Safety and efficacy of photobiomodulation therapy in oncology: A systematic review

Abstract We performed a systematic review of the current literature addressing the safety and efficacy of photobiomodulation therapy (PBMT) in cancer patients. In this systematic review, the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines were used. In vitro, in vivo, and clinical studies, which investigated the effect of PBMT on cell proliferation/differentiation, tumor growth, recurrence rate, and/or overall survival were included. The Medline/PubMed, EMBASE, and Scopus databases were searched through April 2020. A total of 67 studies met the inclusion criteria with 43 in vitro, 15 in vivo, and 9 clinical studies identified. In vitro studies investigating the effect of PBMT on a diverse range of cancer cell lines demonstrated conflicting results. This could be due to the differences in used parameters and the frequency of PBM applications. In vivo studies and clinical trials with a follow‐up period demonstrated that PBMT is safe with regards to tumor growth and patient advantage in the prevention and treatment of specific cancer therapy‐related complications. Current human studies, supported by most animal studies, show safety with PBMT using currently recommended clinical parameters, including in Head & Neck cancer (HNC) in the area of PBMT exposure. A significant and growing literature indicates that PBMT is safe and effective, and may even offer a benefit in patient overall survival. Nevertheless, continuing research is indicated to improve understanding and provide further elucidation of remaining questions regarding PBM use in oncology.


| INTRODUCTION
In 1967, Dr. Endre Mester was the first scientist to discover that a low power laser had a stimulating effect on hair regrowth in mice. 1 Since then, low-level laser (LLL) has been applied for a variety of conditions and to boost physiological function in both humans and animals. In the past decade or so, the term photobiomodulation (PBM) replaced the former low-level laser (LLL), and PBM was introduced as MESH word in PUBMED in 2015. Subsequently, the North American Association for Light Therapy (NAALT) 2 and the World Association for Laser Therapy (WALT) defined photobiomodulation therapy (PBMT) as a form of light therapy that utilizes non-ionizing forms of light sources, including laser diodes (LD), light-emitting diodes (LEDs), and broadband light, in the visible and infrared spectrum. PBM provokes a nonthermal process whereby endogenous chromophores elicit photophysical and photochemical events at diverse biological levels. This process results in positive therapeutic outcomes including the stimulation of tissue regeneration and wound healing, the reduction of inflammation and pain, and immunomodulation. 3 Since 1967, the number of clinical applications of light therapy has increased steadily in multiple medical fields, and in recent years PBM has been widely used for supportive care of cancer patients. 4,5 The best-studied cancer therapy-related complication, for which PBM is recommended, is oral mucositis (OM). The Mucositis Study Group of the Multinational Association of Supportive Care in Cancer/International Society for Oral Oncology (MASCC/ISOO) recommends the use of PBMT in the prevention of OM in patients with head and neck cancer undergoing chemoradiotherapy (CRT) and in stem cell transplant patients treated with high-dose cytoreductive medications. 6 PBM also has beneficial effects in the management of soft tissue necrosis in patients with head and neck cancer (HNC), and therapy-induced bone necrosis. [7][8][9] Also the potential application of PBM for management of xerostomia, dysgeusia, radiodermatitis, post-RT fibrosis, chronic oral graft-versus-host disease (GVHD), and breast cancer-related lymphedema has been reported. [10][11][12][13][14][15][16][17] The basic principle of supportive care in cancer is to provide effective management of complications of cancer treatment without compromising or inducing negative effects on oncology outcomes; also bearing in mind unwanted and dire consequences such as tumor persistence, new secondary tumors, or recurrence of the primary disease. Various in vitro studies have suggested that PBM may induce accelerated growth in some malignant cell lines and/or development of malignancy in dysplastic cells. Due to its increasing utilization in oncology care, and the better understanding of the biologic mechanisms and clinical outcomes, it is important to document the safety of PBM use in oncology settings. The aim of the present systematic review is to evaluate the available literature describing the safety and intervention outcomes in cancer patients receiving PBMT.

