Direct comparison of 5-aminolevulinic acid photodynamic diagnosis and white light inspection for detecting peritoneal dissemination in advanced ovarian cancer: a pilot study

Abstract

Objective

Protoporphyrin IX, a metabolite of 5-aminolevulinic acid (5-ALA), emits red fluorescence. Photodynamic diagnosis (PDD) using 5-ALA has been successfully applied in fluorescence-guided surgery for various malignancies. This pilot study directly compared the diagnostic performance of conventional white light (WL) inspection and 5-ALA-based PDD for detecting intraperitoneal dissemination in advanced ovarian cancer (OC).

Methods

This prospective single-center study included patients with advanced OC undergoing staging laparoscopy. Each patient received oral 5-ALA (20 mg/kg) 4 hours preoperatively. Tumor and non-tumor locations were identified under WL and fluorescent light (FL). Biopsy specimens were collected from tumor locations visible under both WL and FL, only under WL, and only under FL and from non-tumor locations visible under both WL and FL. Pathological examination confirmed malignancy, and results were correlated with imaging findings.

Results

Seven patients were enrolled between January and March 2019 and 71 biopsies were performed. The diagnoses included four cases of high-grade serous carcinoma, one clear cell carcinoma, one grade 3 endometrioid carcinoma, and one pseudomyxoma peritonei. The sensitivity, specificity, positive predictive value, and negative predictive value for detecting malignancy under WL were 87.8%, 59.0%, 82.7%, and 68.4%, respectively, compared with 96.0%, 90.5%, 96.0%, and 90.5% under FL. Additional metastatic lesions were identified in three patients subjected to 5-ALA imaging. No severe 5-ALA-related adverse events were observed.

Conclusion

5-ALA-based PDD demonstrated higher diagnostic accuracy than WL for detecting intraperitoneal dissemination of advanced OC. Its ability to enhance tumor detection may improve surgical precision and outcomes.

Synopsis

This study demonstrated that 5-aminolevulinic acid (5-ALA)-based photodynamic diagnosis (PDD) improves intraoperative detection of intraperitoneal metastases in ovarian cancer. Compared with white light, 5-ALA-PDD showed higher diagnostic accuracy without adverse events, supporting its potential to enhance surgical precision and outcomes.

INTRODUCTION

Ovarian cancer (OC) is among the most lethal gynecological malignancies as it is often asymptomatic in its early stages, leading to the majority of patients being diagnosed only at advanced stages [1]. Despite advancements in treatment, the prognosis of advanced OC remains poor, with high recurrence rates and significantly reduced 5-year survival rates in late-stage disease [1]. Although the introduction of poly(ADP-ribose) polymerase inhibitors as maintenance therapy has improved the survival of some patients, the recurrence rate remains high, necessitating further efforts to improve outcomes [2].

Cytoreductive surgery combined with chemotherapy remains the standard treatment for advanced OC [3]. Achieving no visible residual tumors during surgery has been shown to be a critical factor in improving survival rates [3]. However, peritoneal dissemination, a typical feature of OC, often involves microscopic or small lesions that are challenging to detect using conventional imaging modalities or visual inspection, thereby limiting the effectiveness of surgical resection. Therefore, although diagnostic laparoscopy under conventional white light (WL) is sometimes performed to assess the peritoneal cavity, disseminated lesions are often flat and subtle, making them difficult to detect even under laparoscopic WL visualization [4].

Photodynamic diagnosis (PDD) using 5-aminolevulinic acid (5-ALA) has emerged as a promising method to address these challenges. Clinically, 5-ALA has been used to diagnose bladder cancer and brain tumors, demonstrating its effectiveness [5,6]. Once administered, 5-ALA is metabolized to protoporphyrin IX (PpIX) within the tumor cells, which then selectively accumulates in cancerous tissues [7]. When exposed to specific wavelengths of light, PpIX emits red fluorescence, allowing the detection of otherwise elusive lesions [7]. PDD has demonstrated high sensitivity for identifying microscopic and disseminated lesions in various cancers, making it a potential tool for enhancing the precision of cytoreductive surgery in OC [8,9]. In addition, other fluorescence-guided approaches, such as pafolacianine, a folate receptor-targeted near-infrared fluorescent agent, have been approved by the Food and Drug Administration (FDA) for intraoperative detection of OC, highlighting the growing interest in fluorescence-based imaging to enhance surgical precision [10]. Nevertheless, direct clinical comparisons between conventional WL inspection and 5-ALA-based PDD for detecting peritoneal dissemination in OC remain limited.

