Incidental perineural invasion (PNI) is identified on histopathology in clinically asymptomatic patients. Clinical PNI presents with sensory or motor changes or radiographic evidence of PNI. Management of PNI requires diagnosis, evaluation of extent, consideration of other high-risk tumor features, and treatment. For incidental PNI in basal cell carcinoma (BCC), Mohs micrographic surgery (MMS) or standard surgery with adjuvant radiation therapy provide similar cure rates of around 90% at 5 years. For squamous cell carcinoma (SCC) with incidental PNI, features such as nerve size, extent of invasion, and presence of other high-risk features determine whether adjuvant radiation therapy is recommended after MMS. Clinical P

1. One thing we do not like to see on a histopathology report of a basal or squamous cell carcinoma is the
presence of perineural invasion(PNI).Althoughbynomeansrare,approachesto PNIvaryandthere isno
clear-cut consensus.
An Australiangrouphas beentryingto bring order to this issue.Publishedinthe February2016 issue of
the Australasian Journal of Dermatology (Volume 57, Issue 1 February 2016 Pages 3–13) is an excellent
overview of their work to date. This article is mandatory reading for anyone treating skin malignancy.
The authorsstate that perineural invasion referstotumourgrowthin or around a nerve. Itoccurs bythe
contiguous spread of malignant cells along the potential space between a nerve and its surrounding
sheath. PNI occurs in less than 5% of all cutaneous malignancies. It is more common among squamous
cell carcinomas (SCC), involving 3–14% of cases, compared with 0.18% to 10% of basal cell carcinomas
(BCC).
Why is PNI important? It confers an increased risk of recurrence in both BCC and SCC and of the
development of metastasis in SCC, and a poorer prognosis due to more aggressive tumour behaviour.
However, the risk of death from PNI in BCC is far less. In SCC it’s presence is so important that in the
updated 7th Edition of the American Joint Committee on Cancer staging system for cutaneous SCC, the
presence of PNIisone of the high-riskfeatures that contribute to the upstaging of a T1 SCC to a T2 SCC.
From a clinician’spointof viewwhatdowe do withPNI?The authorsstate that the managementof PNI
requires:
1. Diagnosis through clinical assessment, imaging and histopathological review.
2. Evaluation of its neural distribution and extent.
3. Consideration of other high-risk tumour features present.
4. Decision on management which includes excision (preferably with margin control) and adjuvant
radiotherapy.
The authorsmake the followingpoint“The evidence base forthe managementof PNI consistsmainlyof
institutional observational cohortstudies,withthe inherentassociatedbiasesof suchstudies.Therefore,
our recommendations are based on this level of evidence in combination with expert opinion.”
DIAGNOSIS
Asstatedabove the mostaccurate diagnosisof PNUisbasedonclinical,radiological andhistopathological
assessment.

2. CLINICAL
The classic symptom we were all taught in medical school of PNI is formication or a sensation of ants
crawlingonthe skin. Sadly, 60–70%of patientswithhistologicallyconfirmedPNIare asymptomatic.Thus,
we needtomaintainahighindex of suspiciontodetectPNI,especially fortumoursoverlyingmajornerve
trunks and their branches.
In those who are symptomatic issues reported may include:
 Sensorysymptomsincludevariousformsof dysaesthesia,particularlyformication –the sensation
of ants crawlingunderneaththe skin,tingling,painandhypoaesthesiaornumbness. These most
often are seen in the distribution of the trigeminal nerve.
 Muscle weaknessandfasciculationmaybe describedbythe patientordetectedoncranial nerve
(CN) examination,mostcommonlyinvolvingthe facial nerve and its branches.A misdiagnosisof
Bell’s palsy may be made in these patients.
 Insome patientswithcranialnervepalsiesandsensorydeficitsthere maywell be nodocumented
history of non-melanoma skin cancer with PNI. In such circumstances, imaging followed by a
biopsy confirmation of suspiciously enlarged nerves may be warranted. In other words,
symptomatic PNI may be the first presentation of these malignancies. Imaging followed by a
biopsy confirmation of suspiciously enlarged nerves may be warranted.
Imaging
The authorsstate thatmagneticresonance imaging(MRI) neurographic protocol isthe preferredimaging
modalitytodetectandassessthe extentof macroscopicPNI,demonstratingsuperiorsoft-tissue contrast
andgreatersensitivityandspecificityinthe evaluationof large nerve PNIthancomputedtomography(CT)
and other imaging modalities.
Radiographicfeaturessupportive of perineural spreadinclude enlargementorabnormal enhancementof
the nerve, or obliteration of the normal fat plane surrounding the nerve. There is evidence that spatial
resolution is improved by using 3-Telsa scanners, which have high-field magnets instead of traditional
magneticfieldstrengths.Gadolinium-contrastMRIwithfatsuppressionfurtherincreasesthe radiographic
ability to detect early PNI, thus improving pre-surgical staging.
A negative MRIdoesnot,however,excludePNI,asfalsenegativesmayoccur.Accordingtoone study,the
sensitivityof MRI fordetectionof macroscopicPNIwas 95% but fell to63% for demonstratingthe entire
extentof disease. CTismore useful foridentifyingbone invasionthanchangesinthe nerve itself,andcan
detectthe erosionorenlargementof the foraminaassociatedwiththe involvedCN. Itisa betterimaging
modalitytoassessforthe involvementof regional lymphnodes,andMRIandCT complementeach other.