| Protocol
For this systematic review, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was used. 18

| Eligibility criteria
We considered for inclusion all in vivo and human clinical trials articles dealing with treatment and/or prevention of cancer therapy-related complications, as well as in vitro studies on the safety of PBMT on cell lines. Case reports, cohort studies, case-control studies, systematic and literature reviews, letters to the editor, theses, studies published in a language other than English, monographs, commentaries, conference abstracts, and unpublished data were excluded.

| Information sources
We queried three databases (Medline/PubMed, EMBASE, and Scopus). To detect other potentially eligible reports that could meet the inclusion criteria, the reference list from all selected studies were checked by the reviewers. In addition, the included studies were screened to identify key authors. This allowed us for extra database searches based on author name. The last search was performed in April 2020.

| Study selection
All papers were systematically ordered in a Microsoft Office Excel 2016 document (Microsoft Corporation). The titles were checked and the duplicates excluded. Afterward, titles and abstracts were read for inclusion in the systematic review. Studies were classified into different categories: in vitro studies, in vivo studies, clinical studies, duplicates, no follow-up/ safety information, and language other than English. Two independent reviewers (RJB, JR) reviewed the studies assessed for eligibility in full-text version. The studies lacking relevant methodological information were excluded.

| Data extraction
Data from the included studies were extracted according to the following: (a) Author and publication year; (b) Study type (clinical trials and in vivo/in vitro);(c) PBM properties and treatment protocol; (d) type of animal models; (e) types of cells; (f) patient population; (g) duration of follow-up; (h) outcome measures.

| Data analysis
For this systematic review, a meta-analysis was not feasible due to the great variation in PBM protocols used in the included studies. This systematic review presents a comprehensive qualitative synthesis of the results from the incorporated studies.

| Study selection
The flow diagram ( Figure 1) gives on overview of the selection process of the included studies. In total, 870 studies were collected via the searches on the databases and 5 additional studies via the manual search. After a first review process, 585 duplicate studies were removed. One hundred and nine studies were excluded after reviewing the abstracts, and 114 studies were excluded as they did not meet the inclusion criteria. In total, 67 studies were included in the present systematic review: 43 in vitro, 15 in vivo, and 9 human clinical studies were analyzed.

| Study characteristics
The study characteristics of the in vitro, in vivo, and human clinical studies are presented in Tables 1, 2, and 3, respectively.

