The global sunscreen market, particularly in North America, is dominated by sunscreens that mostly prevent to varying degrees UVB effects, such as sunburn, but offer little or no protection against UVA1 (340-400 nm), the most damaging and deeper penetrating of the UV rays. These types of products are known as UVB-biased sunscreens.

In addition, these types of sunscreens may not be effective in preventing the cellular cascade that is involved in the development of skin cancer and photoaging. Contemporary science suggests that UVA (mostly UVA1) plays a primary role in those processes.¹

The incomplete protection afforded by UVB-biased sunscreens could, therefore, contribute to rising annual skin cancer rates. Of note, skin cancer rates for Non-Melanoma Skin Cancer (NMSC) continue to rise in the USA and worldwide at an average annual rate of 1-2%.^2,3 The National Cancer Institute reports that melanoma rates in the United States tripled between 1975 and 2014.⁴

Skin cancer prevention programs

The Centers for Disease Control and Prevention(CDC) has identified 25 broad-based skin cancer prevention programs including nine Federal, National, or Medical Organizations that publicize skin cancer prevention measures and have extensive consumer education programs that address sun safety and UV exposure, sun protection, or skin cancer prevention.⁵ The exact language may differ, but they all focus on the following key elements:

• Sun avoidance (particularly between 10 AM – 2 PM)
• Never get burned or tan from UV exposure
• Wear protective clothing with a tight weave and UV protective sunglasses when outdoors
• Generously apply sunscreen with sun protection factor (SPF) 15 or higher having both UVA and UVB protection

Sunscreen use is the weakest link in the strategy for reasons including lack of efficacy and poor compliance by consumers. Rising cancer rates suggest the current approach has failed and critical analysis argues that poorly effective UVB-biased sunscreens may be a primary factor.

There is an evidence-based argument that the daily use of balanced sunscreens that provide spectral homeostasis, defined as virtually uniform protection at every wavelength with significant UVA attenuation, could reduce all forms of skin cancer and photoaging.^6,7 The Environmental Working Group (EWG) in 2017 affirms the principle of balanced UVB/UVA protection described by Osterwalder and Diffey as a way to prevent skin cancer.^6,7

Grim statistics

Skin cancer is now the most common cancer in the United States. In North America, it now accounts for more than 50% of all human cancers (i.e., skin cancer cases outnumber all other cancers combined).^2,4 The incidence of BCC (Basal Cell Carcinoma) and SCC (Squamous Cell Carcinoma) increased by 145% and 263%, respectively, from 1976-1984 and 2000-2010. The rate of new melanoma cases among American adults has tripled from 7.9 per 100,000 people in 1975 to 25.2 per 100,000 in 2014.⁴ Melanoma is the leading cause of cancer death in women ages 25-30, the second leading cause of cancer death in women ages 30-35, and the second most commonly diagnosed cancer in ages 15-29.⁸

From 1970 to 2009, the incidence of melanoma increased by 8-fold among young women and 4-fold among young men. In the USA, one person dies of melanoma every 54 minutes. An estimated 9,730 people will die of melanoma in 2017.⁴ Although less common, about 4,000 Americans die in a year from SCC.⁹

In addition to the human cost, the economic outlay is enormous and could double by 2030. Each year in the U.S., over 5.4 million cases of nonmelanoma skin cancer are treated in more than 3.3 million people.² The annual cost of treating skin cancers in the U.S. is estimated at $8.1 billion—about $4.8 billion for NMSC and $3.3 billion for melanoma.¹⁰

A primary role for UVA in photocarcinogenesis

UV-induced photocarcinogenesis involves a complex interplay between ultraviolet radiation (UVR), skin cells, molecular pathways, DNA, and the immune system.¹¹ UVR is both the initiating insult and the ongoing promoter driving the cancer cascade in NMSC, which is relatively unique in carcinogenesis.

UVA radiation generates more oxidative stress than UVB, causing 10 times more lipid peroxidation. UVA is more cytotoxic than UVB; it damages DNA by causing strand breaks and oxidation of nucleic acids. Further, UVA can inhibit DNA repair and induce matrix metalloproteinase (MMP) synthesis, which augments the biologic aggressiveness of skin cancer.¹

More recent literature confirms that both UVB and UVA cause direct DNA mutation of the repair genes p53 and BRAF as well as the KIT oncogenes to different degrees in each type of skin cancer.¹¹ UVA is the main driver of immunosuppression¹² and the generation of reactive oxygen species (ROS) or free radicals.¹³ UVB initiates and modulates the damage cycle, but UVA (particularly UVA-I) completes the process.