3. Histopathology
The commonestway that PNI isidentifiedisas an incidental findingona histopathologyreporti.e.with
no clinical suspicion.
HistopathologicallyPNIisdefinedas“the findingof tumourcellswithinanyof the threelayersof the nerve
sheath (epineurium, perineurium and endoneurium), or tumour cells involving at least 33% of the
circumference of the nerve.” The 33% rule isto differentiate PNI from tumour focally abutting a nerve.
Rememberapathologistonlyseesatinypercentageof anexcisedspecimen.Featuresthatsuggestnearby
PNI include perineural inflammation and nerve-fibre degeneration. If these are seen further sections
shouldbe cut intothe blocklookingfordefinite PNI. Inflammationinvolvingbothneural and non-neural
structurescanindicate non-specificinflammation,whereasisolatedperineural inflammationmaybe more
suggestive of nerve involvement by the tumour.
As ever, using a specialist dermatopathologist is the ideal with skin malignancy. This is especially so as
PNI can be mimicked histopathologically. Over-diagnosis of PNI leads to extra treatment morbidity. The
commonestmimicisperitumouralfibrosis, where concentricringsof fibroustissueare foundadjacentto
tumourcell nests andpresentin approximately5% of SCCand6% of BCCmaybe mistakenfornerve tissue
without the use of additional stains such as S100.
Epithelial sheathneuroma,re-excisionPNI,andreparativeperineural proliferation willall mimicPNIalso.
Epithelial sheath neuromas appear microscopically as discrete nerve complexes in the reticular dermis,
consisting of central nerve trunks enveloped by mature squamous epithelium. There is no associated
carcinoma or scarring fromprevioussurgery.Re-excisionPNIreferstothe presence of benignsquamous
epitheliumin the perineural spacesof previouslybiopsied sites. Thisis likely to reflect aberrant reactive
proliferationof traumatisedeccrine sweatglandsintoa plane of lowerresistance.Reparative perineural
proliferation consists of the appearance of concentric rings of spindle-shaped cells enveloping a nerve
adjacent to scarring and reparation from previous surgery. Immunohistochemistry can distinguish this
process from PNI. Spindle cells demonstrate negative staining for S100 and cytokeratins, but positive
staining for epithelial membrane antigen.
When found the pathologist should report the following descriptors of PNI:
 whether the PNI is intratumoural or extratumoural;
 below dermis or multifocal;
 the distance to the nearest margin; and
 the size of the nerves involved.

