Ann Surg Oncol (2022) 29:1315–1326

https://doi.org/10.1245/s10434-021-10777-6

ORIGINAL ARTICLE – PEDIATRIC ONCOLOGY

Robotic Surgery in Pediatric Oncology: Lessons Learned

from the First 100 Tumors—A Nationwide Experience
Thomas Blanc, MD, PhD1,2,3 , Pierre Meignan, MD1,4, Nicolas Vinit, MD1, Quentin Ballouhey, PhD5,
Luca Pio, MD1, Carmen Capito, PhD1, Caroline Harte, MD6, Fabrizio Vatta, MD1, Louise Galmiche-Rolland, PhD7,
Véronique Minard, PhD8, Daniel Orbach, PhD9, Laureline Berteloot, MD10, Cécile Muller, PhD1,
Jules Kohaut, MD1, Aline Broch, MD1, Karim Braik, MD4, Aurélien Binet, MD4, Yves Heloury, PhD1,2,
Laurent Fourcade, PhD5, Hubert Lardy, PhD4, and Sabine Sarnacki, PhD1,2
1
Department of Pediatric Surgery and Urology, Hôpital Necker-Enfants Malades, Assistance Publique-Hôpitaux de Paris,
Paris, France; 2Université de Paris, Paris, France; 3Département « Croissance et Signalisation », Centre National de la
Recherche Scientifique UMR8253, Institut National de la Santé et de la Recherche Médicale U1151, Institut Necker
Enfants Malades, Université de Paris, Paris, France; 4Department of Pediatric Surgery, Hôpital Clocheville, Tours, France;
5
Department of Pediatric Surgery, Hôpital des Enfants, Limoges, France; 6Department of Pediatric Anesthesia, Hôpital
Necker-Enfants Malades, Assistance Publique-Hôpitaux de Paris, Paris, France; 7Department of Pathology, Hôpital
Necker-Enfants Malades, Assistance Publique-Hôpitaux de Paris, Paris, France; 8Department of Pediatric and Adolescent
Oncology, Institut Gustave Roussy, Villejuif, France; 9Department of Pediatric Oncology SIREDO Oncology Center (Care,
Innovation and Research for Children and AYA with Cancer), Institut Curie, PSL Research University, Paris, France;
10
Department of Pediatric Radiology, Hôpital Necker-Enfants Malades, Assistance Publique-Hôpitaux de Paris, Paris,
France

ABSTRACT Neuroblastoma Group (SIOP/SIOPEN) protocols. Indica-

Background. While robotics has become commonplace in tions were approved by a certified tumor board.
adult oncology, it remains rare in pediatric oncology due to Results. Overall, 100 tumors were resected during 93
the rarity of childhood cancers. We present the results of a procedures (abdomen, 67%; thorax, 17%; pelvis, 10%;
large nationwide experience with robotic oncology, with retroperitoneum, 6%) in 89 children (56 girls). The median
the aim of providing practical and feasible guidelines for age at surgery was 8.2 years (range 3.6–13); 19 children
child selection. (21%) harbored germinal genetic alterations predisposing
Methods. This was a prospective analysis performed over to cancer. No intraoperative tumor ruptures occurred.
a period of 4 years. Treatment was delivered according to Seven conversions (8%) to an open approach were per-
the Société Internationale d’Oncologie Pédiatrique/Inter- formed. Neuroblastic tumors (n = 31) comprised the main
national Society of Paediatric Oncology Europe group (18 neuroblastomas, 4 ganglioneuroblastomas, 9
ganglioneuromas) and renal tumors comprised the second
largest group (n = 24, including 20 Wilms’ tumors). The
remaining 45 tumors included neuroendocrine (n = 12),
adrenal (n = 9), germ-cell (n = 7), pancreatic (n = 4),
Thomas Blanc and Pierre Meignan have contributed equally to this thymic (n = 4), inflammatory myofibroblastic (n = 4), and
work. different rare tumors (n = 5). Overall, 51 tumors were
malignant, 2 were borderline, and 47 were benign. The
 Society of Surgical Oncology 2021 median hospital stay was 3 days (2–4), and five postoper-
First Received: 12 July 2021 ative complications occurred within the first 30 days.
Accepted: 21 August 2021; During a median follow-up of 2.4 years, one child (Wilms’
Published Online: 14 September 2021 tumor) presented with pleural recurrence. One girl with
T. Blanc, MD, PhD Wilms’ tumor died of central nervous system metastasis.
e-mail: [email protected]
1316 T. Blanc et al.

