow-up of these patients.[10]
Cohort studies
The search identified three cohort studies comparing adjuvant RT with no adjuvant RT (surgical control arm), 50 patients in adjuvant RT and 66 in no adjuvant RT.
Fassnacht et al. were the first to publish a matching pair analysis comparing adjuvant RT with no adjuvant treatment
after surgical resection. They screened the Germany databases for patients with ACC treated with surgical resection and adjuvant RT. The authors identified 14 patients submitted to RT, who were matched with 14 surgical controls according to tumor stage, margin status, and tumor size. The results showed a significant impact of adjuvant RT on local recurrence, but with no effect on disease survival and overall survival.[13]
In 2012, researchers from M. D Anderson Cancer Center (MDACC) conducted a retrospective study. In this cohort, they compared the treatment results of adjuvant RT (16 patients) after surgery with patients only submitted to surgery (32 patients). Although the authors balanced the groups according to tumor size, adjuvant mitotane, and margin resection, the main bias of the study was the referral bias. Since the majority of patients of the RT group was treated outside MDACC. The authors concluded that adjuvant RT did not improve any the outcomes in patients treated at the community.[12] Researchers from the Michigan Cancer Center analyzed 20 years of their experience in managing ACC. The authors compared twenty patients treated with adjuvant RT versus twenty patients with the only surgery. In this study, the authors show an impressive impact of adjuvant RT on the local control, but with no effect on disease recurrence in other sites or survival.[14]
More recently, Srougi et al.[16] reported their experience treating twenty ACC with adjuvant RT or with only surgery. In this cohort, patients submitted to adjuvant RT had a significative lower recurrence (40%) rate than only surgery group (60%). No significant difference was observed for overall mortality or disease progression. No late toxicity was noted during the follow-up period.
All these three studies provide data about treatment techniques, margin status, tumor size, RT dose, target treatment volumes, follow-up, and disease recurrence, as showed in Table 2.
Local recurrence
Four cohort studies[12-14,16] compared adjuvant RT with matching surgical controls and reported the local recurrence as an outcome, representing 136 patients. The local recurrence rates were 23.3% (14/60) and 51.3% (39/76) for RT arms and no RT groups, respectively. The relative risk (RR) was 0.46 with CI 95% 0.28-0.75, P = 0.0002. However, the test for heterogeneity, as fixed as the random model, was statistically significant (P = 0.003, I 2 = 78%) not allowing the results to be pooled. To solve the heterogeneity problem, we performed a sensibility analysis excluding the study from MDACC.[11] The reasons for the exclusion were the presence of referral bias, an imbalance in the sample size and different follow-up time between the groups. After that, combining the results of the three cohorts (88 patients) a statistical significance in favor of RT group was observed with a RR = 0.24, IC 95% 0.12-0.49, P = 0.0002. The heterogeneity test was not significant with a P = 0.05, I 2 = 57% indicating that the pooled analysis is valid [Figure 2].
Disease recurrence (any recurrence)
All cohort studies provided information on disease recurrence. The disease recurrence rates were 71.6% (43/60) and 77.6% (59/76) for RT arms and no RT groups, respectively. The RR was 0.95 with IC 95% 0.74-1.22, P = 0.67. The heterogeneity test was not significant with a P = 0.15, I 2 = 44% indicating that the pooled analysis is valid [Figure 3].
Overall mortality
Three cohort studies provided information on overall
mortality.[12,14,16] The overall mortality rates were 38.6% (17/44) and 50% (22/44) for RT arms and no RT groups, respectively. The RR was 0.77 with IC 95% 0.49-1.22, P = 0.3. The heterogeneity test was not significant with a P = 0.27, I 2 = 13% indicating that the pooled analysis is valid [Figure 4]. None of the outcomes evaluated (local recurrence, disease recurrence, and overall mortality) were associated with publication bias [Figure 5].