4. If your lab report mentions PNI but not the presence/absence of these other features you will have to
request it. This information influences management.
Classification
The broad categories of PNI are incidental or clinical.
Incidental PNI is identified only at histopathology in clinically asymptomatic patients with negative
imaging. This is also termed minimal or microscopic PNI.
Clinical PNI is when the patient exhibits sensory or motor changes, or there is radiographic evidence of
PNI. It may also be referred to as extensive or macroscopic PNI. It is uncommon to observe imaging-
positive PNI in an asymptomatic individual.
Doesthismatter?Yes,as itisof greatprognosticandhence managementsignificance. One studyfounda
5-year local control rate of 80% for cutaneous malignancies withincidental PNI,compared with 54% for
clinical PNIdespite aggressivetreatmentwithradiotherapywithorwithoutsurgeryorchemotherapy,or
both. The study population included patients with SCC, BCC and basosquamous carcinomas.
TREATMENT
Incidental PNI
For incidental PNI, Mohs Micrographic Surgery (MMS) has consistently demonstrated superior local
control rates compared to standard excisionfor NMSC with PNI. This is due to significant differences in
the technique used to examine excision specimens. Standard assessment with vertical sectioning
examines <1% of the margins. By using en face sections, MMS enables examination of close to 100% of
the peripheral and deep margins, allowing the better detection of PNI and more complete excision.
Ina recentstudy MMS plus adjuvantradiotherapy improvedcause-specificsurvival andlocal controlrates
inpatientswithincidentalPNI,comparedwithconventional surgical excisionplusART.The 5-yearcause-
specific survival for MMS patients was 84% compared with 68% for non-MMS patients, and the local
control rates were 86% and 76%, respectively. While thisstudyincludedpatientswithbothSCCandBCC,
results were not stratified by histological subtype and most (89%) cases were SCC. This limits the
application of these findings to other NMSC.

5. Remember that an SCC with incidental PNI is more likely to be larger, more poorly differentiated, have
greater subclinical extension i.e. bigger than they look and larger postoperative defectsthan tumours
withoutPNI. Asincidental PNIisassociatedwiththese otherpoorprognosticants, treatmentneedstobe
more aggressive and currently all the evidence points to MMS being the treatment of choice.
Even so, recurrence rates after MMS of tumours with incidental PNI are significantly higher than those
withoutPNI.Thiswasonce felttobe due to‘skiplesions’i.e.areasof uninvolvednervebetweendeposits
of PNI. This has been disproved and PNI is contiguous. Skip lesionsare processing artefacts, most likely
where asymmetrical tumourgrowtharoundanerve andtissuemanipulationduringsectioningmayresult
ina false negative margin. Otherexplanationsforskipareashave includedaninflammatoryreactionwith
immune-mediatedtumourregression,ortrue skippingof regionsof nervesassinglemalignant cellsalong
the perineurium.
After achieving histologically clear margins in tumours with PNI, some Mohs surgeons excise an extra
tissue levelandeitherexamine itonce againwithfrozensectionsorsenditforstandardparaffinsections
because of the possibilityof undetected PNI.Thisapproach,however,isnonspecificanddoesnot confer
assurance of complete tumour removal. Furthermore, the additional removal of tissue with clear
histological margins does not allow for maximal tissue conservation, which is particularly significant in
areas of functional and cosmetic importance.
So,if we want to improve local control ratesandmaximisetissueconservationwe needtoconsiderpost-
operative adjuvant radiotherapy to a wide local field. This can be effective in sterilising microscopic
depositsof cancercellsthatmaybe presentaftersurgical excisionand,inselectcases,electivelytreating
regional nodes without the need for regional surgery.
One of the problemswiththe publisheddataof PNIinNMSCisthat the published studiesoftenlumpBCC
and SCC with PNI together. It is felt that control rates for BCC with PNI are better than with SCC.

6. Treatment of SCC with Incidental PNI
The authorspoint out that in a large seriesof patientswith SCC withincidental PNI manywere curedby
MMS alone i.e. without adjuvant radiotherapy. The mere presence of incidental PNI in a SCC does not
mandate radiotherapy.Thus,theyoutline featuresotherthanPNIthat are importantin determiningthe
needforadjuvantradiotherapy.Inotherwords,whatotherprognosticfeaturesinfluence the decisionto
add radiotherapy to the treatment?
They list the following features;
 Extratumoural PNI, that is, invasion outside the main tumour mass, is associated with more
aggressive tumour behaviour than intratumoural PNI.
 PNI limited to small calibre nerves (<0.1 mm) likely carries a better prognosis with significantly
lower risks of recurrence and metastasis and increased disease-specific and overall survival
comparedto PNIof nerves> 0.1 mm in diameter.Atthe moment, furtherresearchis neededto
determine if 0.1 mm is the ideal cut off.
 The extent of intratumoural perineural involvement in SCC is also likely an important
prognosticant.There are betteroutcomeswhenPNIwaslimitedtoafew smalldermalnervesand
minimal focal disease,than with more diffuse perineural spread within the tumour mass. MMS
doesnotexamine the intratumoralmassandso preoperative biopsyspecimensor examiningthe
central debulking tissue, either with frozen sections at the time of Mohs surgery or subsequent
standard paraffin sections may be needed.
 Depth of invasion is also important with lesions limited to the dermis doing better than those
invading subcutis/muscle.
 PNI in a recurrent lesion carries a worse prognosis than that found at initial presentation.
Then there the adverse prognosticants unconnected to PNI i.e.:
 poor differentiation,
 tumour diameter ≥ 2 cm
 invasion beyond subcutaneous fat.
All of these are more likely in cases of PNI involving larger nerves.
Other factors include:
 Immunosuppression especially in transplant patients. SCCs in the immunosuppressed are more
biologically aggressive with poor differentiation, rapid growth, increased perineural and
lymphaticinvasion,andhigherriskof developingregional anddistantmetastases.Thus,incidental
PNI inan immunosuppressedpatientmayleadone to use adjuvantradiotherapy.Thisis also an
indication of why the level of immunosuppression needs to be kept under review.