Conclusions. Robotic surgery for pediatric tumors is a METHODS

safe option in highly selected cases. Indications should be
discussed by tumor boards to avoid widespread and Robotic-assisted tumor resections were performed by
uncontrolled application. seven senior surgeons in three different university chil-
dren’s hospitals. This prospective observational study
(NCT03274050) received approval from the appropriate
The incidence of global childhood cancer is estimated at Institutional Review Boards and an independent Ethics
400,000 cases per year,1 with 15,780 new cases per year in Committee (Comité de Protection des Personnes Ile de
the US, according to the American Cancer Society.2 Each France VII).
type of tumor can be considered a rare disease. Pediatric At diagnosis, the tumor extent was assessed by com-
cancers also differ from adult cancers in that they are puted tomography and/or magnetic resonance imaging. For
mainly embryonal tumors with high-speed growth and neuroblastic tumors (NBTs), a metaiodobenzylguanidine
huge volume at diagnosis. Treatment usually associates scan was performed at diagnosis to evaluate the disease
neoadjuvant chemotherapy, surgery, and radiotherapy as extension; image-defined risk factors (IDRFs) were retro-
needed. As a result of advances in treatment, the mean spectively established both at diagnosis and after
overall survival is about 80%,3 whatever the tumor type. In neoadjuvant chemotherapy.12 Children received treatment
this context, in the last 15 years a program of de-escalation according to the ongoing International Society of Pediatric
therapy has involved organ-sparing surgery and minimally Oncology (SIOP) protocols. The indications for RAS were
invasive surgery (MIS),4 although open surgery remains based on the size and localization of the tumor, the
the gold standard for most pediatric tumors.5 appraisal of the surgical risk factors according to image
MIS was used first with diagnostic procedures and then analysis with the radiologists, and the surgeon’s experi-
with tumors that may be fragmented, such as neuroblas- ence. Indications were validated by the multidisciplinary
tomas.6 Oncological principles must be reproduced in MIS tumor board, certified by the French National Cancer
as in open surgery (negative margins, prevention of tumor Institute. Surgery was conducted according to the recom-
spillage, extensive lymph node dissection when required) mendations of the specific protocols.
to maintain excellent oncological outcomes.7 In 2010, the Data on demographics, imaging at diagnosis, neoadju-
International Pediatric Endosurgery Group provided vant chemotherapy, preoperative imaging, surgical
guidelines for laparoscopic treatment of adrenal masses in procedure, postoperative complications, histology, adju-
children,8 and, more recently, the UMBRELLA protocol vant treatment, oncologic outcomes, and follow-up were
dedicated a section for indications and contraindications of collected in a prospective database, but data on race/eth-
MIS in pediatric renal tumors.5 nicity is unavailable as the information was not included in
The advantages of robotic-assisted surgery (RAS) in the database. Some patients presented with multiple
addition to laparoscopic surgery are their three-dimen- tumors, for which surgery involved one- or two-step
sional (3D) vision, seven degrees of freedom, tremor procedures.
filtration, and precise camera control.9 The use of robotics To better describe the proportion between the preoper-
in adult oncology has become commonplace but the indi- ative tumor volume and the patient’s age for NBTs and
cations in pediatric oncological surgery are still renal tumors, we provided a ratio between ellipsoid tumor
controversial (the absence of haptic feedback may increase volume (ETV) [width (mm) 9 length (mm) 9 height (mm)
the risk of tumor rupture).10 No recommendations are 9 0.52] and estimated patient blood volume (EPBV; 75
available in North America or Europe to regulate the use of mL/kg for children over age 3 months), a parameter
robotic surgery in pediatric oncology.11 directly related to patient age.13
In this study, we report a large multicentric series of Three or four robotic ports and one Air Seal assistant
pediatric patients with abdominal, thoracic, and pelvic port (Conmed, Utica, NY, USA) were used. Pressure was
tumors removed by RAS. We evaluated the feasibility and maintained at 10–12 mmHg for abdominal and retroperi-
safety of this approach in pediatric oncology, with the aim toneal procedures, and 4 mmHg for thoracic procedures.
of providing a new set of practical and feasible guidelines The tumor was removed in an endoscopic bag and exteri-
for patient selection and technical key points to pediatric orized via the slightly enlarged optical port or via a
surgeons. Given the rarity of pediatric cancers, and thus the Pfannenstiel incision to avoid the need for tissue morcel-
low caseload even in large centers, it is critical to avoid any lation for renal tumors.
inappropriate use of this technique and to maintain good