Accordingly, this pilot study aimed to directly compare the diagnostic performance of conventional WL inspection and 5-ALA-based PDD for detecting intraperitoneal dissemination in patients with advanced OC, and to evaluate the clinical utility and safety of orally administered 5-ALA during diagnostic laparoscopy.

MATERIALS AND METHODS

1. Study design

This single-center, open-label, prospective, exploratory study was approved by the Institutional Review Board of Saitama Medical University International Medical Center (approval number: 17-248). Written informed consent was obtained from all participants before their inclusion in the study. To assess the disease stage, all patients underwent contrast-enhanced chest-to-pelvis computed tomography and pelvic magnetic resonance imaging. Eligible participants were women aged 20 years or older with suspected stage IIIC–IVB epithelial ovarian, fallopian tube, or primary peritoneal carcinoma based on preoperative imaging and clinical assessment according to the 2013 International Federation of Gynecology and Obstetrics (FIGO) staging system. Additional inclusion criteria included an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2 and no evidence of other active malignancies. Exclusion criteria included intolerance to oral administration of 5-ALA, significant comorbidities such as cardiovascular disease, liver cirrhosis, poorly controlled diabetes mellitus, severe pulmonary fibrosis, or active interstitial pneumonia. Patients were also excluded if they had hepatic or renal dysfunction, a known hypersensitivity to porphyrin derivatives, or a diagnosis of porphyria.

The primary endpoint of this study was the diagnostic sensitivity and specificity of 5-ALA-based PDD in patients with epithelial ovarian cancer (EOC).

2. Surgical and 5-ALA-PDD procedure

All patients received an oral dose of 5-ALA (20 mg/kg) 4 hours before the start of surgery and were advised to avoid strong light exposure for 48 hours post-administration to minimize the risk of photosensitivity. Diagnostic laparoscopy was performed under general anesthesia. Fluorescence observation was conducted using the IMAGE1 STM OPAL1^® PDD camera system (KARL STORZ SE & Co. KG, Tuttlingen, Germany). The setup included a D-light C/AF light source equipped with a bandpass filter for transmitting blue light, and the red fluorescence was detected using a HOPKINS II PDD telescope (KARL STORZ SE & Co. KG) with a 30-degree angle and a 10 mm diameter.

Under WL, a nodule was considered positive if it could be visually identified as a tumor, whereas under fluorescent light (FL), a tumor was considered positive if it emitted red fluorescence. The abdominal cavity was first examined under WL to locate the sites of peritoneal dissemination, followed by an observation under FL to identify additional fluorescent areas. Surgeons could seamlessly switch between the WL and FL using a foot pedal during the procedure.

Biopsies were obtained in 4 distinct scenarios: 1) tumor sites identifiable under both WL and FL guidance (WL[+], FL[+]), 2) tumor sites visible only under WL guidance (WL[+], FL[−], 3) tumor sites visible only under FL guidance (WL[−], FL[+]), and 4) non-visible tumor locations under both WL and FL guidance (WL[−], FL[−]). All specimens were sent for pathological analysis to determine the presence or absence of malignancy and to assess the histology. Furthermore, we carefully evaluated the correlation between laparoscopic imaging data and pathological findings to assess the diagnostic performance of 5-ALA-PDD.

3. Safety

The administration of 5-ALA carries a potential risk of liver toxicity owing to the significant accumulation of its metabolite, PpIX [11]. Furthermore, the strong photosensitivity of PpIX can lead to cellular damage in the skin upon exposure to intense light, resulting in photosensitive reactions such as erythema, swelling, and pruritus [12]. Adverse events (AEs) that were considered 5-ALA-related, such as clinical symptoms, hepatotoxicity, and photosensitivity, were evaluated from the time of 5-ALA administration to discharge. Laboratory tests measuring aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were conducted during the screening process and 1 day after surgery to evaluate hepatotoxicity. 5-ALA-related AEs were graded using the Common Terminology Criteria for Adverse Events (version 5.0).

4. Diagnostic efficacy

The observation conditions and locations of the biopsy sites under WL and FL were recorded intraoperatively, and the surgical videos were stored on a hard disk. Each biopsy specimen was numbered, and the observation conditions and locations were documented. All biopsy specimens were reviewed and evaluated by an expert gynecological oncology pathologist. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), false-positive rate (FPR), and accuracy for each lesion under WL and FL guidance were calculated.

5. Statistical analysis

Continuous variables are expressed as medians with ranges. Sensitivity was defined as the proportion of pathologically confirmed metastatic lesions among all the lesions identified as metastatic under WL or FL guidance. Specificity was defined as the proportion of samples without pathological evidence of malignancy among all lesions identified as nontumorous under WL or FL guidance.