Adverse effects
The Germany study reported the acute toxicity graduated
| Variables | Study | |||||||
|---|---|---|---|---|---|---|---|---|
| Michigan[14] | Germany study[13] | MDACC[12] | Brazilian study[x] | |||||
| RT (%) | No RT (%) | RT (%) | No RT (%) | RT (%) | No RT (%) | RT (%) | No RT (%) | |
| Patients | 20 | 20 | 14 | 14 | 16 | 32 | 10 | 10 |
| Age at diagnosis | 49 | 42 | 43 | 48 | 48 | 44 | 40 | 38 |
| Disease stage | ||||||||
| I | - | - | 1 (7) | 1 (7) | NR | NR | ||
| II | 13 (65) | 13 (65) | 7 (50) | 8 (57) | 9 (56) | 17 (53) | ||
| III | 7 (35) | 7 (35) | 3 (21) | 5 (35) | 7 (44) | 15 (47) | ||
| Size, cm (mean) | 10.6 | 12.6 | 11.7 | 11 | 12 | 11 | 12.7 | 10 |
| Surgical margins | ||||||||
| Positive | 4 (20) | 3 (15) | 2 (14) | 2 (14) | 7 (44) | 13 (41) | NR | NR |
| Negative | 11 (55) | 14 (70) | 8 (58) | 8 (58) | 7 (44) | 13 (41) | ||
| Missing | 5 (25) | 3 (15) | 4 (28) | 4 (28) | 2 (12) | 6 (18) | ||
| Mitotane | ||||||||
| Yes | 15 (75) | 15 (75) | 5 (35) | 5 (35) | 4 (25) | 10 (31) | NR | NR |
| RT technique | ||||||||
| 2D | - | 2 (14) | NR | NR | ||||
| 3D | 5 (25) | 12 (86) | ||||||
| IMRT | 15 (75) | - | ||||||
| RT dose (Gy) | 55 (45-60) | 50.4 (41.4-56) | 50.4 (36-59.4) | 54 (45-54) | ||||
RT=Radiotherapy, MDACC=MD Anderson Cancer Center, 2D=Two-dimensional, 3D=Three-dimensional, IMRT=Intensity modulated radiation therapy, NR=No reported
| Study or Subgroup | RT | No RT | Weight | Risk Ratio M-H, Fixed, 95% CI | Risk Ratio M-H, Fixed, 95% CI | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Events | Total | Events | Total | |||||||||
| Fassnacht 2006 | 2 | 14 | 11 | 14 | 30.8% | 0.18 [0.05, 0.68] | ||||||
| Habra 2012 | 7 | 16 | 10 | 32 | 18.7% | 1.40 [0.66, 2.98] | ||||||
| Salboc 2015 | 1 | 20 | 12 | 20 | 33.6% | 0.08 [0.01, 0.58] | ||||||
| Srougi 2017 | 4 | 10 | 6 | 10 | 16.8% | 0.67 [0.27, 1.66] | ||||||
| Total (95% CI) | 60 | 76 | 100.0% | 0.46 [0.28, 0.75] | ||||||||
| Total events | 14 | 39 | ||||||||||
| Heterogeneity: Chi == 13.88, df= 3 (P = 0.003); F == 78% Test for overall effect: Z = 3.14 (P = 0.002) | 0.01 0.1 1 | 10 100 | ||||||||||
| Favours [experimental] Favours [control] | ||||||||||||
| Study or Subgroup | RT Events | No RT | Weight | Risk Ratio M-H, Fixed, 95% CI | Risk Ratio M-H, Fixed, 95% CI | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total | Events | Total | ||||||||
| Fassnacht 2006 | 11 | 14 | 14 | 14 | 28.2% | 0.79 [0.59, 1.06] | ||||
| Habra 2012 | 14 | 16 | 24 | 32 | 31.1% | 1.17 [0.89, 1.53] | ||||
| Salboc 2015 | 10 | 20 | 14 | 20 | 27.2% | 0.71 [0.42, 1.21] | ||||
| Srougi 2017 | 8 | 10 | 7 | 10 | 13.6% | 1.14 [0.69, 1.90] | ||||
| Total (95% CI) | 60 | 76 | 100.0% | 0.94 [0.77, 1.13] | ||||||
| Total events | 43 | 59 | ||||||||
| Heterogeneity: Chi == 5.34, df = 3 (P = 0.15); 1 == 44% Test for overall effect: Z = 0.69 (P = 0.49) | 0.01 | 0.1 | 1 | 10 100 | ||||||
| Favours [experimental] Favours [control] | ||||||||||
Figure 3: Forrest plot including studies comparing adjuvant radiotherapy versus no radiotherapy for disease recurrence
| Study or Subgroup | RT Events | Total | NRT | Weight | Risk Ratio M-H, Fixed, 95% CI | Risk Ratio M-H, Fixed, 95% CI | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Events | Total | ||||||||||
| Fassnacht 2006 | 6 | 14 | 6 | 14 | 27.3% | 1.00 [0.43, 2.35] | |||||
| Salboc 2015 | 4 | 20 | 9 | 20 | 40.9% | 0.44 [0.16, 1.21] | |||||
| Srougi 2017 | 7 | 10 | 7 | 10 | 31.8% | 1.00 [0.56, 1.78] | |||||
| Total (95% CI) | 44 | 44 | 100.0% | 0.77 [0.49, 1.22] | |||||||
| Total events | 17 | 22 | |||||||||
| Heterogeneity: Chi == 2.30, df = 2 (P = 0.32); 1 == 13% Test for overall effect: Z = 1.10 (P = 0.27) | 0.01 | 0.1 1 | 10 100 | ||||||||