oncological results while decreasing the burden of surgical
treatment.
Robotic Surgery for Pediatric Tumors 1317

Statistical Analysis benign; 21% of children harbored germinal genetic alter-

ations predisposing to cancer (Table 1).
Data are expressed as median (interquartile range [25th– Thirty-six children received neoadjuvant chemotherapy,
75th quartiles]) for continuous variables and number including all those with WT (n = 20), 1 with renal undif-
(percentage) for categorical variables. ferentiated sarcoma, 10 with neuroblastoma, 4 with
malignant GCTs, and 1 with rhabdomyosarcoma. One child
RESULTS received anti-anaplastic lymphoma kinase treatment and
one child received corticosteroid therapy for an inflam-
From 2016 to 2020, 100 tumors were resected during 93 matory myofibroblastic tumor. Surgery was performed as a
procedures in 89 children (56 girls). These procedures primary procedure in all other children.
represent 19% of the oncological case load of the three The median age at surgery was 8.2 years (3.6–13), and
centers. NBT tumors were the most frequent group of the youngest child was aged 5 months. One-third of the
tumors (n = 31) in 28 children: 18 neuroblastomas, 4 children were \ 5 years of age, one-third were 5–11 years
ganglioneuroblastomas intermixed, and 9 ganglioneuromas of age, and one-third were [ 11 years. The median weight
(Table 1). Only two patients had an MYCN-amplified was 26 kg (15–47), with the smallest child weighing
neuroblastoma. Tumors were located in the adrenal region 4.6 kg. Twelve children weighed \ 10 kg.
(n = 13, 42%), posterior mediastinum (n = 10, 32%), The median ETV/EPBV ratio was 0.8% (0.2–1.4%) for
presacral region (n = 2, 7%), around the renal pedicle WT, and 0.2% (0.1–0.5%) for NBT (Figs. 1, 2).
(adrenal healthy; n = 2, 7%), and other locations, i.e. inter-
aorticocaval space (n = 1), zuckerkandl ganglia (n = 1), Surgery
latero-aortic (n = 1), and paravaginal region (n = 1). At
preoperative imaging work-up, nine patients had one IDRF Procedures were performed by robotic-assisted laparo-
(contact with the renal pedicle, n = 8; encasement of the scopy for abdominal (n = 62; 67%) or pelvic tumors
renal pedicle, n = 1) and one patient presented with two (n = 9; 10%), robotic-assisted retroperitoneoscopy (n = 6;
IDRFs (encasement of the vena cava and contact with the 6%), or robotic-assisted thoracoscopy (n = 16; 17%). The
renal pedicle). These 10 tumors with IDRFs represented most common procedure was adrenalectomy (n = 28).
32% of the NBTs. Five children had bilateral adrenalectomy: synchronous
The second most frequent group was renal tumors (Carney complex, n = 2; McCune–Albright, n = 2) or
(n = 24), including 20 Wilms’ tumors (WTs), while the metachronous (von Hippel–Lindau, n = 1). Renal tumors
remaining 45 tumors included the following. (n = 24) were treated mainly by total nephrectomy
(n = 18). Six partial nephrectomies were performed using
• Twelve neuroendocrine tumors (six pheochromocy-
a retroperitoneal approach. Eight of 17 WTs scheduled for
tomas and six paragangliomas: one aortocaval, one pre-
total nephrectomy had tumor extension beyond the ipsi-
renal, one bladder, and three latero-aortic).
lateral border of the spinal column, but only one crossed
• Nine other adrenal tumors (one adrenocortical ade-
the midline. Central pancreatectomy with Roux-Y pan-
noma, two bilateral Carney complex, and two bilateral
creaticojejunostomy (n = 3) and distal pancreatectomy
McCune–Albright).
(n = 1) was performed.
• Seven germ-cell tumors (GCTs): two mature teratomas
Seven (8%) conversions to the open approach occurred,
and five malignant GCTs (one seminoma stage III; four
with one emergency undocking for renal vein bleeding
non-seminomatous tumors: one stage II and three stage
(Table 2). Of note, five of seven converted cases were renal
III).
tumors. Median operative time for non-converted cases
• Four pancreatic tumors (one neuroendocrine tumor and
was 215 min (156–282) and median console time was
three focal hyperinsulinisms).
164 min (105–221). Seven patients received intraoperative
• Four thymic tumors (one thymoma requiring comple-
transfusion with red blood cells. Major intraoperative
tion of the thymectomy (two procedures), one
bleeding occurred in only one patient, who subsequently
myasthenia, and one multiple endocrine neoplasia, type
required emergency undocking. The remaining six patients
1).
had received chemotherapy and were anemic (hemoglobin
• Four inflammatory myofibroblastic tumors.
7.8–9.5 g dL-1) at the time of surgery (two NBTs and four
• One of each of the following: embryonal rhab-
WTs).
domyosarcoma, neurofibroma, bronchial carcinoid
No acute hypertensive crisis occurred during
tumor, leiomyoma, and lipoma.
pheochromocytoma resection, and no adjacent organ
Overall, 51 tumors were malignant, 2 were borderline injury, positioning-related injury, or injury due to robotic
(nephrogenic rest and metanephric adenoma), and 47 were arms occurred.
TABLE 1 Tumor histology, stage, genetic predisposition syndrome to cancer, and local recurrence of 100 tumors in children
1318