RESULTS

Seven patients scheduled for diagnostic laparoscopy due to a clinical diagnosis of advanced EOC between January and March 2019 were enrolled in this study. The patient characteristics are summarized in Table 1. The median age of the patients was 53 years (range, 38–70). Preoperative FIGO staging revealed stage IIIC in 6 patients and stage IVB in 1 patient. In one patient, EOC was initially clinically diagnosed, but the final histological analysis resulted in a diagnosis of pseudomyxoma peritonei. Among the 6 patients diagnosed with EOC, one had clear cell carcinoma (CCC), one had grade 3 endometrioid carcinoma, and four had high-grade serous carcinoma (HGSC).

Table 1. Patient characteristics.

Histology	FIGO stage	Age	Total number of biopsies
Endometrioid, grade 3	IVB	49	9
Clear cell	IIIC	38	14
Pseudomyxoma	-	66	9
High-grade serous	IIIC	58	6
	IIIC	65	11
	IIIC	70	12
	IIIC	42	10

FIGO, International Federation of Gynecology and Obstetrics.

A total of 71 biopsy samples were randomly collected from the peritoneum, omentum, and fallopian tube serosa. The intraoperative images of dissemination under WL and FL are shown in Fig. 1. Fig. 1A and B show the typical views of peritoneal and omental dissemination, respectively. In case 3, a nodular lesion was considered metastatic under WL; however, it did not exhibit fluorescence under FL and was pathologically diagnosed as pseudomyxoma peritonei (Fig. 1C). Fig. 1D shows a lesion that was not visible under WL but demonstrated fluorescence under FL, and was pathologically confirmed as a disseminated lesion. Video S1 shows the appearance of the lesion presented in Fig. 1D under the WL and FL conditions.

Fig. 1. Intraoperative images of peritoneal dissemination under WL and FL sources. Disseminated lesions appear as red fluorescence under the FL source. (A) Omental dissemination of clear cell carcinoma. (B) Peritoneal dissemination of high-grade serous carcinoma. (C) A nodule of pseudomyxoma peritonei. (D) No visible peritoneal dissemination under WL; presence of red 5-ALA fluorescence-spots under FL.

Table 2 presents a comparison of the appearance of the biopsy specimens under each light source with their histological evaluation for each case. Consequently, the sensitivity, specificity, PPV, NPV, FPR, and accuracy were calculated using 71 specimens from seven patients, including those from the benign tumors in case 3. The median number of specimens collected per case was 10 (range, 6–14). Three of 7 patients had ≥1 additional metastatic lesion detected via intraoperative 5-ALA imaging, and the FPR was 9.5% (2/21). The diagnostic performance of PDD with 5-ALA for each case under the WL and FL conditions is shown in Table S1. The sensitivity, specificity, PPV, NPV, and accuracy of all specimens from the EOC cases under WL were 87.8% (43/49), 59.0% (13/22), 82.7% (43/52), 68.4% (13/19), and 78.9% (56/71), respectively, while under FL, these values were 96.0% (48/50), 90.5% (19/21), 96.0% (48/50), 90.5% (19/21), and 94.3% (67/71), respectively (Table 3).

Table 2. Comparison of the biopsies under each light system with histological evaluations of the specimens for each case.

WL/FL status	Case 1		Case 2		Case 3		Case 4		Case 5		Case 6		Case 7
	Endometrioid, Grade 3		Clear cell		Pseudomyxoma		High-grade serous		High-grade serous		High-grade serous		High-grade serous
	Cancer	Non-cancer	Cancer	Non-cancer	Cancer	Non-cancer	Cancer	Non-cancer	Cancer	Non-cancer	Cancer	Non-cancer	Cancer	Non-cancer
WL(+) FL(+)	7	1	9	0	0	0	6	0	7	0	7	0	7	0
WL(+) FL(−)	0	0	0	0	0	8	0	0	0	0	0	0	0	0
WL(−) FL(+)	0	0	3	0	0	0	0	0	0	1	1	0	1	0
WL(−) FL(−)	0	1	1	1	0	1	0	0	0	3	0	4	1	1
Total	7	2	13	1	0	9	6	0	7	4	8	4	9	1

FL, fluorescence light; WL, white light.

Table 3. Diagnostic performance under each light system.

Category	Sensitivity		Specificity		PPV		NPV		Accuracy
	WL	FL	WL	FL	WL	FL	WL	FL	WL	FL
All cases	87.8%	96.0%	59.0%	90.5%	82.7%	96.0%	68.4%	90.5%	78.9%	94.3%
	43/49	48/50	13/22	19/21	43/52	48/50	13/19	19/21	56/71	67/71

FL, fluorescence light; NPV, negative predictive value; PPV, positive predictive value; WL, white light.

