Chemotherapy-induced alopecia (CIA) is characterized by absent or incomplete hair regrowth after completion of chemotherapy. Alopecia that persists beyond 6 months after completing chemotherapy is defined as persistent chemotherapy-induced alopecia (PCIA). The incidence ranges from 0.9% to 43%. The drugs most frequently associated with PCIA are busulfan and taxanes (docetaxel/paclitaxel). The clinical spectrum is characterized by a noninflammatory alopecia with diffuse involvement and reduced hair shaft thickness. Trichoscopic evaluation is crucial before, during, and after chemotherapy. Up to 30% of patients, prior to initiating chemotherapy, present findings consistent with miniaturization, anisotrichia, and decreased hair density. Early detection of these findings may predict a guarded prognosis for hair regrowth, particularly in patients receiving adjuvant antiestrogen hormonal therapy. Trichoscopy provides information on the phase of treatment during chemotherapy and the expected regrowth findings after its completion. Currently, there are therapies with evidence supporting their role in preventing hair loss prior to chemotherapy, such as scalp cooling therapies, without increasing the risk of metastatic disease and with a significant impact on quality of life. Topical minoxidil 2–5% and low-dose oral minoxidil (1.25–5mg) have shown promising results in promoting hair regrowth in patients with PCIA. There is evidence that spironolactone promotes hair growth and reduces the progression of alopecia, although there are no studies evaluating it as monotherapy in PCIA. It is essential to standardize treatment response measures with trichoscopic follow-up to guide decision-making regarding therapeutic management in patients with PCIA.
Chemotherapy-induced alopecia (CIA) is characterized by absent or incomplete hair regrowth after completion of a chemotherapy protocol.1 During chemotherapy, hair loss generally occurs in an anagen effluvium pattern that resolves spontaneously within the 6 months following completion of chemotherapy without any treatment.2 Alopecia that persists beyond 6 months after completion of chemotherapy is defined as persistent chemotherapy-induced alopecia (PCIA), which significantly impacts patients’ quality of life.3
EpidemiologyThe incidence rate of CIA is estimated at 65%. Its prevalence and severity are variable and depend on the chemotherapeutic agent and treatment protocol. Frequencies vary, being highest for paclitaxel (80%) and doxorubicin (60–100%).4
The incidence rate of PCIA ranges from 0.9% to 43%.5 One of the main risk factors for PCIA is conditioning regimens with cyclophosphamide (Cy) plus busulfan, associated in 66% of adults receiving it and up to 20% of the pediatric population.6 An incidence rate of PCIA of 30% has been reported in patients treated with taxanes (paclitaxel and docetaxel), with up to an 8-fold increased risk.6,7 Adjuvant antiestrogen hormonal therapy in patients with breast cancer causes alopecia (EIA), although the frequency and pathophysiology leading to progression to PCIA remain unknown.
EtiopathogenesisEtiopathogenesis remains unknown. Single nucleotide polymorphisms in the ABCB1 gene have been reported as being associated with an increased risk of developing PCIA.8 Permanent damage to the hair follicle may result from cytotoxicity affecting stem cells or from detachment of matrix cells from the dermal papilla.9 Chemotherapy affects signaling to the telogen germinal unit (TGU).10 Endocrine changes such as hypothyroidism and hypogonadism have been reported as possible contributing factors.11 However, in case series of patients with PCIA, levels of iron, zinc, thyroid function, and hormonal parameters (androgens and gonadotropins) were evaluated and found to be normal.9
Several hypotheses have been suggested to explain morphological findings in patients receiving taxanes and adjuvant hormonal therapy. Chemotherapeutic agents activate the dystrophic catagen response pathway, causing damage by targeting highly proliferative matrix cells in the hair bulb. The effect of chemotherapy on the hair bulb is directed toward stem cells, leading to collapse of follicular immune privilege, antigen exposure, and a secondary immune response similar to alopecia areata (AA). It has been hypothesized that adjuvant antiestrogen hormonal therapy may maintain activation of the dystrophic catagen response pathway initiated by chemotherapy, resulting in a morphological state of “inert follicles,” with miniaturization and an increased number of hair follicles in telogen resembling female pattern hair loss (FPHL). Hair follicle alteration would follow the following sequence: areata-like peribulbar lymphocytic infiltrate leading to exit from anagen with inflammatory telogen; “inert follicles”; and androgenic miniaturization induced by antiestrogen drugs.12
ClassificationThere is no consensus for assessing the severity of PCIA, although several scales exist (Table 1). Studies exploring therapies for CIA lack a standardized scale to evaluate hair loss and regrowth.13
Comparative table of scales for chemotherapy-induced alopecia (CIA).