Mutation of these repair genes prevents apoptosis (programmed cell death) of atypical keratinocytes or melanocytes and disruption of other repair mechanisms that lead to cancer.¹¹ Acting as a tumor suppressor, p53 plays a central role in cell repair, apoptosis, and cell-cycle arrest. Once the cell cycle has been arrested, DNA damage can be repaired. Base excision repair (BER) involves the recognition and removal of damaged bases, which may not cause significant distortion of the DNA helix. BER is important in protection from oxidative damage of DNA, which is mainly attributed to UVA.¹¹

If cells have sustained too much damage to be repaired, apoptosis is a mechanism in place to destroy mutated cell lines. Those cells with one p53 mutation may still produce p53 protein and undergo apoptosis, so the mutation is not replicated. Cells with two p53 mutations will lack the capacity for apoptosis and continue to proliferate.¹¹ Atypical cell clones that are especially aggressive demonstrate uncontrolled cellular proliferation and continue along the carcinogenic pathway, expanding at the expense of normal skin cells.

Gene mutation is the primary pathway in skin cancer but in photoaging, the disruption involves enzymes like Cathepsin K, G, and Matrix Metalloproteinases (MMP). Any cellular pathway to skin cancer and photoaging requires a UVA insult, and a sunscreen without adequate UVA protection (called “extinction” in the field of dermatology) cannot prevent either to any significant degree.

A joint USA/Australian study confirmed the dominant role of UVA in photocarcinogenesis; there was more fingerprint UVA than UVB mutations in the keratinocytes of the basal layer in human epidermis.¹⁴ Another signature finding was that UVA impaired the p53 controlled cell cycle arrest essential for DNA repair much more than UVB—making UVA-induced pyrimidine dimers more mutagenic than UVB-induced ones.¹⁵

Indoor tanning

The literature in the past decade is replete with papers confirming the primary role that UVA plays in the genesis of skin cancer and photoaging. A detailed list is beyond the scope of this review, but other indirect evidence comes from the role that tanning beds emitting mostly UVA radiation play in an increased risk for all forms of skin cancer.

More than 419,000 cases of skin cancer in the U.S. each year are linked to indoor tanning, including about 245,000 basal cell carcinomas, 168,000 squamous cell carcinomas, and 6,200 melanomas. More people develop skin cancer because of tanning than develop lung cancer because of smoking.¹⁶

A lifetime use of tanning beds 10 or more times increases the risk of developing melanoma by 34%, and first use of a tanning bed before age 35 increases the risk for melanoma by 75%.¹⁷

Standards for tanning beds vary around the globe, and compliance and enforcement are poor, but the radiation profile is usually 0.5-5% UVB and 95% or more of UVA, delivered in output intensity 2-8 times that of sunlight. This is the acute example of asymmetric UVA exposure. The clinical example of insidious chronic asymmetric UVA exposure comes from using a UVB-biased sunscreen for daily protection.

Spectral profiles of a UVB-biased sunscreen

When UVB-biased sunscreens with low UVA-Protection Factor (UVA-PF) values or those with partial broad spectrum protection are used, they provide incomplete protection and the UV insult to the skin is likely to result in higher cumulative exposures than commonly employed sunbed practices. Unless a sunscreen provides up-to-date broad-spectrum protection with high UVA-PF values, a 2-week sunbathing vacation that avoids sunburn, using a UVB-biased sunscreen, can result in double the UV exposure with a greater health risk than a 10-session sunbed course.¹⁸

This landmark analytic photometric study proves that the wrong sunscreen—UVB-BIASED—may be more dangerous than tanning bed exposure.

Just three days in the sun using a UVB-biased sunscreen would be equivalent to two trips to the tanning salon. The similarity of tanning-bed UV exposure to using a UVB-biased sunscreen—either on vacation or every day—eludes most regulators, physicians (including dermatologists), and the cosmetic industry. Both exposures provide asymmetric UVA radiation to your skin.

