Description of the stage of development.
1. Summary of the development of the product.
Since its first therapeutic use in 1978, NSC 631570 (administered either as neoadjuvant
treatment before surgery or as combination therapy or alone) has been the subject of
numerous experimental and clinical tests.
NSC 631570 is a Chelidonium majus L. - thiophosphoric acid derivative, a complex of
Chelidonium majus L.-alkaloids with triethylene-thiophosphoric acid triamide (Thio-TEPA).
The injection solution contains NSC 631570 in a concentration of 1 mg/ml (at least 90%
Chelidonium majus alkaloid-thiophosphoric acid derivative and a maximum of 10% of free
Chelidonium majus alkaloids).
NSC 631570 (active substance: Chelidonium majus L. alkaloid-thiophosphoric acid derivate)
is readily soluble in water. Therefore it is possible to inject the drug intravenously. It has a
very strong affinity to cancer tissue and it accumulates only in cancer cells. This has been
proved by autofluorescence as well as by an American research team using a radioisotope
method (Nowicky et al, 1988; Hohenwarter et al, 1992; Thakur et al, 1992).
The substance is a bright yellow-brown crystalline powder. The injection solution is a
transparent, bright yellow-brown liquid with the aroma of freshly cut grass and a bitter taste.
The preparation comes as a sterile 0.1% (1 mg/ml) aqueous injection solution (pH: 3.5 to 6.5)
in amber-coloured ampoules of 5 ml, with no excipients. Under UV light NSC 631570 shows
a yellow-orange autofluorescence. Due to this autofluorescence NSC 631570 can also be
easily detected in tissues.
Celandine alkaloids and Thio-TEPA are the starting materials used for the production of NSC
631570. The alkaloids are hardly soluble in water. This means that it is not possible to give
intravenous injections. For this reason these drugs are always only administered orally.
However, by this means of administration the drugs cannot accumulate in cancer tissue.
Thio-TEPA is listed in many pharmacopoeia (e.g. UK, Japan, France, USA) and is approved
as a cytostatic in Austria. No free Thio-TEPA or aziridine ring compounds can be detected in
NSC 631570. Ukrain is therefore definitively different from the starting materials.
In vitro activity against cancer cell lines:
In vitro tests by the National Cancer Institute (NCI), Bethesda, USA, demonstrated cytotoxic
activity of NSC-631570 against all of the 8 colon cancer cell lines tested (pancreatic cell lines
were not part of this test program) at molar concentrations between 10-4.5 and 10-5.5
(corresponding to concentrations between ?7.6 ?g/mL and 76.0 ?g/mL). In contrast, 5-FU
barely showed any inhibition of the same cell lines at 100 to 1,000-fold higher concentrations,
not achieving lethal effects even at the highest concentration (10-2.5), in contrast to NSC-
631570 which is lethal at a concentration of 10-3.5, i.e. ?760 ?g/mL. The experiments also
show that the activity profile of NSC-631570 is clearly different from the profile of its basic
components thiotepa and chelidonine hydrochloride, both less active in the majority of the 53
cell lines tested.
NSC 631570 is produced from alkaloids from greater celandine and Thio-TEPA. These two
compounds are approved and clinically widely used. The National Cancer Institute (Bethesda,
Maryland, USA) has proved that NSC 631570 has a completely different effect on malignant
cells to Thio-TEPA (NSC 6396) and chelidonine hydrochloride (NSC 406034; chelidonine is
the main alkaloid of greater celandine and is an ingredient of many oral drugs).
For example:
34
• NSC 631570 is least effective [log(TGI) = -3.4] against leukaemia-HL-60(TB) in
contrast to chelidonine hydrochloride, which is very effective [log(TGI) = -5.4] and to
Thiotepa, which is only moderately effective [log(TGI) = -4.4].
• NSC 631570 is extremely effective [log(TGI) = -5.6] with Non-SmallLung-NCI-H460,
chelidonine hydrochloride less effective [log(TGI) = -4.0] and Thiotepa shows very
little effect [log(TGI) = -4.5].