| Scale | Grade 1 | Grade 2 | Grade 3 | Grade 4 | Grade 5 |
|---|---|---|---|---|---|
| WHO classification, 1981 | Minimal hair loss | Moderate patchy hair loss | Complete but reversible hair loss | – | – |
| CTCAE V3.0, 2006 | Thinning or patchy | Complete | – | – | – |
| Olsen CIA scale, 2007 | Minimal grade 1: 1–24% hair loss | Moderate grade 2: 25–49% hair loss | Grade 3: 50–74% hair loss | Extensive grade 4: 75–99% hair loss | Complete grade 5: 100% hair loss |
| MASCC, 2010 | Grade 1: <50% hair loss, no intervention | Grade 2: local or noninvasive interventionGrade 2A: 50–74% hair lossGrade 2B: ≥75% hair loss, wig use with psychosocial impact | – | – | – |
| CTCAE V4.0, 2010 | Grade 1: <50% hair loss | Grade 2: >50% hair loss with psychosocial impact | – | – | – |
Patterns of hair loss may be diffuse, affecting the entire scalp (Fig. 1a), diffuse with predominance at the vertex14 (Fig. 1b), or diffuse and patchy (Fig. 1c), with or without associated loss of eyebrows, eyelashes, and body hair.12
(a) Diffuse alopecia affecting the entire scalp. Multiple postinflammatory macules due to acneiform reaction from systemic steroids are observed. (b) Diffuse alopecia with predominance at the vertex. Hair is sparse, fine, and short. (c) Diffuse and patchy alopecia with frontotemporal involvement.
The pattern of hair loss in EIA is practically identical to that seen in androgenetic alopecia (AGA), although loss of eyebrows and eyelashes may also occur.6
TrichoscopyTrichoscopy prior to receiving chemotherapy may be normal in up to 70% of patients. In the remaining 30%, anisotrichia, miniaturization, and reduced hair density are observed. Between weeks 2–3 after initiation of chemotherapy, black dots, broken hairs, exclamation mark hairs, flame hairs, circular hairs, vellus hairs, comma hairs, Pohl-Pinkus constrictions, and miniaturization can be observed (Fig. 2). At the end of chemotherapy, 3D yellow dots, short fine regrowing hairs, and scattered black dots are observed, with the most relevant trichoscopic finding being the presence of circular hairs. We detail the trichoscopic findings, their clinical explanation, and the phase of chemotherapy in which they can be found15,16 (Table 2).
Main trichoscopic findings in CIA, clinical interpretation, and phase of the chemotherapy cycle.
| Trichoscopic finding | Interpretation | Chemotherapy phase | Differential diagnosis |
|---|---|---|---|
| Broken hairs | Hair shaft damage at the follicular ostium | Onset | Alopecia areata |
| Exclamation mark hairs | Progressive reduction of follicular mitotic activity without entering telogen phase | Onset | Alopecia areata |
| Flame hairs | Fragmentation and multifocal pigmentation of the hair shaft at the follicular ostium | Onset | Trichotillomania |
| Pohl-Pinkus constrictions | Alternating reduction/increase in keratinocyte proliferation | Throughout treatment | Pseudomonilethrix, alopecia areata |
| Yellow dots | Delay in return to anagen phase | Early weeks | Alopecia areata |
| Circular, vellus, and pigtail hairs | Regrowth in anagen phase | Long treatment protocols | – |
| White distal and black proximal hairs | Reactivation of melanocytes | Post-chemotherapy (3 months) | Alopecia areata |
There are no defined histopathological criteria.17 The TGU described by Headington in 1984 corresponds to the lowest part of the hair follicle in healthy telogen.18 TGU are round or oval in transverse sections. However, in PCIA they are surrounded by fibrosis without terminal hairs attached to basaloid aggregates. In more superficial sections, only dilated sebaceous ducts and cystic structures are observed.17
There is the presence of mild lymphocytic infiltrate, perifollicular and/or perivascular, with a reduction in terminal hairs and an increase in vellus hairs, similar to AGA. In up to 20% of patients, histopathology is normal.19