There is another cautionary aspect to this. The 2-week vacation model is extreme but not unrealistic.¹⁸ It is more acute and intense, but consider the cumulative everyday exposure from using a UVB-biased sunscreen over many years. Many outdoor occupations reach or exceed the vacation exposure.

Critical wavelength

The FDA and Health Canada validate UVA claims with an indirect measure—the Critical Wavelength (CW). CW is used in Europe only as a secondary measure after the UVA-PF (UVA-Protection factor) is determined. They stipulate that a CW of 370 or more is needed to make the label claim of UVA and broad spectrum protection. If the stringent EU standard of actually measuring UVA-PF and ensuring that the UV index (UVA/PF/SPF) exceeds 0.33, the majority of USA sunscreens would get a FAIL.

Critical wavelength >370 nm is not a sufficient, reliable criterion to ensure that a product can provide effective protection against UVA damage, as two sunscreens with the same CW may have very different UVA-PF values.¹⁹ Professor Brian Diffey, the inventor of the CW test, cautioned the FDA that it should not be used alone as a primary determinant of adequate UVA protection.²⁰

The concept of a flat spectral profile or spectral homeostasis from 290-400 nm, as defined by Diffey in 1991, is very rational, since the entire UV band is considered as a carcinogen.⁷ This fundamental concept is now given more credence as there is persuasive evidence that UVA may be the main driver of UVR mediated sun damage to the skin.

Spectral homeostasis: Reducing the intensity of the entire UV band

This argument for spectral homeostasis is more compelling based on the advances made in identifying the action spectra for erythema, melanogenesis, skin cancer, ROS, immunosuppression, and photoaging to different specific bands within the solar UV spectrum. It is logical to reduce the intensity (quantity) of the entire UV band for the best clinical benefits, the way shade or protective clothing works.

For 40 years, the industry and their dermatology consultants have adopted the counterintuitive and irrational approach to use sunscreens that favor UVB attenuation and selectively modify the UV spectrum—mostly quality rather than quantity. The uniform reduction in both UVB and UVA or “spectral homeostasis” assures that the natural spectrum of sunlight is attenuated without altering its quality similar to the protection afforded by a neutral density filter like densely woven textiles or indoor shade.^6,7

Chemical filters that absorb the UV rays

Typical sunscreens rely on chemical UV filters that preferentially absorb a specific por­tion of UV range. Contemporary sunscreens still tend to filter considerably more UVB than UVA radiation and show an extreme UVB bias.⁶ It is now possible to shift the balance in UVR extinction towards UVA and develop actual UVA-biased sunscreens.

Figure 1. Sunscreen evolution from UVB via UVB-biased towards the ideal protection profile or spectral homeostasis.²¹. Reproduced here with permission of lead author.

Figure 1 shows the three categories of sunscreen. The older type sunscreens (top left) from the 1980-1995 era primarily provided UVB protection. They were slowly replaced by products with partial broad-spectrum protection (top right) with the advent of the UVA filter avobenzone and the increasing use of another UVA filter—micronized zinc oxide that produced less whitening on the skin. These two categories of sunscreen still dominate the global market, particularly in North America.

The ideal sunscreen providing optimal truly broad-spectrum filtering (bottom center) has the potential to reduce skin cancer and all forms of sun damage. This type of protection is now very attainable even although the most efficient UVA filters are still awaiting FDA approval. The higher UVA extinction required to do this is only achieved with efficient UVA filters like biscotrizole, ecamsule, or high concentrations of zinc oxide.

UV extinction vs. transmission

There is a tendency to focus on extinction as the amount of UV radiation prevented from passing through a sunscreen film. A more useful principle is to quantify the transmission or amount of harmful UV radiation that actually reaches the skin.

Clothing has a UV Protection Factor of 10-100 or more with uniform attenuation of the entire UVB-UVA spectrum, if made of dense material with medium or deep-color dyes or optical brighteners.⁶

Their extinction profile provides the best model for the ideal and maximally effective sunscreen.^6,7

Many brand name sunscreens using avobenzone and inadequate amounts of zinc oxide that attain the FDA and Health Canada standard of a CW ≥ 370 nm still show considerable UVB bias and allow transmission of undesirable levels of UVA (mostly UVA1 or the most damaging UV rays).