• With Colon-SW-620 NSC 631570 is very effective [log(TGI) = -5.2],chelidonine
hydrochloride is not effective [log(TGI) = -4.0], and Thiotepa is also not effective
[log(TGI) = -4.2].
The enclosed profiles of these three different substances show very clearly that their effects
on the same cell lines are very different. Results of the National Cancer Institute, Bethesda,
USA, Human Cell Line Screen can be seen on the website of the Developmental Therapeutics
Program NCI/NIH (National Cancer Institute, National Institute of Health)
http://www.dtp.nci.nih.gov/.
In a series of experiments with 14 different cell lines of human and animal origin, including
normal and cancer cell lines, the effects of 4 different doses of NSC-631570 (0.1, 1.0, 10,
100mcg/ml) on DNA, RNA and protein synthesis were investigated by measuring the
incorporation of 3-H labelled thymidine, uridine and leucine (Nowicky et al, 1996). Usually, a
dose-dependent inhibition of all anabolic processes, DNA, RNA and protein synthesis was
found that was more pronounced in malignant cells than in normal cells, even in those normal
cell lines known for fast replication rates. According to the authors, no toxic effects were seen
in normal cells treated in doses that are 100% growth inhibitory to cancer cell lines.
Until now NSC 631570 has been tested on more than 100 cancer cell lines and revealed
malignotoxic action against all of them, including pancreas cancer cell lines, cis-platin
resistant cell lines and human tumour xenografts. At the doses at which NSC 631570 kills
cancer cells it does not affect healthy cell lines. The concentration of NSC 631570 which is
toxic for healthy cells is more than 100 times higher than the concentration lethal for all
cancer cell lines. Its therapeutic index is 1250 (Nowicky et al, 1996; Nowicky et al, 1996;
Panzer et al, 1998; Roublevskaia et al, 2000; Cordes et al, 2002).
Mechanism of action:
NSC-631570, in concentrations between 3.5 and 5 ?M (?2.5 to 5 ?g/ml), causes a dosedependent,
reversible arrest of dividing cancer cells (human epidermoid cell lines, American
Type Culture Collection A431 and ME180) in the G2/M phase and to apoptosis; normal
human keratinocytes serving as controls in this experiment remained largely unaffected
(Roublevskaia et al, 2000).
These observations could be confirmed in a recent independent experiment with pancreas
cancer cell lines (Ramadani et al, publication in preparation). It could be demonstrated that
NSC-631570 is a potent mitotic inhibitor, acting through stabilisation of monomeric tubulin.
When four different pancreatic cell lines (American Type Culture Collection cell lines
AsPC1, BxPC3, MiaPaCa2, Panc1) were incubated with NSC-631570, chelidonine or
thiotepa at different concentrations, only NSC-631570 and chelidonine led to a significant
dose and time-dependent accumulation of cells in the G2/M phase and to a reduction of the
proliferation rates in all cell lines in concentrations of ?5?g/ml NSC-631570, and ?0.6?g/ml
chelidonine.
It is worth mentioning that NSC-631570 has clearly different physical properties from
chelidonine: NSC-631570 is freely soluble in water, whereas chelidonine has very low
solubility.
35
NSC 631570 arrests pancreatic cancer cells in prophase via inhibition of tubulin
polymerisation. NSC 631570 reduces the proliferation rate and induces apoptosis in
pancreatic cancer cells (Gansauge et al, 2001).