Patient (n) Tumor (n) Localization Stage Genetic predisposition Local recurrence
syndrome to cancer

Neuroblastic tumors (n = 28)

Neuroblastoma—stroma poor (n = 18) 18 Adrenal (n = 12) L1 n=6
Posterior mediastinum (n = 4) L1-M n=4
Pelvis (n = 1) L1-MS n=1
Zuckerkandl (n = 1)
L2 n=2
L2-M n=4
L2-MS n=1
Ganglioneuroblastoma—intermixed (n = 4) 4 Post mediastinum (n = 3) L1 n=4
Paravaginal (n = 1)
Ganglioneuroma (n = 6) 9 Posterior mediastinum (n = 3) L1 n=5 Phox2B (n = 1)
Renal pedicle (n = 2) L2 n=4
Inter-aorticocaval (n = 1)
Latero-aortic (n = 1)
Pelvis (n = 1)
Adrenal (n = 1)
Renal tumors (n = 24)
Wilms’ tumor (n = 20) 20 Stage 1 n = 12 n=1
Stage 2 n=5
Stage 3 n=3
Intermediate risk n = 15
High risk n=5
Tubulopapillary carcinoma (n = 1) 1 Stage 1, type II (Fuhrman III)
Undifferentiated sarcoma (n = 1) 1 Stage 2 n=1
Metanephric adenoma (n = 1) 1
Nephrogenic rest (n = 1) 1
Neuroendocrine tumors (n = 10)
Pheochromocytoma (n = 4) 6 VHL (n = 3)
Paraganglioma (n = 6) 6 Aortocaval (n = 1) VHL (n = 2)
Pre-renal (n = 1) SDHA (n = 1)
Bladder (n = 1) SDHB (n = 1)
Latero-aortic (n = 3)
Adrenal tumors (n = 5)
McCune-Albright (n = 2) 4 GNAS-1 (n = 2)
Carney complex (n = 2) 4 PRKAR1A (n = 2)
T. Blanc et al.
Table 1 (continued)
Patient (n) Tumor (n) Localization Stage Genetic predisposition Local recurrence
syndrome to cancer
Adrenocortical adenoma (n = 1) 1
Pancreatic tumors (n = 4)
Focal congenital hyperinsulinism (n = 3) 3 Body (n = 2) ABCC8 (n = 3)
Tail (n = 1)
Somatostatinoma (n = 1) 1 Body (n = 1) MEN-1 (n = 1)
Germ cell tumors GCT (n = 7)
Mature teratoma (n = 2) 2 Posterior mediastinum
Robotic Surgery for Pediatric Tumors