This is best described as a monochromatic protection factor (MPF), defined as the reciprocal value of the transmittance (i.e., 1/T, at a particular wavelength). For an SPF 30 sunscreen, the average MPF over the whole UVA I ranges varies from 1 (no protection) to 30 (equivalent to the SPF).

Many sunscreens with 3% avobenzone that meet the CW of 370 nm only provide an average protection factor of around 3-5 in the UVA1 range for an SPF 30 sunscreen.²² This means that 1/3 or more (>33%) of the UVA1 radiation is still transmitted onto the skin, whereas <4% (3.7%) of the UVB radiation (i.e., 1/30) is transmitted through this kind of SPF 30 sunscreen.²² This is the hallmark of the typical brand name UVB-biased sunscreen dominating the market—an extreme UVB protection bias where up to 10 times more UVA than UVB reaches the skin. Asymmetric UVA exposure similar to a tanning bed—less acute and less intense but likely posing the same risks.

Figure 2 (below) shows that UVB-biased sunscreen transmits much more UVA to the skin than a sunscreen with balanced UV protection.

Figure 2. Typical UVB-biased sunscreen transmitting up to 10 times more UVA than the balanced sunscreen with zinc oxide and two other broad-spectrum UVA filters. ²² From BASF sunscreen simulator, BASF SE, Ludwigshafen. 2010. Available at httpss:// www.basf.com/sunscreen-simulator. Accessed 14 November, 2017

The UVB-biased sunscreen is shown on the left. It has has a label SPF of 50 and claims broad-spectrum activity as the CW is 372 nm. The real-life SPF (called the in silico SPF by industry) is only 25 – 50% of what is claimed on the label. The UVA-PF is 9.0 with a UVA-PF/SPF = 9/50 or much less than the 1/3 required in the EU. This sunscreen transmits about 3 times the amount of UVR (mainly UVA) to the skin than would an ideal sunscreen or neutral density filter-like textiles.²²

The ideal sunscreen on the right has a label SPF of 50 and an in silico SPF 47. The UVA-PF is 35 and it performs almost the same as the ideal sunscreen or neutral density filter.²² This level of UVA protection and the proximity in the spectral profile to ideal spectral homeostasis ensures that the sunscreen performs like shade or textiles in reducing the quantity and not the quality of the terrestrial spectrum.

An important takeaway from this discussion is that SPF values and broad spectrum label claims may be seriously misleading. Media reports in 2015-2017 from Consumer Reports, the BBC in the UK, CBS, NBC, and CNN in the USA, have all presented data that,

up to 50% of brand name sunscreens fail to achieve even 50% of their labeled SPF values even if applied properly and at specified intervals.

The assay that uses a UV solar simulator overestimates the SPF that would be expected in natural sunlight. Products labeled SPF 50+ may not be able to achieve a protection against sunlight of more than SPF 25. The popular interpretation of the SPF to mean how much longer skin covered with sunscreen takes to burn in sunlight compared with unprotected skin can no longer be defended.

The premise that SPF can be used to plan your exposure time is based on several invalid assumptions. The lamp used in calculations for label purposes only emits 290-400 nm based on the false assumption that the erythema reaction was mediated only by UVR (UVB and UVA). Visible Light (VL at 400-740 nm) and Infra-red (IR at beyond 740 nm) contribute to the erythema response. Sunlight has more UVA (up to 5X) than the testing lamp emission.

Industry and physicians continue to perpetuate the travesty and advise consumers that the SPF can be used to calculate the safe exposure time for protected skin outdoors. Keen observers know otherwise—fair individuals mostly get sunburned with prolonged outdoor exposure or during tropical vacations, despite using high SPF sunscreens and stringently following all the re-application instructions.

We take a precautionary approach and advise patients to use half the value of the label SPF to calculate your safe exposure time outdoors. Sunscreens with spectral homeostasis and high UVA-PF values will have a Real Life SPF that is closer to the in vivo lab SPF value.

Preventing skin cancer and photoaging

As I have discussed, any cellular pathway to skin cancer and photoaging requires a UVA insult. A sunscreen without adequate UVA extinction cannot prevent either to any significant degree.

Sunscreen manufacturers are not required to measure UVA-PF values in North America. This metric can be determined in silico and provides a useful alternative as a valid substitute for the actual in vitro UVA-PF ISO 24443 test.²⁴ Even with the 3% avobenzone used in most brand name sunscreens for UVA protection, the UVA-PF can only reach 9.0 in silico. The other UVA filter available for general use is zinc oxide. At a maximum level of 25%, zinc oxide can reach a UVA-PF value of 12-15 with innovative dispersion, which is acceptable UVA protection at SPF 30 to approach spectral homeostasis.

