Page 58 - Read Online

Anand et al. J Cancer Metastasis Treat 2019;5:6 I http://dx.doi.org/10.20517/2394-4722.2018.98 Page 3 of 14
[19]
of BCA chest wall lesions . Here, we seek to apply a different combination therapy PDT approach to
BCA. Using pre-clinical murine models and subsequent clinical trials in patients with skin cancer, we have
established that certain neoadjuvantal agents can significantly improve PpIX levels and distribution within
tumors when combined with ALA-PDT [9,20] . In this study, we explore an approach that combines two FDA-
approved drugs [capecitabine (CPBN) and ALA] and a standard light source used in dermatology, to deliver
a safer, more efficacious, and more convenient treatment for localized BCA tumors and metastases in mice.

The rationale for our approach is based upon several relatively new scientific principles. In the past,
several PDT regimens have been explored with different PS molecules, routes and doses of PS, and various
wavelengths, illumination protocols and oxygen concentrations (hypoxia vs. normoxia), in attempts to
[7-9]
improve PDT responses . However, since cell physiology (particularly the state of tumor differentiation)
plays an important role in PDT outcome, we considered biomodulation with neoadjuvants given prior to
[9]
PDT as another avenue for optimization . Towards this goal, we pioneered a concept called “differentiation-
enhanced” or “combination PDT”. Thus, cancer cells of different tissue origins, when pre-treated with
differentiation promoting agents, show elevated accumulation of PpIX and enhanced cell death after ALA-
[23]
[24]
[9]
PDT . Three such agents, methotrexate [21,22] , vitamin D (calcitriol) and 5-fluorouracil (5-FU) were
shown to improve tumor responses in skin, prostate and BCA models when used as neoadjuvants for
PDT. Pretreatment with any of these agents resulted in 3- to 5-fold upregulation of mitochondrial PpIX
[9]
production . The effects on PpIX levels were selective for tumor cells and did not occur in adjacent skin [24,25] .
We recently worked out the molecular mechanisms for this effect. Coproporphyrinogen oxidase, a heme
enzyme located upstream of PpIX, is upregulated; ferrochelatatse located downstream of PpIX, is decreased;
and the net effect favors PpIX buildup [23,26] . An important feature of these neoadjuvants is that they drive
cancer cells toward a more highly differentiated state, as determined by the elevated expression of E-cadherin
in tumors. Using a BCA model (MDA-MB-231 cells implanted in breast fat pads of nude mice), we showed
that pretreatment with calcitriol (the active form of Vit D) prior to ALA-PDT resulted in an enhanced
[25]
therapeutic response relative to ALA-PDT alone . Using topical 5-FU, we recently completed a clinical
trial in patients with actinic keratosis (AK; pre-SCC), and showed that combination PDT with topical 5-FU
[20]
significantly enhanced the PDT efficacy as determined by increased clearance of the AK lesions .

A major reason for PDT treatment failure is the heterogeneity of PpIX distribution within the target tissue.
Some microanatomical pockets, located deep inside the tumor, produce minimal PpIX due to poor PS
[26]
penetration and suboptimal physiological response, thereby escaping PDT photodamage . When 5-FU
is given as a neoadjuvant prior to ALA, PpIX levels are raised in the vast majority of tumor cells, resulting
[24]
in enhanced cell death following PDT in both skin tumors and in murine breast tumors . However,
the systemic dose of 5-FU required to achieve this effect is quite high and might be toxic in humans. In
the current study, in order to circumvent toxicity, we used an alternative drug that is FDA-approved for
treatment of metastatic BCA [27-29] . CPBN is a precursor form of 5-FU that is selectively converted into the
active end product, 5-FU, primarily in cancer cells. CPBN is converted to 5-FU in three enzymatic steps
requiring carboxylesterase, cytidine deaminase (CDA), and thymidine phosphorylase (TYMP) [28,30] . Cancer
cells have relatively high levels of CDA and TYMP, explaining the increased sensitivity of cancer cells to
[28]
CPBN, and the low adverse effects of this drug on non-tumoral tissues . To test and develop a combination
approach with CPBN and PDT, we used a mouse BCA model (the 4T1 cell line; a triple-negative BCA
equivalent to human stage IV) that has already been previously utilized in many studies of BCA [31,32] . The
long-term goal of our study was to develop a better mechanistic understanding of the CPBN-PDT approach,
for possible translation of this concept to the clinic for treating localized BCAs.

METHODS
Cell culture
4T1, a murine breast carcinoma line (a triple-negative breast cancer and stage IV human BCA equivalent)
and Bioware® Brite Cell Line 4T1-Red-FLuc, were purchased from animal type culture collection and Perkin