In another study of PCIA patients treated with taxanes for breast cancer plus adjuvant hormonal therapy, 100% showed preserved follicular units, reduced hair density, with an increased anagen:telogen ratio of 3.1:1 and an increased number of vellus follicles (terminal:vellus ratio 1:1), with preservation of sebaceous glands, miniaturization, and presence of terminal avascular fibrous tracts.14 Additionally, 60% showed pigment deposits in telogen follicles, dysmorphic TGU in 90%, trichomalacia in 30%, and peribulbar lymphoid infiltrate in 20%. This diversity of findings can be observed in both AGA and AA, depending on the stage of PCIA evolution. One study found that follicular density is significantly reduced in PCIA vs AGA.20
PrognosisMost CIA cases are reversible once chemotherapy has been completed. Patients treated with busulfan, cyclophosphamide, and taxanes are at higher risk for PCIA. Other risk factors include age and prior exposure to radiation.4 After 6 months to 1 year following the last chemotherapy session, up to 20% of patients show no increase in the number of terminal hairs or follicular units, with persistence on trichoscopy of vellus and circular hairs, predominantly in androgen-dependent areas. This may be due to latent FPHL or damage to hair follicle stem cells.15
Quality of lifeCIA represents one of the adverse effects that most negatively impacts patients’ quality of life.21,22 The degree of hair loss does not correlate with the impact on quality of life. Up to 8% of women decide not to undergo chemotherapy due to fear of hair loss.9 PCIA in the pediatric population increases rates of anxiety and depression.14
TreatmentPreventive and therapeutic management strategies are limited, with variable efficacy, and aim to reduce hair loss and stimulate regrowth.
Preventive treatmentScalp cooling during chemotherapy reduces alopecia by decreasing blood perfusion to the hair follicle, with efficacy of up to 71.7%.16 In a recent meta-analysis, no evidence of increased risk of scalp metastatic disease was demonstrated. PCIA.23 There are 2 automated devices approved by the FDA (Food and Drug Administration of the United States) for use during chemotherapy in solid tumors, based on prospective clinical trials in breast cancer patients: DigniCap® and PAXMAN®.24 In these trials, success was defined as less than 50% hair loss. Results of these studies are variable, with success rates ranging from 39.3% to 71%, with better outcomes in patients treated with non-anthracycline regimens. The strength of recommendation for scalp cooling to reduce CIA severity in breast cancer is B according to the SIGN system. Results of these studies in patients with early breast cancer are presented (Table 3).
Clinical trials evaluating the efficacy profile of different scalp cooling devices.
| Trial | Device | Chemotherapy received | Success rate | SIGN level of evidence |
|---|---|---|---|---|
| Rugo HS et al. (2017) | DigniCap® | Taxanes (n=101) | 66.3% | 2++ |
| Nangia J et al. (2017) SCALP trial | Paxman® | 182 patients randomized 2:136% anthracyclines64% taxanes alone or combination | 50.5% (overall)a 16% in anthracycline group | 1++ |
| Smetanay K et al. (2019) COOLHAIR trial | DigniCap® | Taxanes and/or anthracyclines (n=79) | 39.3% (no difference in anthracycline group) | 2++ |
| Munzone E et al. (2019) | DigniCap® | Anthracyclines+taxanes (n=139) | 43% | 1++ |
| Giarratano T et al. (2020) | DigniCap® | Anthracyclines and/or taxanes (n=135) | 60% (71% in taxane group vs 54% with anthracyclines) | 2++ |
Currently, several studies include groups of patients with different neoplastic conditions (predominantly breast cancer) and different treatment regimens with secondary PCIA.6 Various treatment regimens have been used with topical minoxidil 2–5% and/or oral minoxidil 1–5mg (LDOM), with or without spironolactone 50–100mg, for periods ranging from 6 weeks to 20 months, with variable responses in hair regrowth. Some studies report no response to 5% minoxidil after more than 3 months of use, while others describe better results, particularly in protocols combining LDOM. Study limitations include small sample sizes and heterogeneity in underlying disease treatments. The strength of recommendation for restoring hair density using topical or oral minoxidil in PCIA is D according to the SIGN system (Table 4).