The Committee for Orphan Medicinal Products (COMP) of the European Agency for the
Evaluation of Medicinal Products (EMEA) suggested that Now Pharm AG could establish a
biological test procedure for the determination of the biological activity of its product Ukrain
(NSC-631570), as confirmation of the stability of Ukrain in addition to the data on its primary
structure obtained by chemical, biochemical and physical methods. With this aim a
biological test procedure was developed. The experiments and procedures were performed to
1) characterise the effects of Ukrain on pancreatic cancer cells and 2) develop a routine test
procedure to ensure comparable biological activity in each new batch of Ukrain. Test criteria
were: influence on the cell cycle of pancreatic cancer cells, effect on the proliferation rate of
pancreatic cancer cells, inhibition of cell division by mitotic arrest in prophase, disruption of
tubulin filaments and inhibition of tubulin polymerization in vitro. All cell lines investigated
showed reduced proliferation rates following incubation with NSC-631570 in a time- and
dose-dependent manner indicating that NSC-631570 affects replication of these cells. In cell
cycle analysis, a G2/M arrest was observed in all cell lines tested, whereas the number of
cells in G1 or S phase remained nearly stable. Since the fraction of cells in sub-G1 increased
continuously under the influence of NSC-631570, with a time delay to the increase of the
G2/M fraction, it was supposed that a significant number of cells undergo apoptosis following
G2/M phase without re-entering the G1 phase. In order to further establish this hypothesis,
Giemsa stains of the nuclei were performed and indeed a highly significant arrest of cells in
the mitotic prophase was observed, indicating that NSC-631570 does not act on DNA- or
RNA-levels but is involved in the cell division process. Since many mitotic inhibitors act
through interaction with tubulin as the major component of the mitotic spindle apparatus,
fluorescence immunostaining with antibodies against tubulin was performed. Disruption of
the microtubule network in cells incubated with NSC-631570 was observed. To further clarify
how NSC-631570 acts on the formation of microtubules, in vitro polymerization experiments
with monomeric tubulin were performed. It was shown that NSC-631570 inhibits GTPdependent
and paclitaxel-mediated tubulin polymerization by stabilising monomeric tubulin.
The stability and similarity of biological activity of all tested Ukrain batches was absolutely
confirmed.
In a study “Enhancement of macrophage tumouricidal activity by the alkaloid derivative NSC
631570. In vitro and in vivo studies” by Sotomayor et al. (University of Miami School of
Medicine, Miami, Florida, USA, 1992) various doses of NSC 631570 and various routes of
administration (intravenous, intraperitoneal, subcutaneous) were tested.
The optimal administration route was judged to be intravenous and the optimal dose inducing
the best remission was estimated to be 4 ?g per mouse. This dose corresponds to a human
single dose of about 7-10 mg for 70 kg body weight (Sotomayor et al, 1992).
In vivo activity:
Fluorescence microscopic examinations of malignant cells (NSC-631570 has a marked
autofluorescence) show that NSC-631570 has a strong affinity to elements of the nuclei of
cancer cells but not to those of normal cells. In breast cancer patients treated with a total of 10
injections of 5mg NSC-631570 every second day before mastectomy, NSC-631570 could still
be detected in tissues removed 1 week after the last dose.
When tissues of breast cancer patients treated with NSC-631570 were examined under the
electron microscope, massive changes were found in comparison to an untreated control
group (Uglyanica et al., 1996): Under the influence of NSC-631570 the endoplasmatic
reticulum underwent fragmentation, and mitochondria became swollen with the cristae
36
damaged. In addition, the cytoplasm was also swollen with an increased number of
lysosomes, phagolysosomes and myelin bodies indicating destruction of the cancer cells.
Ultrastructures of other cells however were not affected. Treatment with NSC-631570 also
resulted in a markedly higher number of fibroblasts and extracellular connective fibres as
enzyme content, in particular in those enzymes that are key factors in the Krebs-cycle
(tricarboxylic acid cycle) and therefore in the flow of cell-respiration; these enzymes are
responsible for the generation and transfer of energy in the form of ATP, e.g., NADH, SDH,
LDH. On the other hand, the activity of glucose-6-phosphate-dehydrogenase and acid
phosphatase was increased, indicating an enhanced destruction process of cancer cells.
Susak (2004) performed a study to define histological features of pancreatic ductal
adenocarcinoma after Ukrain administration. Six non-smoking, male, 57±5 years old,
patients with histological verified pancreatic ductal adenocarcinoma with localisation in the
mid part of the gland were operated on duodenal impassability. All the patients had
previously received palliative surgical treatment with subsequent chemotherapy with
gemcitabine or 5-fluoruracil. Due to extremely strong adverse events chemotherapy was
discontinued. All the patients then received 2±1 courses Ukrain (30 mg weekly, 120 mg per
course). The last injection of Ukrain was performed 10-12 hours before the operation.