7. Below is the algorithm for management of PNI in SCC.
A low risk lesion where surgery alone might be all that is needed includes incidental PNI that is
intratumoural, focal, involves small nerves, is limitedto dermis,and occurs in primary tumours. MMS is
the preferred primary excision strategy where PNI is diagnosed on biopsy.
Adjuvant radiotherapy should be considered if patients are immunosuppressed, or additional high-risk
tumour-related features are present such as tumour size ≥ 2cm, or poor tumour differentiation.
The medium-riskgroupincludesincidental PNIwithanyof the high-riskfeaturesidentified above.These
patients would possibly benefit most from ART regardless of the primary surgical approach used.
High-riskpatientsare those withclinical PNIinwhom managementmayinvolve nerve resection,usually
by a skull base surgeon,withthe additionof ART or alternativelydefinitive RT; these shouldgotoa head
and neck clinic.

8. The authorsaddressthe role of sentinel lymphnode biopsyinhighriskSCC.Asexpecteditiscontroversial.
It would need to be addressed on a case by case basis.

9. BCC with incidental PNI.
The authors state that cure rates for MMS and standard surgery plus adjuvant radiotherapy are similar
forthese muchlowerrisklesions(comparedtoSCC). Around90% 5 yeardisease freesurvival isseenwith
either. There are no data specificallyevaluating the outcomes of incidental PNI in BCC managed with
standard surgical excision alone.
BCC with PNI still has a higher recurrence rate than BCC without PNI.PNI is more frequently associated
withaggressiveBCCsubtypes,includinginfiltrative, morphoeic,sclerosingandmicronodularvariants.The
locally invasive nature of these BCC subtypes may warrant consideration of local ART, especially in the
setting of incidental PNI.
The algorithm below nicely summarises the authors approach to BCC with PNI.

10. What about justradiationtherapyas definitive treatmenti.e.withoutanyinitial surgerybeyondbiopsy?
Asexpectedthere isdebate.Itiscertainlyauseful modalityinnon-surgical candidates.The authorspoint
out that although traditionally restricted to those over 60 due to concerns re long term sequelae it has
been used in younger patients. It is contraindicated in naevoid basal cell carcinoma (Gorlins) and
xeroderma pigmentosum. These patients need surgical treatment and close follow up.