Studies reporting response rates in PCIA with topical minoxidil and/or LDOM with spironolactone.
| Study | Primary cancer | Chemotherapy (n patients) | Adjuvant endocrine therapy (Yes/No) | Clinical subtype/PCIA grade | Treatment response | SIGN level of evidence |
|---|---|---|---|---|---|---|
| Iorizzo et al. | Breast, lung, myelodysplasia, Hodgkin lymphoma | Taxanes (n=10); Cyc (n=10); non-taxanes (n=16) | No | Grade 1: 47%Grade 2: 53% | Oral minoxidil 1.5–2.5mg for 6–12 months; Grade 1: 100% improvement; Grade 2: 75% improvement | 2+ |
| Kang et al. | Breast | Taxanes (n=30); fluorouracil+doxorubicin+Cyc (n=1) | YesSERM (n=7)AI (n=4) | Diffuse: 45%AGA-like: 55%Grade 1: 71%Grade 2: 26% | Oral minoxidil 1.25–5mg+topical minoxidil for 17–20 months | 2+ |
| Lyakhovitsky et al. | Non-Hodgkin lymphoma | Busulfan, Cyc, cytarabine (n=1) | No | Diffuse alopecia, fine short hair | Oral minoxidil 1.25–5mg for 6 months with complete regrowth | 3 |
| Bhoyrul et al. | Breast | Taxanes (n=92); Cyc (n=8)±epirubicin, methotrexate, fluorouracil | Yes Tamoxifen (n=49)AI (n=47) | Diffuse: 39%AGA-like: 60%Eyebrows/eyelashes: 18–32% | Topical minoxidil 2–5%+oral minoxidil 0.5–10mg+spironolactone 25–100mg or bicalutamide 10mg or flutamide 50–75mg for 6–12 months; improvement from Sinclair 4 to 3 | 2+ |
| Slaught et al. | Breast | Docetaxel+carboplatin+trastuzumab (n=1); paclitaxel+doxorubicin+Cyc (n=1); Cyc+methotrexate+fluorouracil+docetaxel (n=1) | Yes Tamoxifen (n=1) AI (n=1) | Diffuse: 33%AGA-like: 66%Eyebrows/eyelashes: 33% | Topical minoxidil 2% for 10 months with mild improvement in 1 patient | 3 |
| Freites-Martínez et al. | Breast, hematologic, ovarian, others | Taxanes (n=80)Cyc+methotrexate+fluorouracil (n=13)Others (n=5) | No | Diffuse: 41%AGA-like: 59%Eyebrows/eyelashes: 37%Fine hair: 37% | Minoxidil 5% and/or spironolactone with moderate to significant improvement in 36/54 (67%) | 2+ |
| Werbel et al. | Breast | Docetaxel+pertuzumab, carboplatin, trastuzumab (n=1) | Yes (tamoxifen) | AGA-like+eyebrows, eyelashes, axillary, pubic, and upper lip hair | Topical minoxidil 5% for 10 months with significant improvement | 3 |
| Fonia et al. | Breast | Taxanes (n=10) and/or epirubicin, doxorubicin, Cyc, carboplatin, fluorouracil | Yes Tamoxifen (n=3) Herceptin (n=5)AI (n=2) | Diffuse: 40%AGA-like: 50%Patches: 10%Eyebrows/eyelashes: 30%Body hair: 50% | Minoxidil 5% for 6 months; 1/10 complete regrowth | 2+ |
| Yang et al. | Acute myeloid leukemia | Busulfan+Cyc (n=1) | No | Diffuse | Minoxidil 1mg for 6 weeks with cosmetically significant regrowth | 3 |
| Kluger et al. | Breast | Taxanes (n=20) and/or epirubicin, Cyc, fluorouracil | Yes Tamoxifen (n=8)AI (n=9)Exemestane n=3) | Ludwig scaleI: 5%II: 63%III: 32%Eyebrows/eyelashes: 100%Body hair: 80–90% | Minoxidil 2–5% for >3 months in 14/14 with no regrowth | 2+ |
| Prevezas et al. | Breast | Docetaxel (n=1); paclitaxel (n=1) | Yes Letrozole (n=1) | Diffuse: 50%AGA-like: 50%Eyebrows/eyelashes/body hair: 50% | Minoxidil 5% with mild improvement in 1 patient | 3 |
| Tosti et al. | Myeloproliferative syndrome, multiple myeloma | Busulfan (n=2); Cyc (n=1); melphalan (n=1) | No | Diffuse: 100% | Minoxidil 5%, slight hair thickening without cosmetic improvement | 3 |
AGA: androgenetic alopecia; Cyc: cyclophosphamide; SERM: selective estrogen receptor modulator; AI: aromatase inhibitor.