Necrosis areas squares were increased by 50-70% compared with existing previous (before
Ukrain administration) stains. Tissue sclerosis was present as well as perivascular that is not
common in spontaneous processes in the pancreas. In the sites where parenchymatic elements
predominated on stromal necrosis, the areas were especially bright. Microcirculation
disorders in the form of perivascular and perineural haemorrhages, tissue infiltration with
blood, connective tissue disaggregation, marginal erythrocytes standing and pathological
changes of vascular walls were present. The signs of fibrinoid infiltration and percolation of
malignant tissues with fibrin is a universal phenomenon in therapeutic pathomorphosis of
pancreatic cancer under Ukrain influence that leads to “immuring” of cancer cells and
preventing metastasis. Neocollagenogenesis that follows fibrinoid infiltration separate single
malignant cells or even little groups of malignant cells leading to their dystrophic changes,
for example, impossibility of mucus secretion. The other important feature was tissue
prosoplasia – increase of differentiation grade in previously less differentiated malignant
cells. This event occurred predominantly in tumours rich in vascularisation and parenchyma
elements.
From these observations it may be concluded that NSC-631570 has direct effects on cancer
cells in humans as it can be found in the cytoplasma but also that it is indirectly cytotoxic via
immunological processes, possibly changing the antigenic expression of tumour cells.
Increased cell-respiration may be the underlying mechanism of the sensation of heat in the
area of tumours reported by many patients after treatment.
Toxicity:
NSC-631570 has a low acute toxicity. The LD50 in rats after i.v. application is 43 and
76mg/kg b.w. (males and females respectively), in mice 80 and 68 mg/kg b.w. (unpublished
report of the Austrian Research Centre, Seibersdorf, Internal study code A-4483, Oct. 1998
and L-0400, May 2000). This is at least 300 times above the usual therapeutic dose in man.
NSC-631570 has no cumulative toxicity and is - in cases where no tumour is present – rapidly
excreted.
In a 6-month i.v. toxicity study with rabbits (0-negative control, 0 -negative control recovery,
0.07 -low dose, 0.30 -mid dose, 0.70 -high dose and 0.70 mg NSC 631570 /kg -high dose
recovery, groups of 6 animals each), statistically significant differences between dosed groups
and the control group were observed with regard to bone marrow (sternum) with
37
hypocellularity (mid dosed males and females, high dosed males), karyorrhexis (mid dosed
males and females), inactive megakaryocytes (high dosed males), pyknosis (mid dosed
females), cytolysis (mid dosed males) and with regard to the kidneys with proximal tubuli
epithelium degeneration (high dosed males and females). Differences also occurred in white
blood cells, with a slight increase of leukocytes, lymphocytes and bands in the high dose
group (both sexes) after 4 months. Haematocrit and reticulocytes were also slightly increased
in the high dose group. Occasionally, other differences between the groups were observed but
can be considered as not medically relevant (Austrian Research Centre Seibersdorf, 2001).
As was previously reported by Benninger et al, 1999, greater celandine drugs can lead to toxic
liver damage when given per os. Studies were recently performed to detect the possible
hepatotoxic activity of Ukrain. In several recent studies Ukrain was demonstrated to be free of
hepatotoxicity. In a recent study Ukrain administered at a daily dose of 2 mg/kg to male
Wistar rats had a slight activating effect on the drug metabolising enzymes of the liver
(Zverinsky et al, 2003). It should also be mentioned that in more than 20 clinical studies
performed with Ukrain no signs of toxic effects on the liver were found. Quite the contrary,
the compound can be successfully used to protect the liver from toxic damages in the
acetaminophen-induced hepatitis model in rats (Levina et al, 2004).
The aim of the study performed by Muller (2004) was to examine to potential for Ukrain
solution for injection (Ampoule 5 mg/5 ml) to induce hepatotoxicity in the rat. The test
substance was administered by intraperitoneal injection to two groups of 5 male and 5 female
Sprague-Dawley rats for 5 consecutive days. Injections were made at the same time each day.