| In vitro studies
A total of 43 in vitro studies examining the effect of PBMT on diverse cancer cells types with varying PBM parameters were included (Table 1).  The potential for PBM to negatively influence tumor growth and/or reaction to cytotoxic treatment has had a limited evaluation to date and is currently unresolved. Conflicting data refute or support the potential for PBM to impact tumor activity and responsiveness to treatment. As noted above, given the lack of uniformity, which characterizes tumor biology, it seems probable that different tumors vary in reaction to the range of biomodulatory activities associated with PBM exposure. PBM may affect various pathways linked to negative tumor behavior, including cell proliferation and anti-apoptosis effects. Different malignant cell lines have been used in in vitro studies to investigate the effects of PBM on cell proliferation and differentiation. They showed conflicting data by exploiting a wide diversity of PBM parameters and tumor cell lines. 48,49,52,54,61,65 Concerning the effect of PBM on malignant transformation in non-cancer cells, an in vitro study applied PBM (660 nm, 350 mW, 15 min) during three consecutive days to epithelial cells and/or fibroblasts and no change in cell behavior was shown. 45 Besides, in an in vitro study with normal breast epithelial cells, no malignant transformation was detected when different PBM doses and wavelengths were applied during numerous exposures. 58 When PBM parameters were used outside the range recommended in oncology (GaAIAs laser, 809 nm, 1.96-7.84 J/cm 2 ), a clear proliferation was seen in laryngeal carcinoma cells. 52 Another study applying PBM at different wavelengths (685 and 830 nm) to Hep2 carcinoma cells, clearly showed an increase in proliferation. 55 The differential effect of PBM on normal and cancer cells was tested in a study with osteoblasts and osteosarcoma cells. The study demonstrated that only a laser diode (830 nm) at 10 J/cm 2 was able to increase osteoblast proliferation. On the contrary, a laser diode (780 nm) decreased osteoblast proliferation at energy densities of 1, 5, and 10 J/cm 2 . PBM did not affect osteosarcoma cells by using an 830 nm laser, while a minor proliferative effect was detected at 670 nm. 57 In another study, human breast carcinoma and melanoma cell lines were used to investigate the effects of diverse doses of PBM at different wavelengths on cancer cells 58 : the proliferation of breast carcinoma cells increased at specific PBM doses, while numerous exposures had either no effect or reduced cell proliferation. An in vitro study on oral cancer cell lines with 1 J/cm 2 PBM (660 nm) showed a nonsignificant increase in the invasive potential of these cell lines. 40 Another in vitro study in oral dysplastic and oral cancer cells suggested that PBM (660 nm or 780 nm, 40 mW, 2.05, 3.07, or 6.15 J/cm 2 ) could modulate the Akt/mTOR/ CyclinD1 signaling pathway linked to more aggressive cell behavior. 42 Another report of PBM exposure of three head and neck cancer (HNC) cell lines was reported to increase the proliferation of cells in each tumor line, but not in normal tissue control. 29    While the limits of basing broad-reaching conclusions on in vitro assays have been noted, collectively the reports suggest it would be irresponsible to ignore the possibility that PBM could negatively impact tumor behavior. Therefore, understanding how PBM may modify tumor biology, both positively and negatively, is a research priority. 66 Direct investigation of the radio-modulatory effects of PBM as it affects tumor response is limited, but as with other types of cytotoxic cancer therapy, PBM may affect tumor response to radiation by the dose, fractions, and timing of PBM or radiotherapy (RT). While the data are sparse and limited to in vitro studies, the evidence suggests that PBM may act as a radiosensitizer. 30 Another in vitro study with three HNC cell lines suggests that PBMT can enhance the sensitivity of cancer cells to chemotherapy (CT). 27 Conversely, PBM's reported induction of cell survival suggests a potential pathway for tumor self-preservation. 67 An in vitro study with oral squamous cell carcinoma (OSCC) demonstrated that PBM induced apoptosis in the absence of radiation. Moreover, PBM did not induce anti-apoptotic effects that might stimulate tumor cell resistance to cancer therapy. 45 When PBM (810 nm, continuous wave, 20 mW/ cm 2 , 1.5 J/cm 2 ) was applied to human osteosarcoma cells before NPe6-mediated photodynamic therapy, increased apoptosis was detected as a result of a higher uptake of the photosensitizer and an increased cellular ATP. 68 The potential enhancement of ionizing RT and CT in OSCC, was seen in PBM when applied shortly before RT and suggested that increased loco-regional blood flow may have contributed to local tissue oxygenation, which may translate into enhanced tumor effect. 69 Several in vitro studies demonstrated that PBM could also inhibit the proliferation of malignant cells. A study using PBM (805 nm, 4 J/cm 2 or 20 J/cm 2 ) in gingival SCC demonstrated a decrease in mitotic rate. 49 In an in vitro study with osteosarcoma cells, PBM (830 nm) did not influence cell proliferation or protein expression. 59 A decreased proliferation of human hepatoma cells was detected after PBM (808 nm; 5.85 and 7.8 J/cm 2 ). 60 A study with glioblastoma and astrocytoma cells showed a minor reduction in mitotic rate after PBM (805 nm and 5-20 J/cm 2 ). 51 Comparably, glioblastoma cell proliferation was inhibited by PBM (808 nm, 5 J/cm 2 ). 43 PBM at rather high cumulative doses resulted in growth inhibition of various malignant cell lines. 46 This suggested the hypothesis that PBM may have favorable effects in the treatment of lung cancer. 70 An in vitro study in B16F10 melanoma cells showed that high-dose PBM (50 J/cm 2 ) seemed safe, with only insignificant effects on proliferation. Furthermore, no noteworthy effect on tumor growth in a melanoma mouse model was shown. PBM at a high power density (2.5 W/cm 2 ) with a very high dose of 1050 J/cm 2 induced tumor growth in the melanoma mouse model. 62 The wide variety of PBM parameters utilized in these studies constitutes an obstacle to arriving at meaningful conclusions, especially when the parameters are outside the scope of the MASCC/ISOO guideline that recommended PBM therapy in cancer care. Additionally, even studies using similar parameters can have differing or contradictory results. To wit, some in vitro studies suggest that PBM may favor tumor progression of oral SCC cells by activation of Akt/mTOR pathway, 42 cellular proliferation, 29,40 and cellular migration, 28 while other studies report a reduction in tumor growth. 25,28,63 It is also important to comment that the results suggesting PBM tumor enhancement were not replicated in other studies. Generally, in vitro studies have limited applicability when compared with in vivo studies where various physiologically active cells and systems interact in the targeted tissue. There is a concomitant effect of light on endothelial, epithelial, mesenchymal, and immune cells, which must be studied together to identify real-time effects.

