Key words
cervical cancer - radical hysterectomy - LACC
Ramirez et al. published the results of their randomized study comparing minimally
invasive and abdominal radical hysterectomy for early cervical cancer (laparoscopic
approach to cervical cancer, LACC) in the New England Journal of Medicine [1]. The primary objective of this international randomized multicenter phase-III study
was to investigate the hypothesis that laparoscopic or robot-assisted radical hysterectomy
is noninferior to abdominal radical hysterectomy with regard to disease-free survival
(DFS) after 4.5 years. Secondary objectives were the rate of recurrence and overall
survival (OS).
The study was designed as a noninferiority study and aimed to compare a minimally
invasive arm with a standard abdominal arm. The planned number of cases was 740 patients,
who would subsequently be followed up for 4.5 years. If the disease-free survival
rate in the abdominal arm was 90%, the minimally invasive arm needed to have a power
of 87% to be classed as noninferior to open surgery, based on a 7.2% noninferiority
boundary for differences in disease-free survival. Patients with FIGO stage IA1 (with
LVSI), IA2 or IB1 (without lymph node metastasis) cervical cancer histologically classified
as primary squamous-cell carcinoma, adenocarcinoma or adenosquamous carcinoma, who
showed no signs of metastasis on PET/MRT/CT and underwent type II or III (Piver classification)
radical hysterectomy, were included in the study. Participating centers had to submit
10 documented cases of laparoscopic/robot-assisted radical hysterectomy and two non-edited
complete video recordings to the study committee. Every surgeon carried out both minimally
invasive and open surgery. Adjuvant chemotherapy, radiotherapy and radio-chemotherapy
was administered in accordance with standard local practice.
The study started in June 2008 and was stopped prematurely for safety reasons in June
2017 by the Data Safety & Monitoring Committee after 631 patients (85%) had been recruited
because of the significant inferiority of the laparoscopy arm.
Overall, with 33 centers participating worldwide, fewer than 20% of patients were
recruited in North American centers; the other patients were recruited in centers
in South America, India, China, Australia, Italy and Bulgaria. 312 women were randomized
to the abdominal arm and 319 women were randomized to the minimally invasive arm.
Inclusion characteristics, rates of intraoperative complications and rates of adjuvant
therapies were comparable for both study arms; only the rate of tumors with superficial
stromal invasion was higher in the minimally invasive arm.
After a median follow-up of 2.5 (0.0 – 6.3) years, 59.7% of the survival data at 4.5
years was available at the time of analysis (84% power): there were 27 recurrences
(41% of them in the fornix/pelvis) and 19 deaths in the minimally invasive group and
7 recurrences (43% of them in the fornix) and 3 deaths in the open surgery group.
The rate of disease-free survival after 4.5 years was 86% in the minimally invasive
group and 96.5% in the abdominal group (difference − 10.6%; 95% CI: − 16.4 to − 4.7;
p = 0.87 for noninferiority); in this respect the data did not provide evidence of non-inferiority. The results were consistent with those obtained
from 45 robot-assisted cases (difference − 10.6%; 95% CI: − 16.4 to − 4.7). The differences
in the per-protocol evaluation were also almost identical.
After 3 years, the percentage of disease-free patients (DFS) was significantly higher
(97.1%) in the standard abdominal arm compared to 91.2% in the laparoscopy arm (hazard
ratio [HR] 3.7; 95% CI: 1.6 – 8.6). The abdominal approach was significantly better
(99 vs. 93.8%) with regard to overall survival (OS: HR 6.0; 95% CI: 1.8 – 20.3), disease-specific
death (4.4 vs. 0.6%; HR 6.6; 95% CI: 1.5 – 29.0) and locoregional recurrence (94.3
vs. 98.3%; HR 4.3; 95% CI: 1.4 – 12.6). The results for robot-assisted and laparoscopic
surgery were similar.
Recurrence for tumors with diameters of 2 – 4 cm was around 5 times more frequent
than for tumors with diameters of less than 2 cm. However, compared with the respective
relative risk in the open surgery cohort, the minimally invasive cohort did not have
a lower relative risk with regard to the recurrence rate for the subgroup “tumor size
less than 2 cm” compared to the subgroup “tumor size 2 – 4 cm”. But the study was
not powered for further analysis (e.g., tumor size </> 2 cm, depth of invasion < 10 mm,
no lymphovascular invasion, node-negative).
In addition to this study, a retrospective evaluation of two North American Cancer
Registers (NCDB, SEER) was also published in the same issue of the NEJM [2]. Women with FIGO stage IA2 to IB1 cervical cancer treated with open surgery had
significantly better survival rates (4-year mortality 9.1 vs. 5.3%; HR 1.6; 95% CI:
1.2 – 2.2; p = 0.002). The introduction of minimally invasive radical hysterectomy
in 2006 coincided with a continuous and statistically significant reduction in survival
rates (mortality increased by 0.8% annually), whereas a continuous improvement in
the survival rates of women with FIGO stage 1A2 to IB1 cervical cancer had been noted
in preceding years.
It was not possible to establish a precise association between minimally invasive
surgery vs. open surgery and mortality in the subgroup with tumor diameters of less
than 2 cm and few deaths. However, the 95% confidence interval shown in the corresponding
graph includes the null value (HR 1.5; 95% CI: 0.7 – 3.0), meaning that there was
no significance at the alpha level of 0.05, i.e., there was no difference. The differences
between the subgroups with tumors larger than 2 cm were significant.
In their discussion of the LACC study the authors noted that the very good results
of their open surgery arm with its recurrence rate of only 2.4% after 4.5 years (DFS
of 97.6% instead of 90% as posited in the study hypothesis) was responsible for the
lack of noninferiority of the minimally invasive surgery arm (which had a recurrence
rate of 13% and a DFS of 87.1%) and attempted to qualify this argument by citing three
retrospective studies in which the 5-year DFS in the open surgery arm was between
93.3 and 94.4% [3], [4], [5]. However, in those studies, the DFS in the minimally invasive arm was between 90.5
and 92.8%, i.e. of a similar magnitude. The argument that after 2008 there was a learning
curve for the minimally invasive approach was countered by noting that the surgeons
in the study had extensive professional experience with this technique. The routine
use of uterine manipulators with the potential to disseminate tumor cells in the abdomen
during the laparoscopic opening of the vagina and the possibly negative effect of
CO2 insufflation were cited as potential reasons for the inferiority of minimally invasive
surgery. There is, in fact, a study in which vaginal colpotomy was associated with
a lower risk of recurrence than intraabdominal colpotomy [6].
In the summary the authors of the LACC study and US Cancer Registry Evaluation stated
that minimally invasive radical hysterectomy was associated with a higher rate of
recurrence and poorer overall survival compared to open radical hysterectomy. Previous
meta-analyses, which were based on retrospective studies, showed no significant superiority
for open surgery.
The results were not interpreted as signaling the end of minimally invasive surgery
to treat cervical cancer; instead, it was recommended that studies be done on how
to avoid the use of uterine manipulators and the dissemination of cancer cells by
ensuring more effective vaginal closure using a standardized approach.
The Uterus Commission of the AGO and the AGE notes that patients with FIGO stage IA1
(with LVSI), IA2, IB1 cervical cancer must be informed about the results of the LACC
study prior to making a decision about the planned means of access for radical hysterectomy.