The doses selected were equivalent to the maximum (0.3 mg/kg/day) and 5 times the maximum
(1.5 mg/kg/day) human daily dose. Followed investigations were performed: body weights
and body weight gain, observations in life, gross liver pathology, liver histopathology and
plasma hepatic enzymes levels. All animals survived until the scheduled termination of the
study. Body weights and body weight gain were normal. All animals were normal throughout
the study. There were no increases in plasma levels of alanine aminotransferase (ALT),
aspartate aminotransferase (AST) or alkaline phosphatase (AP) that were considered
treatment-related. Gross liver pathology was normal. Histopathological examination of the
liver revealed no lesions considered to be treatment-related. There were no increases in
plasma levels hepatic enzymes that were considered treatment-related.
There were no treatment-related sex differences. It was concluded that intraperitoneal
injection of Ukrain at 0.3 mg/kg/day or 1.5 mg/kg/day for 5 consecutive days did not induce
hepatotoxicity in the rat.
The aim of the study by Zverinsky (2004) was to compare the effects of thiotepa, greater
celandine alkaloids and Ukrain on the morphology of the liver and activity of liver enzymes in
rats. Total 88 Wistar rats, weighed 180-220 g were used in the study. Rats were kept on the
standard laboratory feed, at natural light and room temperature 23-250?. Animals were
divided into 11 groups, 8 animals each, and were treated as follows: I and II – control; III –
thiotepa, 3 mg/kg/day, intraperitoneal, for 10 days; IV – alkaloids, 20 mg/kg/day,
intraperitoneal, for 10 days; V - alkaloids, 10 mg/kg/day, intraperitoneal, for 10 days; VI –
alkaloids, 5 mg/kg/day, intraperitoneal, for 10 days; VII – Ukrain (concentrated) 20
mg/kg/day, intraperitoneal, for 10 days; VIII – Ukrain (ampoules) 10 mg/kg/day,
intraperitoneal, for 10 days; IX – Ukrain (ampoules) 5 mg/kg/day, intraperitoneal, for 10
days; X – thiotepa, 11 mg/kg/day, intraperitoneal, for 3 days; XI – thiotepa, 6 mg/kg/day,
intraperitoneal, for 3 days. Thiotepa solution as well as celandine alkaloids solution were
prepared immediately before administration. The injections was performed daily at noon. The
animals were sacrificed 24 hours after the last administration. Blood was sampled. A part of
the liver was removed for morphology and biochemistry examination, the remainder was
38
washed out with 1.15% potassium chloride solution. Microsomal and cytosolic fractions were
then separated by means of differentiated centrifugation. Activity of acid phosphatase and
alkaline phosphatase, triglycerides and total cholesterol, choline esterase activity, thymol test
and endogenic intoxication were measured.
Administration of thiotepa at a dose of 11 mg/kg body weight for 3 days caused the death of
all treated animals on the day 4. Administration of thiotepa at the dose of 6 mg/kg caused the
death of all animals on the day 6 or 10. The treatment with thiotepa at all doses caused
significant decrease of the body weight. In the group X (thiotepa, 11 mg/kg i.p. for 3 days)
total cholesterol, triglycerides and acid phosphatase activity increased by 88, 133 and 196%,
respectively (all values p<0.0001). Low molecular tyrosine- and tryptophan-containing
peptides increased by 168 and 65%, respectively. Significant decrease in the concentration of
reduced glutathione (GSH) by 34% was also revealed. In the groups III and X, superoxide
dismutase activity increased significantly compared to the control group. A minor decrease of
this enzyme activity was revealed in the group XI. Other parameters did not statistically
differ compared to the control group.
In the groups treated with celandine alkaloids (5, 10 or 20 mg/kg/day), the most pronounced
changes of biochemistry parameters were revealed in the group IV (20 mg/kg/day). The body
weight increase was by 36% lesser compared with the control group. Serum choline esterase
activity decreased by 38%, indicating the damage to the liver cells. Low molecular
tryptophan peptides increased by 43% and GSH decrease was revealed. Serum triglycerides
were significantly increased. Superoxide dismutase activity was increased by 36% in the
cytosolic liver fraction. All alkaloids treated groups revealed decrease of GSH in the liver
and positive thymol test. Thymol test is considered as one of the most sensitive among liver
function tests. Such results of the thymol test indicate the damage to the liver cells in the
groups treated with alkaloids. Other parameters did not statistically differ from these in the
control group.