| In vivo studies
A total of 15 in vivo studies investigating the safety of PBMT in different animal cancer models were identified (Table 2). 62,63,[71][72][73][74][75][76][77][78][79][80][81][82][83] In a study with a chemically induced OSCC hamster cheek pouch model, PBM (660 nm, 30 mW, 424 mW/cm 2 , 56.4 J/cm 2 , and 133 s, 4 J) led to tumor progression. 76 In contrast, a study with a mouse model of multiple UV-induced skin tumors, full-body PBM (670 nm, twice a day, 5 J/cm 2 for 37 days) did not enhance tumor growth in comparison with sham-treated animals. Moreover, the tumor area slightly decreased after PBM, possibly associated with PBM-induced antitumor immune activity or a local photodynamic effect. 77 Similar results were seen in a rat study demonstrating that PBM was able to reduce and let even completely disappear small tumors. 82 This led to the hypothesis that the upregulation of ATP signaling by PBM stimulated differentiation of tumor cells and apoptosis, leading to an inhibition of tumor proliferation. 84,85 A normal cell produces ATP via the process of oxidative phosphorylation. This gives a yield of around 32-38 ATPs per glucose molecule. Cancer cells naturally change from "cellular respiration" to the very ineffective glycolysis for their ATP needs (i.e. Warburg effect). Cancer cells perform anaerobic glycolysis, which implies that they produce most of their energy from glycolysis. This produces only two ATPs per glucose molecule. 86 The potential of PBM to promote anti-inflammatory and repair of normal tissue while not enhancing tumor cell proliferation may be related to this differential effect.
PBM was tested to stimulate hair regrowth in an animal model with leukemia, which developed chemotherapy-induced alopecia (CIA). PBM did not alter the efficacy of CT, as 22% of the PBM-treated rats and 20% of the control rats remained leukemia-free. 83 An in vivo study showed that PBM reduced the tumor growth and invasiveness in xenograft OSCC and melanoma mouse models. The authors suggested that PBM may have impacted tumor proliferation by stimulating antitumor immunity and normalizing tumor vessels. 63 A recent study in an orthotopic animal model of head and neck squamous cell carcinoma (HNSCC) suggests that the use of PBMT does not safeguard tumor cells against the cytotoxic effect of RT. 72 The described results indicate that diverse malignant cells may react differently to specific PBM parameters and doses. A possible explanation for this lies in the dissimilarities in the cellular microenvironment between different tumor models, as PBM has a clear effect on the cellular signal transduction pathways. The microenvironment of cancer cells differs between in vitro studies. Moreover, it is not possible to compare the microenvironment of cell culture studies with that in animal models. To improve the understanding of the dissimilarities in tumor response to PBM and how pretreatment molecular and genomic characterization of tumors can be used to establish the most appropriate PBM parameters, more in vitro and in vivo studies are needed. 87

| Clinical human studies
We identified nine clinical trials studying the safety of PBMT in patients with cancer, reporting disease-free survival, overall survival, and recurrence rates (Table 3). [88][89][90][91][92][93][94][95][96] In a prospective, randomized-controlled trial (RCT) with HNC patients (SCC of the nasopharynx, oropharynx, and hypopharynx) undergoing CRT, PBM was used to prevent OM. The average follow-up time was 18 months (range 10-48 months). Results showed that PBM treated patients had improved loco-regional disease control, and a better progression-free and overall survival. 93 In a retrospective study with 152 advanced OSCC patients PBM (660 nm, 40 mW, 10 J/cm 2 ) applied for the prevention of OM. In overall, PBM did not affect the treatment result of the primary tumor, relapse rate, development of new primary tumors, and overall survival. 90 Similarly, a retrospective study of 222 patients with HNC who were treated with RT with or without cisplatin-based CT investigated the safety of PBM in the management of OM. PBM did not affect the time to local recurrences, the disease-free survival, and the overall survival. 89