| Favours [experimental] Favours [control] | |||||||||||
Figure 4: Forrest plot including studies comparing adjuvant radiotherapy versus no radiotherapy for overall mortality
by common toxicity criteria (CTC) criteria in 14 patients. In this study, the authors observed eight nausea Grade I or II, and four dermatitis Grade I. Late toxicity was also reported with 1 case of impaired kidney function and 1 case of partial Budd-Chiari syndrome.[13]
The MDACC study described the acute effects of 16 patients treated outside the institution. They observed nine nausea Grade I or II, five fatigue Grade I or II, and one abdominal pain. Habra et al. did not provide any information on late toxicity.[12]
The study of the University of Michigan evaluated the toxicity in twenty patients treated with adjuvant RT. They used the CTC criteria to graduate the acute toxicity. Sabolch et al.[14] observed that 16 patients experienced nausea Grade I, and only one patient developed nausea Grade III. The authors did not observe any late toxicity during the study follow-up.
The Brazilian series evaluated the toxicity in 10 ACC patients treated with adjuvant RT. They used RTOG criteria to graduate acute and late toxicity. Most of acute toxicity was mild (Grade I and II) with no late toxicity was observed.
Treatment techniques
The Germany study treated their patients with conformal RT or conventional RT, delivering a median dose 50.4 Gy (range 36-59.4 Gy) in 25 fractions (range: 20-30). They used a linear accelerator with 6-18 MeV.[13] The study from Michigan used intensity-modulated RT (IMRT) in 15 patients, and conformal RT to the others. The median dose was 55 Gy (range 41.4-56 Gy). All patients were simulated with TC simulation, and after 2008, all patients had four-dimensional movement assessment of the target volume.[14] Habra et al. only reported the median dose and fractionation utilized to treat their patients. The authors did not give any details on treatment technique, target volume, or treatment machine.[12]
Srougi et al.[16] only also reported the median dose used (54 Gy median, range: 45-54 Gy) with no information about treatment technique, target volume, or treatment machine.
DISCUSSION
The basic premise to perform this SR was to present the
experience accumulated in the last 40 years about the role of adjuvant RT for ACCs. ACCs are rare tumors, and consequently, a large-scale analysis of the benefits of adjuvant RT is extremely difficult to conduct. Therefore, a SR including all the evidence available can be useful to point out the benefits and harms of the RT treatment. Moreover, historically, the ACCs have been considered a radioresistant tumor and adjuvant RT has often been omitted in ACC patients. For instance, in the United States, recent data of the surveillance, epidemiology, and end results have estimated that RT is delivered in approximately 10% of ACC treatments.[17] These data are particularly concerning given the difficulty of attaining a complete surgical resection due to the location of these tumors. Data from the National Cancer Database have shown that about 20% of patients are left with microscopic or gross residual margins.[17]
From a historical perspective, the explanation for this low rate of adjuvant RT is related to the mix results from case reports or case series published between 1976 and 2006. In this period, the majority of information on the adjuvant RT came from studies with poor quality, no details on RT technique, treatment volume, toxicity, and time of follow-up [Table 1]. In the last ten years, the publication of cohort studies with better quality has changed the view of the role of RT for operated ACCs.