After administration of Ukrain at the doses of 5, 10 or 20 mg for 10 days, superoxide
dismutase activity was increased in blood as well as in the liver; GSH was decreased in the
liver. the These effects were non-dose-dependent. Other parameters did not statistically differ
from the control group.
It was concluded, that administration of Ukrain oppositely to thiotepa and alkaloids in similar
doses has no hepatotoxic activity. It confirms the previously findings that the drug possesses
other pharmacological properties comparing with the start components for its synthesis (see
Results of the Ukrain, thiotepa and alkaloids testing at the NCI, Maryland, Bethesda, USA).
Reproduction studies have given no indications of teratogenic, mutagenic or cancerogenic
properties of the preparation, even in doses, which were 100 times larger than the therapeutic
dose. NSC 631570 does not induce sensitisation and is also not genotoxic (Chlopkiewicz et al,
1992; Wyczolkowska et al, 1992; Austrian Research Center Seibersdorf, 1999; Austrian
Research Center Seibersdorf, 2000).
Pharmacokinetics:
In a pilot study, NSC-631570 was administered to 6 healthy men at a dose of 20 mg / 20 ml,
undiluted, as a slow intravenous injection; plasma concentration was determined 5, 15, 30, 45,
60, 90, 120, 150, and 180 min after administration, urine was collected over 24 hours. In this
study the half life of NSC-631570, t1/2 ??was 27.55?2.45 minutes and the apparent volume of
distribution (V) was 27.93?1.38 l. Around 47% of NSC-631570 was found in the urine, more
than half of the amount being eliminated during the first 6 hours (Uglyanica, Karavay,
Nefyodov, et al., Unpublished report 1999). No significant changes with regard to results of
physical examination, laboratory parameters and ECG were reported.
39
Binding to human plasma proteins seems to be insignificant at around 2% (Doroshenko et al.,
2000).
In another study (Danysz et al, 1992) NSC 631570 was administered to 19 healthy volunteers
intramuscularly or intravenously at doses from 5-50 mg every one, two or three days for up to
40 days. In all cases NSC 631570 was generally well tolerated. Some volunteers reported
localised pain with a burning sensation during intramuscular injection. The pain disappeared
spontaneously after about two minutes. Drowsiness during the day was also reported by some
volunteers. There were no notable changes in clinical conditions. All haematological,
chemical and urine parameters studied revealed only minimal fluctuations within normal
range.
Spasmotic and cholagogic actions of the preparation were reported by two volunteers who
had mild dyspepsia. During the study period these symptoms disappeared.
It is worth emphasising that during the period of NSC 631570 administration, numerous
catarrhal and influenza infections were prevalent in the study area. However, no such
infections were observed in any of the volunteers taking NSC 631570. A tendency to an
increase in the CD4/CD8 cell ratio was noted.
After drug administration all volunteers were in good or even better general states of health
than before therapy. At the beginning of NSC 631570 administration some volunteers felt
slight fatigue, a slight increase in body temperature and increased thirst and enhanced
urination. The results of this study showed no evidence that NSC 631570 had any harmful
side effects.
From animal experiments it may be concluded that NSC-631570 concentrations are highest in
tumour tissues (2.84-fold higher than in plasma) followed by normal liver and kidney tissues;
the lowest concentration was found in muscles and the brain. NSC-631570 does not
significantly cross the blood-brain barrier (Doroshenko et al, 2000).
NSC-631570 can be detected in tumour tissues within minutes after i.v. injection and
concentrates in the nucleoli of tumour cells; healthy cells remain unaffected. The presence of
NSC-631570 in tumour tissues can be demonstrated up to 19 days after injection by means of
its autofluorescence under UV light. However, NSC-631570 is rapidly excreted from healthy
tissues.
No dose-limiting signs of accumulation were observed during repeated injections of NSC-
631570.