Paravaginal
Non-seminomatous GCT (n = 4) 4 RPLND/Ovary (n = 3) FIGO stage II (n = 1)
RPLND/Testis (n = 1) FIGO stage III (n = 3)
Seminomatous GCT (n = 1) 1 RPLND/Testis (n = 1) FIGO stage III
Thymus tumors (n = 3)
MEN-1 (n = 1) 1 MEN-1 (n = 1)
Thymoma (n = 1) 2 (1 totalization) Stage Ia, type B1
Myasthenia (n = 1) 1
Inflammatory myofibroblastic tumors (n = 3) 4 (1 totalization) Stomach
Pelvis
Greater omentum
Embryonal rhabdomyosarcoma (n = 1) 1 Broad ligament IRS III
Neurofibroma [NF1] (n = 1) 1 Pre-sacral NF1 (n = 1)
Bronchial carcinoid tumor (n = 1) 1 T2N1M0
Leiomyoma (n = 1) 1 Mesocolon PTCH1 (n = 1)
Lipoma (n = 1) 1 Broad ligament
ABCC8 ATP-binding cassette, subfamily C, member 8, FIGO International Federation of Gynecology and Obstetrics, GNAS1 guanine-nucleotide binding protein a-subunit, INRG International
Neuroblastoma Risk Group, IRDFs image-defined risk factors, IRS-III Intergroup Rhabdomyosarcoma Study, L1 stage L1 (locoregional tumor without IDRFs), L2 stage L2 (locoregional tumor
with one or more IDRFs), M stage M (distant metastatic disease, except MS), MEN-1 multiple endocrine neoplasia type 1, MS stage MS (INRG stage L1 or L2 tumor with metastatic disease
confined to the skin and/or liver and/or bone marrow), NF1 neurofibroma type 1, PHOX2B paired-like homeobox 2B, PRKAR1A protein kinase A type I-a regulatory subunit, PTCH1 Patched 1,
RPLND retroperitoneal lymph node dissection, SDHA Succinate Dehydrogenase Complex Flavoprotein Subunit A, SDHB Succinate Dehydrogenase Complex Iron Sulfur Subunit B, VHL von
Hippel-Lindau
1319
1320 T. Blanc et al.

FIG. 1 Evaluation of Wilms’ 5

tumor volume/patient age. 4.5
EPBV estimated patient blood 4 conversion
volume (75 mL/kg for children
over age 3 months), ETV 3.5

ETV/EPBV (%)
ellipsoid tumor volume [width 3
(mm) 9 length (mm) 9 height 2.5
(mm) 9 0.52] 2 Emergency
1.5 undocking
1
0.5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Case number

FIG. 2 Evaluation of 1.8

neuroblastic tumor 1.6
volume/patient age. EPBV 1.4
estimated patient blood volume
ETV/EPBV (%)

1.2
(75 mL/kg for children over age
1
3 months), ETV ellipsoid tumor
volume [width (mm) 9 length 0.8 conversion
(mm) 9 height (mm) 9 0.52] 0.6
0.4
0.2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Case number

TABLE 2 Converted procedures

Procedure Histology Cause

Right nephrectomy Wilms Renal vein injury

Right nephrectomy Wilms Misdiagnosed renal vein tumor thrombus
Right nephrectomy Wilms Poor respiratory tolerance after diaphragmatic resection
Left nephrectomy Wilms Sliding Hem-O-Lok clip on renal vein (emergency undocking)
Left nephrectomy Sarcoma Difficult renal hilum dissectiona
Thoracic tumor resection Neuroblastoma Difficult dissection (narrow space)
Retroperitoneal tumor resection Ganglioneuroma Difficult dissection (vascular involvement)
NTRK neurotrophic tyrosine receptor kinase
a
Due to the nature of the tumor and the anti-NTRK inhibitor treatment, there was a thick coat around the renal pedicle that prevented the isolation
of vessels for ligation

No intraoperative tumor rupture occurred during adrenalectomy for Cushing syndrome due to the need for
robotic-assisted dissection. Two spillages were reported medical therapy. The median length of hospitalization for
with WT: one renal vein thrombus discovered after renal non-converted procedures (n = 86) was 3 days (2–4).
vein control and one after conversion, owing to a tumor
rupture during dissection of the component extending into Outcomes
the liver (see details regarding this patient in the Outcomes
section below). Five postoperative surgical complications were
One-third of children were discharged at day 1 or 2; observed within the first 30 days in four patients and
however, a longer hospital stay was required after complex subsequently required an additional procedure (Clavien–
procedures (such as ‘en bloc’ radical nephrectomy, Dindo III). None of these complications were due to
splenectomy, and distal pancreatectomy) or after bilateral oncological problems (Table 3).
Robotic Surgery for Pediatric Tumors 1321

TABLE 3 Postoperative complications requiring a new procedure (Clavien–Dindo grade III)

Procedure Histology Complication Time Management

Neuroblastic thoracic Ganglioneuroma Pneumothorax Day 1 Chest tube

tumor resection
Total gastrectomy Inflammatory myofibroblastic tumor Anastomotic stenosis Day 8 Endoscopic balloon dilatation
Adhesions Day 30 Laparotomy
Adhesiolysis
Bilateral adrenalectomy Carney complex Retroperitoneal collection Day 10 Percutaneous drainage
Lobectomy Bronchial carcinoid tumor Pneumothorax Day 28 Chest tube