| DISCUSSION
Evidence from the clinical follow-up, and animal models and in vitro data indicate that the possible negative effect of PBM on tumor biology is not clinically relevant at the doses applied in the management of cancer therapy-related complications. It is key to understand the effect of different PBM parameters (wavelength, fluence, energy, and time) and experimental models before applying and evaluating PBM correctly. As there is a clear contrast between in vivo and human studies versus in vitro cell culture studies, related to the tumor microenvironment. 66,97,98 In vitro studies strengthen the meaning of dosimetry and exact description and control of PBM parameters for clinical, oncologic use. The benefits of the use of PBM and levels of theoretical risk should be considered for the best patient care. In our opinion, current evidence supports the use of PBM in accepted indications. The Mucositis Study Group of the MASCC/ISOO has completed a systematic review and recommended the use of PBM for management of OM in stem cell transplant and HNC patients receiving stomatotoxic cancer therapies. 6 NICE guidelines in the UK recommend PBM also in those indications. 99 For more than 20 years, PBM has been used in the management of OM in HNC patient and no significant adverse effects have been documented. Clinical studies have assessed tumor outcomes in patients treated with PBM. [88][89][90][91][92][93] While these clinical trial outcomes support safety in the clinical application of PBM for oral/oropharyngeal complications of cancer therapy, continuing research is warranted due to the potential diverse biological impact of PBM, and due to tumor heterogeneity and evaluating tumor behavior and overall response to therapy. Indeed tumor heterogeneity and the tumor microenvironment may have been reflected in contradictions observed in some in vitro studies. For example, 35% expressing dysregulation in the PI3K pathway, a putative site of action of PBM in OSCC. 100 Animal and clinical trials allow assessment of the effect of PBM on the tumor and the tumor microenvironment (e.g. immune function, the surface microenvironment, and epithelial and connective tissue interactions). While additional clinical trials and follow-up are desirable, the current evidence supports the safety of PBM in the established protocols through consensus guidelines in the management of cancer therapy-related side effects.

| Limitations
Due to heterogeneity of the used PBM parameters and outcome measures, a meta-analysis of the reported data was not possible. Moreover, in this systematic review, in vitro, in vivo, and clinical trials were considered and therefore it was impractical to find a rigorous model to determine the risk of bias. This would be implemented in case the review only included clinical trials.

| CONCLUSION
PBM (in the red or NIR spectrum by definition) appears safe and successful in the management of cancer therapy-related side effects. Therefore, PBM should be considered as part of the standard of care for specific oncology purposes. 10 Clinicians should be able to prescribe PBM by guideline recommendations using appropriate approved parameters of PBM in a clinical setting. 101 The safety of PBM concerning the effect on the tumor response and most importantly the benefit of PBM to patients in the management of cancer treatment-related toxicities have been shown in in vivo studies and clinical trials. Vigilance remains warranted for applications that have not been adequately documented or without guidelines due to a lack of evidence. Therefore, studies concerning the effect of PBM on malignant cell protection or enhancement of tumor growth are still needed.
The increasing number of published data in OM prevention in HNC and stem cell transplantation patients suggest that PBM does not influence tumor or treatment outcomes and overall survival. In recognition of the complexities, which govern tumor responsiveness, 102 it remains obligatory upon the clinician to notify patients of the potential benefits and risks related to PBM. Based on the demonstrated value of PBM in supportive cancer care, continuing research may also be directed to elucidate its effect on respondents and nonrespondents to PBM, like any other treatment or mitigation modality in modern precision oncology.

DISCLAIMER
This systematic review has been performed by an international multidisciplinary panel of clinicians and researchers, with expertise in the area of PBM. This article is informational in nature and should be used with the clear understanding that continued research and practice could result in new insights and recommendations. In no event shall the authors be liable for any decision made or action taken in reliance on the protocols included in this review paper.