Therefore, the main finding of this SR is to indicate that adjuvant RT abruptly reduces the local recurrence with a low rate of acute/late toxicity. The pooled results of four cohort studies (136 patients) demonstrate that adjuvant RT provides an absolute risk reduction of local recurrence of 28% in 5 years. This impressive reduction in the absolute risk reduction of local failure yields a number need to treat of 3.5 to prevent a failure. Interestingly, this impressive effect over local recurrence appears to be for both stages, Stage II and III. Unfortunately, we did not perform a subgroup analysis by stage group, or margin status. Evaluating the proportion of patients per stage, 50%-65% and 21%-44% in the RT group, versus 53%-65% and 35%-47% in no RT group had Stage II or III, respectively [Table 2]. Thus, it is possible to think that even with the inclusion of a great proportion of patients with better prognostic (Stage II), the effect of RT remains significant. This rational is also valid for the margin status. Since the majority of patients in RT group had negative margins (50%-58%) and in two studies,[12,14] the RT group had slightly more positive
local control
o
SEGODERRD
0.2
0.4
0.6
O
0.8
8.61
0.2
10
100
a
disease recurrence
o
SECIOR(RR])
:
0.1
O
4
0.2
D
0.3
0.4
o
8.61
1
RR
100
b
Overall mortality
o
SE(Og[RRD
>.2
O
O
>.6
2.8
1
.
RR
c
margins. Consequently, even including patients with worst prognostic, or a great proportion of patients who theorically would have a low benefit from RT, the adjuvant treatment remains effective in reducing the local failure. Thus, although a comparison by subgroup has not been possible, our results suggest that adjuvant RT is appropriate for patients with Stage II or III with or without positive margins. The heterogeneity introduced by the MDACC study is knowledge in the literature. Fassnacht et al. compared the outcomes in patients with Stage II ACC with referral bias identified the German ACC database.[18] The author concluded that patients treated in the specialized centers had a better prognosis than who were treated outside
a specialized center due to a major referral bias. Thus, our conducted of excluding the MDACC study of the sensibility analysis was correct.
Concerning overall mortality and disease recurrence, the present study did not detect a significant difference between those treated with RT versus those treated with surgery alone. The pooled results from three cohort studies indicate that locoregional control does not significantly affect survival. However, a prospective study with an adequate sample is necessary to confirm or refute the absence of a survival benefit with adjuvant RT.
The rate of adverse effects produced with RT is another point of this SR that deserves attention. In all cohort studies included in this SR, the RT adverse effects were considered mild and self-limited. The main acute toxicity described in those studies was nausea and fatigue, and only one study reported one case of impairment of kidney function during the follow-up. Moreover, nausea, emesis, and fatigue are often adverse effects reported with the administration of mitotane. Due to similarities between the collateral effects of RT and mitotane, it would be reasonable to think that the combined treatment would be intolerable by the most patients. However, there was not a significant difference in toxicity profile comparing studies with a higher versus lower proportion of patients receiving adjuvant RT and mitotane. For instance, in the Michigans’ study, 75% of patients received mitotane and adjuvant RT. The most common acute adverse effect was nausea Grade I with only one case (1/20) nausea Grade III. In the other two studies, which included 25%-35% of patients with RT and mitotane, nausea Grade I or II was also the most often acute toxicity event. Regarding the RT dose and RT technique, all the three cohort studies used a median dose ≥50 Gy with fractions dose of 1.8-2.0 Gy, with the majority of patients treated with IMRT or three-dimensional-RT (3D-RT). In the Michigan’s study, no patient received conventional RT, in the Germany study only two patients received adjuvant RT with the conventional technique, and the MDACC study no details on techniques was provided. Therefore, evaluating the evidence available, we were not capable of observing a relation between low toxicity/mitotane usage with RT technique or RT dose.
For us, it is important to stress that this SR is subordinate to limitations common to retrospective analyzes, such as the heterogeneity of patient characteristics, treatments, timing of radiation, and the referral bias.
The strengths are in the historical overview, including all available information on adjuvant RT for ACCs, and in the pooled analysis of three homogeneous cohort studies which provides a uniform base for the adjuvant treatment of ACC.
CONCLUSIONS
This SR and meta-analysis suggests that adjuvant RT significantly reduces the risk of local recurrence in patients with ACC. Evaluating the acute and late toxicity described in all available studies, the adjuvant RT produces only mild and self-limited symptoms, resulting in an excellent therapeutic index, even when combined with mitotane. It is not clear whether the therapeutic index has any relation to RT technique or RT dose. Given the reduction of local recurrence, the low rate of toxicity and the morbidity of a local failure, adjuvant RT could be considered for high-risk ACC patients Stage II or III (positive margins and tumor >10 cm) undergone to surgical resection. In addition, based on the experience accumulated over the decades, is prudent recommend adjuvant RT using IMRT or 3D-RT with doses ranging from 50 to 55 Gy
to guarantee the therapeutic index. Multi-institutional, prospective, randomized trials are necessary to confirm or refute a benefit for overall survival or disease progression with RT.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
1. Kebebew E, Reiff E, Duh QY, Clark OH, McMillan A. Extent of disease at presentation and outcome for adrenocortical carcinoma: Have we made progress? World J Surg 2006;30:872-8.