Three patients presented with oncological events. • One patient with a cystic lesion of the thymus under-
went partial thymectomy with a presumed diagnosis of
• One patient with WT infiltrating the liver required a
teratoma. Histology of the specimen revealed a thy-
diaphragmatic resection for a suspected extension. The
moma (B1), and the thymectomy was completed using
opening of the diaphragm was poorly tolerated during
a robotic approach according to tumor board
the robotic procedure. A conversion was performed,
recommendations.
and tumor rupture occurred during dissection of the
• One patient presented with recurrence of hypoglycemia
tumor extension in the liver. The patient received
after resection of focal congenital hyperinsulinism,
treatment as for stage III, along with postoperative
required completion of the pancreatectomy, and was
radiotherapy and chemotherapy. She presented with a
cured.
massive pleural recurrence on therapy, 4 months after
surgery, and required thoracoscopy for drainage and The median number of lymph nodes for WT was 6 (2–10).
biopsy for molecular analysis. The patient is currently All patients resumed the postoperative scheduled
receiving second-line chemotherapy, with complete chemotherapy on time and with no delay. At a median
response seen on imaging at the 15-month follow-up. follow-up of 2.4 years (1.5–3.4), all patients were alive,
• The patient with a multimetastatic renal sarcoma apart from one girl with WT who died from central nervous
(converted case) who received targeted therapy with a system metastasis.
neurotrophic tropomyosin-related receptor tyrosine
kinase (NTRK) inhibitor (LOXO101) presented with a DISCUSSION
new lesion 10 months after the first R0 surgery, which
was treated with another NTRK inhibitor (LOXO195) This report describes the largest prospective series of
and resected using an open approach. The patient is still RAS in pediatric oncology, with the aim of evaluating the
receiving targeted therapy and was in complete remis- feasibility and safety of robotic surgery in pediatric
sion at the 2.5-year follow-up. oncology. Our critical analysis demonstrated that proper
• One patient with bilateral nephroblastomatosis pre- case selection is key to maintaining good oncological
sented with a small nodule of the right kidney at 18 results while decreasing the burden of surgical treatment.
months after completion of treatment with vincristine/ Given the rarity of pediatric cancers, and thus the low
actinomycin. He underwent a partial nephrectomy caseload even in large pediatric centers, it is critical to
using a robotic-assisted retroperitoneal approach, with avoid any inappropriate use of this technique. We provide
clear margins. The patient presented with a new nodule clear and practical guidelines and limitations for proper
2 years later and underwent open surgery after neoad- case selection (Table 4), which should help surgeons when
juvant chemotherapy. He received radiotherapy and starting a robotic program. Pediatric surgery manages a
1 year of vincristine/actinomycin chemotherapy, and large variation of rare diseases. Despite laparoscopy being
was in complete remission at 3-year follow-up. developed more than 30 years ago for use in pediatric
patients, this variation explains why no randomized
Apart from these patients, no tumor recurrences or port-site
prospective studies have yet compared laparoscopic simple
metastases occurred.
non-cancer procedures with the open equivalent in chil-
Two patients required additional surgery.
dren. This problem becomes even more acute because even
the most common pediatric cancers, such as NBTs and
1322 T. Blanc et al.

TABLE 4 Guidelines for patient selection based on tumor histology and location
Robotic surgery may be Relative contraindications Formal contraindications
considered

Thoracic tumors Paravertebral neuroblastoma Age \ 2 years (limited access) Encasement of vessels
OR AND/OR
Tumor limited to the thymic Extension to the median mediastinum
bed (teratoma, thymoma) (pericardium, esophagus, trachea)
OR
Lung resection (single
metastasis)
Renal tumors Tumor not crossing the Tumor with a thin rim of normal Tumor crossing the midline
ipsilateral border of the spine parenchyma OR
column OR Tumor infiltrating extrarenal structures,
AND Tumor crossing the ipsilateral border of especially liver or diaphragm
Tumor with a thick rim of the spine column but not the midline infiltration
normal parenchyma OR OR
AND ETV/EPBV 1.5–2% Tumor with encasement of renal vessels
Tumor without any sign of (e.g. sarcomas, carcinomas)
infiltration of extrarenal OR
structures
ETV/EPBV [ 2%
AND
ETV/EPBV \ 1.5%
Neuroblastic tumors Paravertebral (thoracic, One or two IDRFs More than two IDRFs
abdominal, or pelvic) OR OR
neuroblastoma without
Paravertebral (thoracic, abdominal, or Any IDRF number involving median
foramen extension
pelvic) neuroblastoma with foramen vessels (celiac artery, superior
OR extension but without a spinal mesenteric artery) and/or both renal
Adrenal tumor component pedicles
OR OR OR
Neuroblastoma of the ETV/EPBV 1–2% Paravertebral (thoracic, abdominal, or
zuckerkandl ganglia pelvic) neuroblastoma with foramen
OR extension and an intraspinal
component
ETV/EPBV \ 1%
OR
AND
ETV/EPBV [ 2%
No IDRF12
Paragangliomas Encasement of major vessels
Pheochromocytomas
Adrenocortical All adrenocortical carcinomas
carcinomas
Solid All solid pseudopapillary tumors
pseudopapillary
neoplasms
ETV/EPBV ellipsoid tumor volume [width (mm) 9 length (mm) 9 height (mm) 9 0.52] and estimated patient blood volume (75 mL/kg for
children aged [ 3 months), IDRF image-defined risk factor

WTs, are still rare and heterogenous in their presentation. series, Cundy et al. identified 40 robotic-assisted resections
Contrary to adult oncology, these characteristics preclude of solid tumors (malignant in 35%) in children \ 18 years
any attempt at controlled clinical trials of children com- of age (85% were teenagers) (Table 5). Two oncological
paring laparoscopic tumor resections with the open complications occurred: one tumor spillage and one
equivalent, as confirmed by a recent Cochrane review.11 incomplete resection. In 2018, in a stepwise collaboration
Regarding robotics in pediatric cancer, Meehan published a at a children’s hospital, one adult and one pediatric urol-
review highlighting the scarcity of studies using robotic ogist performed RAS for renal tumors in six teenagers and
surgery.9 In 2014, in a review of 21 case reports and 2 case two children.14 Finally, our group reported a preliminary
Robotic Surgery for Pediatric Tumors 1323

TABLE 5 Published cases series of robotic surgery in pediatric oncology

Study, year No. of procedures Median age, years Conversion rate (%)