In a study of 54 patients, most of whom had breast cancer and developed PCIA after taxane-based chemotherapy, 67% (n=36) showed moderate to significant improvement after treatment with topical minoxidil with spironolactone or topical minoxidil alone.25 Although this study does not evaluate spironolactone monotherapy in PCIA, another study shows that spironolactone as monotherapy at a dose of 200mg/day promotes hair growth and reduces alopecia progression in patients with FPHL.26 Recent studies suggest that spironolactone may reduce cancer rate. More studies are required to confirm its efficacy in CIA and evaluate its potential adverse effects.27 The strength of recommendation for restoring hair density using spironolactone in PCIA is D according to the SIGN system.
A botanical-based lotion (CG428) containing lemon, cocoa, guarana, and onion has been described, which prolongs the anagen phase, improving hair density and thickness. Although there was an improvement in hair density of 34.7% vs 24.9% vs placebo, the result was not statistically significant (P=.37) in a population of women with PCIA (SIGN level of evidence 2−).28
The use of immunomodulators such as trichloroammonium tellurate (AS101 0.01% topical) has been described, with efficacy demonstrated in murine models and in humans in topical form. This drug inhibits pro-inflammatory cytokines such as IL-10, delaying entry into catagen and inducing anagen. In 1 of the 3 patients with PCIA in whom it was used, greater than 50% improvement in hair regrowth was observed.29 (SIGN level of evidence 2−).
Vasoconstrictive agents (epinephrine/norepinephrine), immunosuppressants (cyclosporine), cytokines and growth factors (interleukin 1), cell cycle modulators (calcitriol, CDK2), apoptosis inhibitors (caspase-3), and monoclonal antibodies (MAD11) have been evaluated in animal models, showing promising results for promoting hair regrowth or preventing hair loss.16 (SIGN level of evidence 3, strength of recommendation D)
In ex vivo models, activation of the PPARγ signaling pathway in a prophylactic manner reduces cytotoxicity to the hair bulb caused by 4-hydroperoxycyclophosphamide (4-HC), protecting hair follicle stem cells from apoptosis and preventing their depletion.30 (SIGN level of evidence 3, strength of recommendation D)
Non-pharmacological therapiesThe use of wigs, camouflage, or micropigmentation are tools that mitigate the impact of chemotherapy on the self-esteem of patients with CIA and PCIA.
Low-level laser therapy (LLLT), or photobiomodulation, is a noninvasive tool useful in CIA, with success rates similar to those observed in AGA and AA in animal models, although controlled human studies are lacking. Platelet-rich plasma (PRP) and microneedling lack controlled studies in patients with CIA.16 Their efficacy profile has not been validated, partly due to their autologous nature and variability in preparation protocols. An ongoing study is evaluating the efficacy of PRP in the treatment of EIA and PCIA in patients with breast cancer, with results pending.27 Hair transplantation is ineffective in PCIA because alopecia is diffuse and the donor area is poor. All these approaches have a SIGN level of evidence 3 and strength of recommendation D.
ConclusionsCIA is a frequent and expected event during chemotherapy that may become permanent. Knowledge of the drugs associated with higher risk, along with trichoscopic evaluation prior to chemotherapy to detect undiagnosed conditions, helps the dermatologist establish a prognosis regarding hair recovery, preventive measures, and expected response to subsequent treatments. Prospective randomized clinical trials are required to standardize treatment response measures with trichoscopic follow-up, enabling better decision-making regarding pharmacological options for patients with PCIA. Finally, we propose an evidence-based algorithm for the management and treatment of oncology patients at risk of CIA and those who develop it (Fig. 3).