2. .. Else T, Kim AC, Sabolch A, Raymond VM, Kandathil A, Caoili EM, et al. Adrenocortical carcinoma. Endocr Rev 2014;35:282-326.
3. Percarpio B, Knowlton AH. Radiation therapy of adrenal cortical carcinoma. Acta Radiol Ther Phys Biol 1976;15:288-92.
4. King DR, Lack EE. Adrenal cortical carcinoma: A clinical and pathologic study of 49 cases. Cancer 1979;44:239-44.
5. Didolkar MS, Bescher RA, Elias EG, Moore RH. Natural history of adrenal cortical carcinoma: A clinicopathologic study of 42 patients. Cancer 1981;47:2153-61.
6. Henley DJ, van Heerden JA, Grant CS, Carney JA, Carpenter PC. Adrenal cortical carcinoma - A continuing challenge. Surgery 1983;94:926-31.
7. Magee BJ, Gattamaneni HR, Pearson D. Adrenal cortical carcinoma: Survival after radiotherapy. Clin Radiol 1987;38:587-8.
8. Markoe AM, Serber W, Micaily B, Brady LW. Radiation therapy for adjunctive treatment of adrenal cortical carcinoma. Am J Clin Oncol 1991;14:170-4.
9. Pommier RF, Brennan MF. An eleven-year experience with adrenocortical carcinoma. Surgery 1992;112:963-70.
10. Hermsen IG, Groenen YE, Dercksen MW, Theuws J, Haak HR. Response to radiation therapy in adrenocortical carcinoma. J Endocrinol Invest 2010;33:712-4.
11. Sabolch A, Feng M, Griffith K, Hammer G, Doherty G, Ben-Josef E, et al. Adjuvant and definitive radiotherapy for adrenocortical carcinoma. Int J Radiat Oncol Biol Phys 2011;80:1477-84.
12. Habra MA, Ejaz S, Feng L, Das P, Deniz F, Grubbs EG, et al. A retrospective cohort analysis of the efficacy of adjuvant radiotherapy after primary surgical resection in patients with adrenocortical carcinoma. J Clin Endocrinol Metab 2013;98:192-7.
13. Fassnacht M, Hahner S, Polat B, Koschker AC, Kenn W, Flentje M, et al. Efficacy of adjuvant radiotherapy of the tumor bed on local recurrence of adrenocortical carcinoma. J Clin Endocrinol Metab 2006;91:4501-4.
14. Sabolch A, Else T, Griffith KA, Ben-Josef E, Williams A, Miller BS, et al. Adjuvant radiation therapy improves local control after surgical resection in patients with localized adrenocortical carcinoma. Int J Radiat Oncol Biol Phys 2015;92:252-9.
15. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177-88.
16. Srougi V, de Bessa J Jr., Tanno FY, Ferreira AM, Hoff AO, Bezerra JE, et al. Adjuvant radiotherapy for the primary treatment of adrenocortical carcinoma: Are we offering the best? Int Braz J Urol 2017;43:841-8.
17. Bilimoria KY, Shen WT, Elaraj D, Bentrem DJ, Winchester DJ, Kebebew E, et al. Adrenocortical carcinoma in the United States: Treatment utilization and prognostic factors. Cancer 2008;113:3130-6.
18. Fassnacht M, Johanssen S, Fenske W, Weismann D, Agha A, Beuschlein F, et al. Improved survival in patients with stage II adrenocortical carcinoma followed up prospectively by specialized centers. J Clin Endocrinol Metab 2010;95:4925-32.