None of the patients experienced nausea or vomiting within 4 hours administration of 5-ALA. One patient reported mild (grade 1) nausea after surgery, which resolved on the first postoperative day, and no vomiting was observed. In addition, blood tests on the first postoperative day showed no elevation in AST or ALT levels and no photosensitivity. Furthermore, no serious adverse events (SAEs) occurred during hospitalization.

DISCUSSION

This study demonstrated that 5-ALA-based PDD achieved a sensitivity of 96.0% and a specificity of 90.5% for the visualization of EOC. Notably, microlesions that could not be identified under WL were successfully detected under FL in three out of 6 cases. Thus, PDD facilitates the detection of microlesions that are difficult to identify using conventional WL observations or imaging techniques. These findings suggest that 5-ALA-based PDD is valuable for the initial diagnosis of EOC. In particular, it is reliable in detecting the negative cases (based on the high NPV), making it useful for accurately assessing the extent of EOC spread during surgery.

Many previous studies have reported that 5-ALA-based PDD demonstrates high diagnostic accuracy for bladder and brain tumors [5,6]. Similarly, case reports and studies with small patient populations in EOC have suggested that the oral administration of 5-ALA for PDD achieves a diagnostic accuracy comparable to that observed in other cancer types [13,14,15]. These studies reported a sensitivity of 75%–95% and a specificity of 88%–100% for detecting peritoneal metastasis [13,14,15]. No cases of photosensitivity or SAEs were reported, indicating a high safety profile [13,14,15]. Previous studies on 5-ALA-based PDD for OC primarily focused primarily on observations under FL [13,14,15,16]. However, no study has systematically evaluated how lesions observed under WL appear under FL or vice versa. In the present study, we initially observed the target areas under WL and subsequently evaluated the same biopsy sites under FL. To the best of our knowledge, this is the first study to directly compare and evaluate identical regions under both lighting conditions. Furthermore, in addition to confirming the effectiveness of 5-ALA-based PDD as reported in previous studies, we demonstrated its applicability to HGSC and other histological subtypes, including CCC and grade 3 endometrioid carcinoma.

Case 3, diagnosed with pseudomyxoma peritonei, presented with a nodular lesion that appeared metastatic under WL but did not fluoresce under FL. This case underscores the advantage of fluorescence diagnosis, as benign lesions, such as pseudomyxoma peritonei, may not exhibit fluorescence, potentially yielding a true-negative result. Although FL is highly sensitive for detecting malignant EOC lesions, discrepancies between WL and FL findings, such as when a lesion is visible under WL but does not fluoresce under FL, highlight the need for cautious interpretation of the FL findings. Indeed, recent studies have highlighted the subjective nature of intraoperative assessment using 5-ALA-based PDD [17]. Although our present study did not formally assess interobserver variability, previous reports from glioblastoma and bladder cancer surgeries suggest that fluorescence interpretation may vary depending on the observer’s specialty and level of experience [17,18]. Greater concordance has been observed among experienced surgeons, with studies indicating that 20–30 cases are needed to achieve reliable interpretation [17,18]. These findings underscore the importance of surgical experience and structured training. In the context of EOC, where 5-ALA-based PDD remains an emerging technique, future studies should incorporate standardized training modules and objective image evaluation protocols to reduce interobserver bias and enhance clinical applicability.

Intraoperative video recordings can be reviewed postoperatively to validate whether the real-time decision-making was appropriate. These recordings may demonstrate that fluorescence signals often guided biopsy site selection and enabled detection of occult lesions not visible under white light, thereby supporting the added diagnostic value of 5-ALA-based PDD during surgery. Accumulating such video data may also contribute to refining future intraoperative decision-making by providing valuable feedback and educational material for surgical teams.

Regarding the safety of using 5-ALA for fluorescence imaging, no SAEs have been reported in the study. Patients did not experience nausea or vomiting within 4 hours of administration. Only 1 patient reported mild nausea that resolved on the first postoperative day. Additionally, there were no significant changes in the liver function tests (AST/ALT), and no photosensitivity was observed, indicating that FL is a safe technique for intraoperative use.