Milestone studies from Australia point the way to effective prevention of photocarcinogenesis and photoaging. Regular daily use of an SPF 15 or higher sunscreen reduces the risk of developing squamous cell carcinoma by about 40%,²⁵ the risk of developing melanoma by 50%, and invasive melanoma by 71%.²⁶ The sunscreen used in these studies contained 8% of Ethylhexyl methoxycinnamate and 2% butyl methoxydibenzoylmethane (avobenzone)—a SPF of 16. The CW was only 366 nm and UVA-PF in silico was only 4.1.²² The UV index would be 0.25 or ¼. The product would not meet the present EU criteria or the FDA standard of 370 nm. Despite low SPF value and lack of adequate UVA protection, the investigators demon­strated protective effects. The same group showed that daily use of sunscreens with SPF 15 or higher had 24% less skin aging than those who do not use daily sunscreen.²⁷

Another landmark Australian study found that sunscreen applied to the skin before exposure to 2 MED solar-simulated ultraviolet radiation (SSUVR) completely blocked the effects of DNA damage, p53 induction, and cellular proliferation in both melanocytes and keratinocytes.²⁸

The implications for the future are promising since properly applied sunscreen can protect the repair p53 and inhibit the cellular basis of all three forms of skin cancer. The sunscreen used in the study was apparently a commercial product with zinc oxide and titanium dioxide as the active ingredients. No other details on SPF, CW, and UVA-PF are available. And since the active ingredient concentrations are unknown, these metrics cannot be calculated in silico.

It still shows that a sunscreen with a UVA/UVB filter in zinc oxide can completely inhibit the cellular cascade in the genesis of skin cancer. The sunscreen likely had at best a modest UVA-PF and reinforces the Green et al. studies that daily use of a UVB-biased sunscreen with low UVA-PF levels effectively prevented UVR mediated damage.^25,26 Zinc oxide levels of 15-25% can attain UVA-PF values of 12- 15, higher than typical 3% avobenzone sunscreens with a UVA-PF around 9-10.²²

Since label values of SPF and UVA/Broad Spectrum in North America may be inaccurate, consumers need to look for specific UV filters and their concentrations to assure that they get the best UVR protection, particularly in UVA1 (340-400 nm).

Related content: A Little Lump Or Bump On The Eyelid? Beware, It Could Be Skin Cancer

First, determine if the filter is likely to be absorbed into the blood?

The first driver of consumer choice should be whether a UV filter has any chance of dermal penetration to achieve blood and tissue levels. The author believes that even minimal scarcely detectable levels of a UV filter are too much where human safety is concerned. Soluble filters with a molecular weight (MW) <500 Daltons are absorbed by the skin and reach the blood,²⁹ potentially causing side effects in humans.

Many popular UV filters are <500 Daltons in MW and should be prudently avoided, for the risk of absorption. These include but are not limited to

• avobenzone
• oxybenzone
• homosalate
• octisalate
• octocrylene
• 4-methyl-benzylidine camphor
• regular (non-encapsulated) octinoxate.

Only avobenzone is a real UVA filter, and a typical sunscreen using these filter combinations (see Figure 2) gives undesirable UVB-biased protection.²²

To avoid dermal absorption, look for products with insoluble or particle type filters that have an MW > 500 Daltons:

• zinc oxide
• titanium dioxide
• encapsulated octinoxate
• ecamsule (Mexoryl SX™)
• drometrizole (Mexoryl XL^™)
• bemotrizinol (Tinosorb S™)
• biscotrizole (Tinosorb M™)
• Parsol SLX™
• iscotrizinol
• octyl triazone
• bisdisulizole disodium

“Newer” UVA and broad spectrum filters—bisoctrizole, bemotrizinol, and drometrizole used globally for over 15 years—are still being blocked by the FDA. Zinc oxide, titanium dioxide, encapsulated octinoxate, and ecamsule (patented to L’Oreal) are approved in North America. Ecamsule was approved in 2006 for limited use below a specific concentration.