| in epidemiology checklist | for systematic review | Criteria | Brief description of how the criteria were handled in the meta-analysis Data extracted from each of |
|---|---|---|---|
| Criteria | Brief description of how the criteria were handled in the meta-analysis | Rationale for the selection and coding of data | |
| Reporting of background should include | ACC is a rare malignal tumor with an aggressive behavior The role of adjuvant radiotherapy for ACC historically is controversial. Thus, the potential of adjuvant radiotherapy on resected | the studies were relevant to the population characteristics, study | |
| Problem definition | Assessment of confounding | design, and outcomes Conducted sensitivity analyses by eliminating studies that had not adjusted for possible confounders such as sample size, treatment technique and referral bias | |
| ACC remains to be summarized quantitatively | Assessment of study quality, including blinding | Sensitivity analyses by several quality indicators such as timing | |
| Hypothesis statement | Adjuvant radiotherapy reduces the risk of ACC recurrence Recurrence of ACC | of quality assessors; stratification or regression on possible predictors of study results | of diabetes assessment relative to tuberculosis, method of diabetes and tuberculosis diagnosis, control selection, adjustment factors, |
| Description of study outcomes | |||
| Type of exposure or intervention used | Adjuvant radiotherapy | potential duplicate data, use of convenience samples | |
| Type of study designs used | Due to rarity of disease, we included case series studies and cohort studies. We excluded studies of case reports and duplicate publication studies | Assessment of heterogeneity | Heterogeneity of the studies was explored within two types of study designs using Cochrane's Q test of heterogeneity and /2 statistic that provides the relative amount |
| Study population Reporting of search strategy | Patients with ACC undergone to surgical resection and submitted to adjuvant radiotherapy with or without a surgical group as control The credentials of the two investigators GV and BV are | Description of statistical methods in sufficient detail to be replicated | of variance of the summary effect due to the between-study heterogeneity Description of methods of meta-analyses, sensitivity |
| should include Qualifications of searchers | analyses, and assessment of publication bias are detailed in the methods | ||
| indicated in the author list PubMed from 1965 - March 2015 EMBASE from 1974 - March 2015 Lilacs | Provision of appropriate tables and graphics | We included two boxes detailing the studies' characteristics, one flowchart, three forest plot of all studies, and one funnel plot | |
| Search strategy, including time period included in the synthesis and keywords | |||
| Scopus Cochrane | Reporting of results should include | ||
| Databases and registries searched | PubMed, EMBASE, Lilacs, Scopus, and Cochrane | Graph summarizing individual study estimates | Figure 2-4 |
| Search software used, name and version, including special features | We did not employ a search software. Endnote was used to merge retrieved citations and eliminate duplications | and overall estimate Table giving descriptive information for each study included | Table 1-2 |
| Use of hand searching | We hand-searched bibliographies of retrieved papers for additional | Results of sensitivity testing | Figure 3 |
| references | Indication of statistical uncertainty of findings | 95% CIs were presented with all | |
| List of citations located and those excluded, | Details of the literature search process are outlined in the | summary estimates, I 2 values and results of sensitivity analyses | |
| including justifications | flowchart. The citation list is available upon request | Reporting of discussion should include | |
| Method of addressing articles published in languages other than | We placed no restrictions on language; local scientists fluent in the original language of the article were contacted for translation We did not contact any authors for unpublished studies on the adjuvant radiotherapy We did not contact authors | Quantitative assessment of bias Justification for exclusion | Sensitivity analyses indicate heterogeneity in strengths of the association due to most common |
| English | biases in observational studies | ||
| Method of handling abstracts and unpublished | We excluded studies that had not adjusted for a potential confounder | ||
| studies Description of any contact with authors | and used different exposure or outcome assessment for the comparison groups | ||
| Reporting of methods should include | Assessment of quality of included studies | We discussed the results of the sensitivity analyses and | |
| Description of relevance or | Detailed inclusion and exclusion | potential reasons for the observed | |
| appropriateness of studies assembled for assessing the hypothesis to be tested | criteria were described in the methods section | Reporting of conclusions should include | heterogeneity |
Contd …
| Criteria | Brief description of how the criteria were handled in the meta-analysis |
| Consideration of alternative explanations for observed results | We discussed that potential unmeasured confounders such as referral bias may have caused residual confounding, but the measured factors that are correlated with such confounders would have mitigated the bias |
| Generalization of the conclusions | The calculation of risk ratio, absolute risk reduction, and number need to treat gives a realistic estimate of the reduction of ACC recurrence |
| Guidelines for future research | We recommend future studies on the effect of adjuvant radiotherapy for ACC |
| Disclosure of funding source | No separate funding was necessary for the undertaking of this systematic review |
ACC=Adrenocortical carcinoma, CIs=Confidence intervals, GV=Gustavo viani, BV=Bruno viana