This study utilized 5-ALA as a fluorescent agent that accumulates in cancer cells and fluoresces under specific lighting conditions, aiding tumor visualization. In comparison, pafolacianine is an FDA-approved imaging agent that targets folate receptor alpha (FRα), and its clinical utility in detecting FRα-positive OC lesions during cytoreductive surgery has been demonstrated [10]. In a phase II trial, pafolacianine-guided imaging showed 85.9% sensitivity and 88.1% PPV [19]. A phase III trial showed that fluorescence imaging identified additional lesions in 33% [10]. However, its efficacy is largely restricted to tumors with high FRα expression, particularly HGSC, and its diagnostic performance in FRα-negative subtypes remains unclear [10]. Furthermore, van Dam et al. [20] reported that fluorescence intensity depends on FRα expression levels, with weaker fluorescence observed in tumors with low FRα expression, potentially limiting detectability. By contrast, 5-ALA functions as a metabolic imaging agent that accumulates as PpIX in malignant cells regardless of surface marker expression. This mechanism allows for broader histologic applicability. In the present study, 5-ALA was effective in visualizing HGSC, CCC, and grade 3 endometrioid carcinoma, highlighting its versatility across subtypes. Moreover, unlike intravenously administered pafolacianine, 5-ALA is taken orally, offering a less invasive option for intraoperative imaging. In a phase III trial for non-muscle-invasive bladder cancer, 5-ALA-related AEs, excluding laboratory abnormalities, were limited in scope and infrequent, with no SAEs or photosensitivity reported [21]. As a naturally occurring mitochondrial amino acid, 5-ALA has low toxicity, which may contribute to its favorable safety profile [22]. In contrast, a phase II trial of pafolacianine reported that pafolacianine-related AEs, particularly gastrointestinal symptoms and infusion reactions, occurred at a higher incidence compared to those associated with 5-ALA [19]. Although both agents appear to be generally well tolerated, the lower frequency of gastrointestinal AEs and the oral route of administration suggest that 5-ALA may represent a safer and more practical option for intraoperative imaging.

The results of this study demonstrated that 5-ALA-based PDD is highly effective for the diagnosis and surgical support of disseminated lesions in OC. In particular, its ability to detect residual microscopic tumors, enabling complete resection, suggests its potential in reducing the risk of postoperative recurrence. Improvements in complete resection rates, contributing to better prognoses, have been corroborated in other cancer types. For example, randomized controlled trials have reported reduced recurrence rates in non-muscle-invasive bladder cancer and improved 6-month progression-free survival in malignant gliomas [6,23].

Real-time fluorescence-guided surgery facilitated by 5-ALA offers significant advantages in OC surgery. Complete gross resection is a key prognostic factor for advanced OC requiring primary debulking surgery. Real-time fluorescence imaging improves surgical precision by allowing the detection of metastatic lesions that may be missed by conventional observation. It also aids in swift transition to adjuvant chemotherapy in cases where complete resection is not achievable, thereby reducing unnecessary surgical invasiveness. Similarly, interval debulking surgery after neoadjuvant chemotherapy (NACT) benefits from this technology because NACT can shrink tumors to undetectable levels through macroscopic observation and palpation, thereby increasing the risk of microscopic residual disease. Real-time fluorescence imaging provides critical assistance in identifying such lesions. Furthermore, staging surgeries for suspected early-stage OC involve systematic lymph node dissection and peritoneal biopsy. Real-time fluorescence imaging can detect metastases that are invisible under WL conditions, potentially identifying lesions that may otherwise be missed. Additionally, the omission of procedures on non-fluorescent tissues could minimize damage to normal tissues, thereby reducing surgical invasiveness and enhancing patients’ quality of life (QoL). These findings underscore the potential of 5-ALA-based PDD not only in enhancing the surgical outcomes but also in contributing to improved patient prognosis and QoL.

This study has several limitations. First, it was conducted at a single high-volume center by a specialized surgical team in the initial phase of using 5-ALA-guided procedures for advanced EOC. Therefore, the findings may not be directly generalizable to other institutions where surgical experience and resource availability may differ. Second, the sample size was very limited (n=7), which prevents evaluation of diagnostic performance based on tumor histology, disease distribution, or anatomical localization. These limitations underscore the exploratory nature of our findings. To validate and extend these results, further large-scale, multicenter studies are warranted in order to assess diagnostic accuracy across diverse clinical settings. Notably, a phase III clinical trial (NCT05804370) is currently ongoing to investigate the safety, diagnostic performance, and clinical usefulness of 5-ALA for the real-time detection and visualization of peritoneal dissemination during debulking surgery, in patients with suspected newly diagnosed or recurrent disease.

The findings from our present study support the potential clinical utility of 5-ALA-based PDD for detecting peritoneal dissemination in patients with EOC. The results of the ongoing phase III trial may help establish PDD as a standard intraoperative diagnostic tool in the surgical management of OC.

SUPPLEMENTARY MATERIALS

Video S1

Enlarged view of the disseminated lesions in Fig. 1D under FL.

Table S1

Diagnostic performance of each case under each light system