11. Thenwe come to clinical PNIi.e.the patientswithsymptoms/signsof PNIor PNIdetectable onMRI. You
will notcome acrosstoomanyof these butequallyyoudonow wanttomissthemwhentheydopresent.
These patientsreceive treatmenttailoredto theircase based on the extent andresectabilityof the PNI,
and alsothe patientprofile.A multidisciplinaryapproach,oftenincludingstagingandplanningwithMRI,
surgical excision under general anaesthetic, preferably with frozen section control, and RT is often
required.MMSof the primarytumourmay be a valuable supplement;however,patientswithclinicalPNI
oftenhave involvementof structuresthatprecludesthisapproach.Obviously, factorssuchaspatient age,
performance status and preference, together with the presence of comorbidities, also need to be
considered when determining appropriate management.
While the natural history of PNI usuallyinvolves central progression, this can be an unpredictable,long
and indolent process. Many years may elapse from diagnosis of a head and neck primary cutaneous
tumour with PNI before invasion into the central nervous system. Furthermore, the patient group that
appearsto be most at risk of PNIare olderCaucasianmen,many withcomorbiddisease anda historyof
multiple previousskincancers. These become particularlysignificantconsiderationswhencontemplating
aggressive treatment and weighing the potential survival benefit against the side-effect profile in this
subset of patients.
SCC portendsamore unfavourable prognosisthanBCCas for incidentalPNI.A recentstudyreporteda5-
year relapse-free survival of 39% for patients with clinical PNI and SCC versus 80% for BCC when both
patientgroupswere treatedwithradiotherapyorsurgery,orboth.ThusmanagementvariesfromSCCto
BCC with clinical PNI.
SCC with clinical PNI
Resection with margin control plus adjuvant radiotherapy offers the best chance of cure. What is
resectable nowisa differentkettle of fishtopreviously.Forexample,SCCwithextensiveintracranialPNI
involvingcranial nerves up to the Gasserian (Trigeminal) ganglion (zone 2) may be operable, and this
treatment potentiallyoffers improved survival rates with acceptable morbidity. This is of course very
major surgery and is limited to selected patients in specialised units.
Extension beyond the Gasserian or Geniculate ganglion (zone 3) generally deems a patient to be
inoperabledue tothe risksof exposingcerebrospinalfluidtotumourcells,andsubsequentseedinginthe
brainstemandspinal cord. Patientswithsuchextensive PNI,aswell asthose who are medicallyunfitfor
surgery,maybe managedwithRTalone (definitiveorpalliative),oralternatively,bestsupportivecare.RT
is often effective in palliating debilitating neuropathic pain and preventing, or delaying,the progression
of intracranial disease and its associated consequences. High-dose definitive RT alone can also offer the
chance of cure in ∼50–60% of suitable patients, but with associated acute and late side-effects.

12. For radiotherapy to get best results in these patients it is advised that the target volume include the
portionsof the nerve proximal anddistaltothe tumoursite,skininnervatedbythe diseasednerve,major
communicatingbranchesandthe compartmentinwhichthe nerve isembedded,suchasthe parotidgland
forCN VII.Thisisto treatbothantegrade andretrograde PNI,aswell ascrossoverspreadfromone major
nerve or branch to another. Regional nodes may also be included.
As you would expect with such dangerous lesions there are also treatments as yet considered
experimental.These include adjuvantordefinitive chemoradiotherapy,withplatin-basedchemotherapy
(as a radiosensitiser), in the management of clinical PNI in cutaneous SCC in select patients.
Where the tumourisconsideredinoperable,cetuximab(epidermal growthfactorreceptorinhibitor) may
provide palliative relief.
BCC with clinical PNI
The principlesare similaras for SCC withtreatmentcomingdown to surgery plus adjuvantradiotherapy
or definitive radiotherapy for inoperable lesions. Interesting there is a dearth of literature defining the
role of radiation treatment for BCC according to this article.
Experimental approaches
For advancedBCC withPNI,the use of Hedgehog(Hh) pathwayinhibitorssuchasVismodegibrepresents
an as yetunexploredtreatmentoption.VismodegibhasrecentlybeenTherapeuticGoodsAdministration
approved for the treatment of adult patients with metastatic BCC or with locally advanced BCC where
surgery and radiotherapy are not appropriate. A phase II trial found a response rate of 43% in patients
with locally advanced BCC, with a complete response reported in 21%. Remaining challenges with this
therapyinclude asignificantadverseeffectprofile,developmentof resistance anddisease reboundafter
the drug is discontinued. To overcome this, consideration is being given to the possible role for Hh
inhibitors as adjuvant therapy to surgical excision or MMS. The rapid effect of Hh inhibitors in shrinking
tumoursize mayenabledefinitivemanagementinanatomicallydifficultorhigh-risksitessuchasthe head
and neck.NewHhinhibitorsare alsobeingdeveloped,includingSaridegibandSonidegib,withthe aimof
reducing adverse effects while maintaining clinical efficacy.
I thinkthisisa must readarticle forall of usinvolvedinskincancermanagement.The presenceof PNIon
a histopathologyreportshouldtriggeryouto considermore aggressive management.Thisisdefinitelya
situation where it is good to ‘phone a friend’ or refer on. In my practice I tend to either call a radiation
oncologistwithextensiveskincancerexperience orone of the specialisedheadandneckclinicspecialists.