Meehan and Sandler 200831 14 NA 29

Meignan et al. 201832 11 7.6 8
Varda et al. 201814 8 14 0
15
Blanc et al. 2019 10 5 30
Navarrete Arellano and Garibay González 201924 5 NA NA
Present series 93 8.2 8
NA not available

study comparing open and robotic surgery for WT, with no to not only young children but also to some tumors that
differences in terms of age at surgery, tumor stage, lymph have never been described in MIS, such as paravaginal
node sampling, and operative time, but a longer hospital ganglioneuroma or teratoma of the thymic region.
stay for open surgery (p = 0.01).15 As expected, the homogenous cancer groups were the
In contrast, the literature on pediatric robotic urology is most frequent solid malignant tumors in the thorax and
abundant, and our robotic program was originally set up abdomen (e.g. NBTs and WTs; 31% and 20% of the whole
mainly for these indications.16 Because of the fear of series, respectively). A large number of NBT cases of MIS
rupture or spillage due to the absence of haptic feedback, has been reported, especially for adrenal lesions.18 The
we were very cautious and did not anticipate including so robot was advantageous for huge adrenal tumors with
many patients with malignant tumors. However, these IDRFs,12 especially on the renal pedicle, and for pelvic
drawbacks were balanced by the availability of three tumors. The indication accounted for the neuroblastoma
robotic arms, the 3D-HD vision, and the camera control by risk group, considering that the robotic approach should be
the surgeon. High-definition stereoscopic optics were able to achieve complete resection, especially in localized
indeed reported to contribute to enhanced recognition of high-risk neuroblastomas (MYCN-amplified). Hence, only
the operative field and subsequent tissue manipulation.17 two children with MYCN amplification had surgery in this
The development of our unique experience was also series. It was also a good option for large painful and/or
possible with the historic high case load in pediatric compressive benign ganglioneuromas with IDRFs because
oncology surgery and a strong collaboration between it definitively solved the problem with limited burden.
pediatric oncologists, radiologists, and pediatric surgeons The use of MIS for WTs was reported as early as
with proficiency in MIS. The dual console allowed simul- 2004,19 but the risk of tumor rupture and spillage, which
taneous use by an experienced pediatric oncology surgeon leads to upstaging, intensified chemotherapy and radio-
and the robotic operating surgeon. Each indication of RAS therapy, was for a long time an obstacle to the development
was validated by the multidisciplinary tumor board and of this approach. This risk prevented us from performing
certified by the French National Institute of Cancer (https:// laparoscopy for WT before the start of the robotic pro-
www.oncorif.fr/prises-en-charge-specifiques/adolescents-e gram.15 Owing to the increased attractiveness of MIS for
t-jeunes-adultes-aja/rcppi_canpedif2/). This rigorous WT, the UMBRELLA SIOP-RTSG protocol included a
patient selection allowed us to develop expertise without contraindication for MIS.20 In a recent analysis of a mul-
compromising patient safety, and to extend boundaries. ticentric series of 50 patients with WT, MIS was feasible in
Despite the high activity, as expected indications were up to 55% of patients, with results comparable with those
heterogeneous in terms of age, weight, tumor type, volume, with open surgery.21 Of note, 32% of indications violated
and localizations. With this specificity linked to the unique the UMBRELLA protocol because tumors extended
features of pediatric oncology, every procedure is unique beyond the ipsilateral border of the spinal column. This
and requires strong involvement and good communication situation is not considered a formal contraindication in our
among the whole robotic team. experience (47% in this series), but we did consider the
Patient age and weight, which may be a limiting factor tumor crossing the midline as a formal contraindication
in MIS pediatric oncology, did not seem a contraindication because it prevents the control of the renal vessels, as
in this series. Thus, our series differs from the current lit- illustrated by the emergency undocking. The other con-
erature because 72% of children were \ 12 years of age traindications indicated in the protocol, such as liver or
and 13% weighed \ 10 kg. The added value of the robotic diaphragm infiltration, should be respected, as illustrated
system was clearly an advantage to expand the indications
1324 T. Blanc et al.

by the case with pleural recurrence. Considering the low earlier. MIS alleviates the stigmatizing aspect of the scar,
number of other tumor types in this series, it is difficult to which grows with the child and is associated with increased
develop precise guidelines for each kind of lesion. distress due to negative self-esteem or self-image.29,30
We did not observe any specific robotic complications, The future of surgery will be image-guided surgery.
such as adjacent organ injury, positioning-related injury, or Robotic surgery will greatly facilitate the integration of
injury due to robotic arms. The objectives expected images to plan the surgical strategy and to anticipate dif-
regarding resection were always reached and confirmed by ficulties. These advances have been developed for adult
postoperative imaging. No recurrence linked to the surgical surgery, but pediatric genetic and developmental diseases
approach was observed (except for WT with liver infiltra- have not benefitted from advanced simulation, computer
tion). Of note, surgery was not used for adrenocortical assistance, and automation. This is a detriment because this
carcinoma or solid pseudopapillary neoplasms of the pan- population requires particular attention in terms of the
creas because, in contrast to WT, the literature for these small size of critical organs and structures compared with
tumors is poor and only adverse events (high risk of tumor adults and the great need to reduce the sequelae of surgical
rupture and increase in treatment intensity) were repor- procedures to avoid long-term morbidity. Because of the
ted.22,23 Even if we do not consider the 8% conversion rate rarity of the disease and the specific nature of the pediatric
as a failure of the technique, such a rate of conversion population, pediatric oncology has been pioneering per-
shows that this technique is feasible in highly selected sonalized medicine. RAS offers precision surgery with
cases. The conversion rate in this series is higher than the many benefits if indications are strictly regulated by a
conversion rate of 4% reported by Navarrete Arellano certified tumor board.
et al.24 Of note, the conversion rate for our entire robotic
cases was 3%. This situation shows a high concern to not CONCLUSIONS
upstage the local tumor by overestimating our skill and/or
the performance of robotic surgery. It also compares Robotic surgery for pediatric tumors is a safe option in
favorably with the previously published case series (see highly selected cases. Indications must be discussed in the
Table 5). The conversion rate was higher at the beginning framework of certified tumor boards with medical oncol-
of the learning curve for renal tumors. Therefore, we rec- ogists, pathologists, and radiologists to avoid widespread
ommend starting the experience with smaller tumors and and uncontrolled application. Surgery must be performed
converting in cases of difficult dissection, because the main with respect to oncological surgical rules. As with any new
objective is to respect the oncological surgical principles. emerging technique, careful patient selection is crucial and
In Table 4, we provide guidelines for patient selection further evidence must be sought to confirm its limits and
based on tumor histology and location. indications. Children with a genetic predisposition to can-
An increasing number of pediatric cancers due to cer are at high risk of recurrence or secondary tumor, and
genetic predisposition syndromes have been described. A thus iterative surgery. A minimally invasive approach such
recent study of more than 1000 patients estimated that as robotic surgery seems a good option to expand the
about 8% of children with cancer harbor a hereditary pre- possibilities of complex resection in pediatric cancer
disposition.25 Multiple and/or metachronous tumors compared with classic endoscopic procedures, while min-
requiring reoperation are likely in these patients, who imizing the burden of treatment in these patients with an
represented 21% in our series. Paragangliomas are already improved condition.
demonstrated in this way: recurrence is associated with
young age and genetic germinal alteration of one of the ACKNOWLEDGMENT The authors thank Deborah Nock (Med-
ical WriteAway, Norwich, UK) for medical writing support, and
known susceptibility genes (SDHB, VHL RET, HIF2A,
Caroline Elie, Sandra Colas, Emilie Ervilus, Mégane Régina, and
SDHC, SDHD or NF1).26 About 75% of pediatric cases Sarah Bouchard (URC/CIC Paris Descartes Necker Cochin, Paris,
show an accumulation of both risk factors27 and the risk of France) for their help in the protocol management. This project was
recurrence within 10 years after surgery is about 50%.26 funded by a grant from Necker-Enfants Malades Hospital (Assistance
Publique–Hôpitaux de Paris, Clinical Research and Innovation Del-
The robotic approach is a good option to resect iterative
egation) and is registered at ClinicalTrials.gov (ID: NCT03274050).
lesions that will appear in this genetic context with mini-
mal burden,28 thus lightening the psychological burden of a
genetic predisposition syndrome. FUNDING No sources of funding were used to assist in the
Treatment for pediatric cancer is mainly multimodal, preparation of this study.
associating chemotherapy, surgery, and sometimes radio-
therapy. The robotic approach is not only a technical
DISCLOSURES Thomas Blanc is an official proctor for Intuitive
improvement but also allows the length of hospital stay to Surgical. Laurent Fourcade has a financial relationship with Intuitive
be reduced,15 thus resuming postoperative chemotherapy
Robotic Surgery for Pediatric Tumors 1325