Second, determine the filter concentration for good UVA1 protection

The next driver of consumer choice should be to look for a specific filter in the concentration that will give the best UVA1 protection. Zinc oxide is approved worldwide as a UVA filter. Its action spectrum declines beyond 370 nm, but products with 22% zinc oxide have a CW 377 nm, a UVA-PF of 12, and a PASS in the EU for a SPF 30 sunscreen that will provide better protection than the best USA sunscreen using avobenzone.²² A product with 3% avobenzone, 6% oxybenzone, 5% octisalate, 15% homosalate, and 10% octocrylene has a CW of 372 nm, UVA-PF of 10, UV Index of 0.22, or a FAIL in the EU.²²

The author has developed a new formula with 25% zinc oxide, now in the approval process at Health Canada, with a CW 377 nm and UVA-PF of 18 (ISO 24443), raised beyond its usual in silico level of 13 by a proprietary dispersion method, using an Ecocert™ certified organic material that elevates UVA attenuation up to 60% at the same ingredient concentration (patent pending).

Since 2000, there are new global UV filters that are truly broad spectrum with large MW above 500 Daltons, ideal from every aspect, with biscotrizole as the most efficient UVA1 filter. Canada has given special approval for bemotrizinol and biscotrizole to specific manufacturers.

The USA has not benefited from these innovations, despite the Sunscreen Innovation Act (SIA), signed into law by President Obama in 2014, that defined time limits for the different steps of the approval process. The FDA inexplicably continues to question the safety of these filters used for almost 20 years around the entire globe without issues. They cite concerns about dermal penetration of the pending filters. The basis for this is a mystery, as most of the soluble organic UV filters they approved prior to 1997 have definite cutaneous absorption with MW <500 Daltons, while several of the pending UV filters have a large MW >500 Daltons, and should not have any dermal penetration based on basic physiology.

Additional comments and concerns

• For doctors involved with rejuvenation, prescribing a balanced sunscreen with adequate UVA1 protection is a responsible practice standard. This is an integral part of care after rejuvenation procedures. Without balanced protection, patients resume accelerated photoaging from UVA1 exposure and the treatment benefit may be quickly lost.

• Tanning beds cause higher risks for skin cancer and providing a patient with a UVB-biased sunscreen to prevent the increased risks for skin cancer related to tanning beds should also be considered as an undesirable measure.

• Caregivers rarely advise pregnant women and parents about the absorption of soluble sunscreen filters through skin into blood and their possible hormone disrupting and carcinogenic effects. The CDC in the USA reported that 96.8% of Americans had benzophenone (oxybenzone) in urine possibly from its pervasive use in sunscreens and cosmetics.³⁰ Eighty-five percent of nursing mothers in an EU population had at least one UV filter in breast milk.³¹ Oxybenzone was found in the urine (99%) and in amniotic fluid (61%) of patients having amniocentesis.³² In humans, any level of exposure at all may cause endocrine or reproductive abnormalities, particularly if exposure occurs during a critical developmental window. In human endocrinology, low doses may even exert more potent effects than higher doses.

• The combinations of particulate filters in optimal concentrations are now technically possible without any solubility and stability issues. Consumers can look for zinc oxide alone in concentrations >20% or 15-20% combined with 7.5% titanium dioxide or encapsulated octinoxate—both safe and effective UVB filters. In Figure 2 (above), the author has developed a dispersion with an ingredient load of 31% combining zinc oxide, bemotrizinol, and biscotrizole that attains an SPF of 30, UVA-PF of >35, both in silico and in vitro (ISO 24443), and a UV Index of >1.²²

• Health Canada has already specially approved about 10 sunscreens with bemotrizinol and biscotrizole, so this ultra-UVA sunscreen could soon be available to Canadians, but the FDA will likely still withhold ratification. Global citizens outside North America have access to this type of sunscreen. Imagine the possibilities for the prevention of skin cancer and photoaging with this level of ultra-UVA protection, and a flat spectral curve meeting the principle of spectral homeostasis proposed by Diffey 25 years ago. Such a sunscreen has an actual UVA bias that finally closes in on the protection afforded by textiles. The author believes high UVA sunscreens may potentially prevent up to 90% of NMSC and improve the 50% prevention rate already achieved for melanoma.

Related content: Dispelling Misconceptions About the Regulation of Cosmetics in the U.S.