Surgical as a proctor, helping teams performing robotic surgery. 15. Blanc T, Pio L, Clermidi P, et al. Robotic-assisted laparoscopic
Pierre Meignan, Nicolas Vinit, Quentin Ballouhey, Lucas Pio, Car- management of renal tumors in children: preliminary results.
men Capito, Caroline Harte, Fabrizio Vatta, Louise Galmiche- Pediatr Blood Cancer. 2019;66(Suppl 3):e27867. https://doi.org/
Rolland, Véronique Minard, Daniel Orbach, Laureline Berteloot, 10.1002/pbc.27867.
Cécile Muller, Jules Kohaut, Aline Broch, Karim Braik, Aurélien 16. Cundy TP, Harley SJD, Marcus HJ, Hughes-Hallett A, Khurana
Binet, Yves Heloury, Hubert Lardy, and Sabine Sarnacki have no S. Global trends in paediatric robot-assisted urological surgery: a
conflicts of interest to declare. bibliometric and Progressive Scholarly Acceptance analysis. J
Robot Surg. 2018;12(1):109–15. https://doi.org/10.1007/s11701-
017-0703-3.
REFERENCES 17. Shakir F, Jan H, Kent A. 3D straight-stick laparoscopy versus 3D
robotics for task performance in novice surgeons: a randomised
1. Ward ZJ, Yeh JM, Bhakta N, Frazier AL, Atun R. Estimating the crossover trial. Surg Endosc. 2016;30(12):5380–7. https://doi.org/
total incidence of global childhood cancer: a simulation-based 10.1007/s00464-016-4893-y.
analysis. Lancet Oncol. 2019;20(4):483–93. https://doi.org/10.1 18. Irtan S, Brisse HJ, Minard-Colin V, Schleiermacher G, Canale S,
016/S1470-2045(18)30909-4. Sarnacki S. Minimally invasive surgery of neuroblastic tumors in
2. Ward E, DeSantis C, Robbins A, Kohler B, Jemal A. Childhood children: Indications depend on anatomical location and image-
and adolescent cancer statistics, 2014. CA Cancer J Clin. defined risk factors. Pediatr Blood Cancer. 2015;62(2):257–61. h
2014;64(2):83–103. https://doi.org/10.3322/caac.21219. ttps://doi.org/10.1002/pbc.25248.
3. Gatta G, Mallone S, van der Zwan JM, et al. Cancer survival in 19. Duarte RJ, Dénes FT, Cristofani LM, Giron AM, Filho VO, Arap
Europe 1999–2007 by country and age: results of EUROCARE- S. Laparoscopic nephrectomy for Wilms tumor after
5—a population-based study. Lancet Oncol. 2014;15(1):23–34. h chemotherapy: Initial experience. J Urol. 2004;172(4 I):1438–40.
ttps://doi.org/10.1016/S1470-2045(13)70546-1. https://doi.org/10.1097/01.ju.0000138230.51134.65.
4. Christison-Lagay ER, Thomas D. Minimally invasive approaches 20. Van Den Heuvel-Eibrink MM, Hol JA, Pritchard-Jones K, et al.
to pediatric solid tumors. Surg Oncol Clin N Am. Position paper: rationale for the treatment of Wilms tumour in the
2019;28(1):129–46. https://doi.org/10.1016/j.soc.2018.07.005. UMBRELLA SIOP-RTSG 2016 protocol. Nat Rev Urol.
5. Vujanić GM, Gessler M, Ooms AHAG, et al. The UMBRELLA 2017;14(12):743–52. https://doi.org/10.1038/nrurol.2017.163.
SIOP–RTSG 2016 Wilms tumour pathology and molecular 21. Bouty A, Blanc T, Leclair MD, et al. Minimally invasive surgery
biology protocol. Nat Rev Urol. 2018;15(11):693–701. https://d for unilateral Wilms tumors: multicenter retrospective analysis of
oi.org/10.1038/s41585-018-0100-3. 50 transperitoneal laparoscopic total nephrectomies. Pediatr
6. Mirallié E, Leclair MD, Lagausie P, et al. Laparoscopic Blood Cancer. 2020;67(5):e28212. https://doi.org/10.1002/pbc.
adrenalectomy in children. Surg Endosc. 2001;15(2):156–60. h 28212.
ttps://doi.org/10.1007/s004640000335. 22. Irtan S, Galmiche-Rolland L, Elie C, et al. Recurrence of solid
7. Fuchs J. The role of minimally invasive surgery in pediatric solid pseudopapillary neoplasms of the pancreas: results of a nation-
tumors. Pediatr Surg Int. 2015;31(3):213–28. https://doi.org/10. wide study of risk factors and treatment modalities. Pediatr Blood
1007/s00383-015-3660-9. Cancer. 2016;63(9):1515–21. https://doi.org/10.1002/pbc.25986.
8. Group IP. IPEG guidelines for the surgical treatment of adrenal 23. Wu K, Liu Z, Liang J, et al. Laparoscopic versus open
masses in children. J Laparoendosc Adv Surg Tech. adrenalectomy for localized (stage 1/2) adrenocortical carcinoma:
2010;20(2):VII–IX. https://doi.org/10.1089/lap.2010.9999. experience at a single, high-volume center. Surgery.
9. Meehan JJ. Robotic surgery for pediatric tumors. Cancer J. 2018;164(6):1325–9. https://doi.org/10.1016/j.surg.2018.07.026.
2013;19(2):183–8. https://doi.org/10.1097/PPO.0b013e31828948 24. Navarete Arellano M, Garibay González F. Robot-assisted
6c. laparoscopic and thoracoscopic surgery: prospective series of 186
10. Cundy TP, Marcus HJ, Clark J, et al. Robot-assisted minimally pediatric surgeries. Front Pediatr. 2019;7:200. https://doi.org/10.
invasive surgery for pediatric solid tumors: a systematic review of 3389/fped.2019.00200.
feasibility and current status. Eur J Pediatr Surg. 25. Zhang J, Walsh MF, Wu G, et al. Germline mutations in pre-
2014;24(2):127–35. https://doi.org/10.1055/s-0033-1347297. disposition genes in pediatric cancer. N Engl J Med.
11. van Dalen EC, de Lijster MS, Leijssen LGJ, et al. Minimally 2015;373(24):2336–46. https://doi.org/10.1056/nejmoa1508054.
invasive surgery versus open surgery for the treatment of solid 26. Parasiliti-Caprino M, Lucatello B, Lopez C, et al. Predictors of
abdominal and thoracic neoplasms in children. Cochrane Data- recurrence of pheochromocytoma and paraganglioma: a multi-
base Syst Rev. 2015;1(1):CD008403. https://doi.org/10.1002/ center study in Piedmont, Italy. Hypertens Res.
14651858.CD008403.pub3. 2020;43(6):500–10. https://doi.org/10.1038/s41440-019-0339-y.
12. Brisse HJ, McCarville MB, Granata C, et al. Guidelines for 27. de Tersant M, Généré L, Freyçon C, et al. Pheochromocytoma
imaging and staging of neuroblastic tumors: consensus report and paraganglioma in children and adolescents: experience of the
from the international neuroblastoma risk group project. Radiol- French Society of Pediatric Oncology (SFCE). J Endocr Soc.
ogy. 2011;261(1):243–57. https://doi.org/10.1148/radiol. 2020;4(5):bvaa039. https://doi.org/10.1210/jendso/bvaa039.
11101352. 28. Krielen P, Stommel MWJ, Pargmae P, et al. Adhesion-related
13. Howie SRC. Les volumes des échantillons sanguins dans la readmissions after open and laparoscopic surgery: a retrospective
recherche en matière de santé Infantile: Examen des limites de cohort study (SCAR update). Lancet. 2020;395(10217):33–41. h
sécurité. Bull World Health Organ. 2011;89(1):46–53. https://doi. ttps://doi.org/10.1016/S0140-6736(19)32636-4.
org/10.2471/BLT.10.080010. 29. Pintér AB, Hock A, Kajtár P, Dóber I. Long-term follow-up of
14. Varda BK, Cho P, Wagner AA, Lee RS. Collaborating with our cancer in neonates and infants: a national survey of 142 patients.
adult colleagues: a case series of robotic surgery for suspicious Pediatr Surg Int. 2003;19(4):233–9. https://doi.org/10.1007/s00
and cancerous lesions in children and young adults performed in 383-002-0760-0.
a free-standing children’s hospital. J Pediatr Urol. 30. Kinahan KE, Sharp LK, Seidel K, et al. Scarring, disfigurement,
2018;14(2):182.e1-182.e8. https://doi.org/10.1016/j.jpurol.2018. and quality of life in long-term survivors of childhood cancer: a
01.003. report from the childhood cancer survivor study. J Clin Oncol.
1326 T. Blanc et al.

2012;30(20):2466–74. https://doi.org/10.1200/JCO.2011.39. tumors: a bicenter experience. J Robot Surg. 2018;12(3):501–8. h

3611. ttps://doi.org/10.1007/s11701-017-0773-2.
31. Meehan JJ, Sandler A. Pediatric robotic surgery: a single-insti-
tutional review of the first 100 consecutive cases. Surg Endosc. Publisher’s Note Springer Nature remains neutral with regard to
2008;22(1):177–82. https://doi.org/10.1007/s00464-007-9418-2. jurisdictional claims in published maps and institutional affiliations.
32. Meignan P, Ballouhey Q, Lejeune J, Braik K, Longis B, Cook
AR, et al. Robotic-assisted laparoscopic surgery for pediatric