Vol. 31, No. 7 (2019)
Isolation and Structural Characterization of Degradation Products of Finasteride 1515
The LC-MS data was collected using an Agilent 1290
Infinity LC System (Agilent, Santa Clara, CA, USA) with
ACQUITY BEH C18 (2.1 × 50 mm, 1.7 µm particle size)
column; column temperature was maintained at 45 °C; Binary
pump with mobile phaseA: 0.1 % formic acid in water; mobile
phase B: 100 % Acetonitrile; T/% of B: 0.0/2.0, 0.2/2.0, 1/20,
5.0/80, 6.0/98, 8.0/98, 9.0/2, 10/2; Diluent: Mobile phase; flow
rate 0.8 mL/min; Detection: 215 nm.
Preparative HPLC was carried out usingWaters quaternary
gradient module 2545, Waters photo diode array detector
module 2998; Waters Auto sampler 2707; Waters Fraction
collector III (Waters Corp., Milford, MA, USA); column: Sym-
metry C8 (300 × 19 mm) 7 µ, mobile phase A: 0.1 % formic
acid (Aq); mobile phase B: Acetonitrile: T % of B: 0.0/10,
8.0/50, 11/50, 11.1/98, 12/98, 12.1/10, 15/10; flow rate: 20
mL/min.
0.5 N HCl
Finasteride
500
400
300
200
100
0
DP-2
DP-3
DP-1
2
4
6
8
min
1000
800
600
400
200
0
Finasteride
0.5 N NaOH
2
4
6
8
min
Finasteride
5 % H2O2
1000
800
600
400
200
0
2
4
6
8
min
HRMS data of all the degradation products was acquired
with Waters micro mass Q-TOF equipped with an electrospray
Ionization (ESI) source. Samples were operated in positive
and negative mode to enable detection of degradants. Leucine
encephalin (m/z: 555.62268 Da) was used as reference lock
mass calibrant to achieve the typical mass accuracies with Mass
Lynx software. The optimal conditions of analysis were as
follows: capillary voltage of 3.5 kV, sample cone at 25 V and
the scan range from 100 to 2200 Daltons. The source tempe-
rature was 120 °C and the desolvation temperature was set at
350 °C. Nitrogen was used as the nebulization gas at flow rate
of 750 L/h.
Fig. 1. Acid, base and hydrogen peroxide degradation chromatograms of
finasteride drug substance
Isolation of acid degradation products: Preparative
HPLC fractions containing degradants under acidic condition
were collected and lyophilized. The products were labeled as
DP-1, DP-2 and DP-3. The extensive analysis of HRMS and
1D, 2D NMR data confirmed the structures of the degradation
products. Literature survey revealed that DP-1, DP-2 and DP-
3 are novel compounds. Structures of finasteride and its degra-
dation products in various stress conditions are shown in Fig. 2.
Structure elucidation of DP-1: The HRMS showed a
protonated molecular ion peak at m/z 336.2178 [M+H]+ corres-
ponding to m.f. C19H29NO4. The 1H NMR spectrum revealed
that DP-1 had one carboxylic acid proton (observed at 11.9
ppm) and one hydroxy proton (observed at 4.92 ppm). The
absence of olefinic and tertiary butyl protons in the NMR of
DP-1 suggested hydration of the alkene and hydrolysis of the
tertiary butyl amide in the side chain under acidic condition.
To ascertain the position of hydroxyl 13C HMBC was performed
and the DP-1 shown in Fig. 3 was found to be the correct struc-
ture. In 13C HMBC, H-6 (3.67 ppm) showed correlation with
C-2 (169.7 ppm), C-1 (38.2 ppm), C-5 (38.8 ppm), C-4 (52.7
ppm) and C-18 (11.8 ppm) as shown in Fig. 4. H-6 showed
correlation with C-18 indicating that hydroxyl group is located
on C-6 and not on C-1. The complete assignments of 1H and
13C chemical shift values for DP-1 are tabulated in Table-2.
The observed regioselectivity is in line with the established
mechanistic prediction of reactivity of α,β-unsaturated carbonyl
compound. The nucleophilic oxygen of water molecule is
expected to add to the electrophilic β carbon of the lactam.
Structure elucidation of DP-2: The HRMS showed a
protonated molecular ion peak at m/z 317.2232 [M+H]+ corres-
ponding to m.f. C19H28N2O2. DP-2 showed all protons that are
present in finasteride except tertiary butyl group protons in
the side chain. DP-2 indicated primary amide protons (observed
at 6.71 ppm and 6.91 ppm) suggesting the hydrolytic cleavage
of the tertiary butyl amide –CONHC(CH3)3 to –CONH2. This
can be attributed to the ready formation of stable tertiary butyl
carbocation under acidic condition. Carbonyl carbon (21-C)
of the primary amide was seen at 173.8 ppm in 13C NMR. All
proton versus proton correlations in COSY and all proton
versus carbon correlations in HSQC and HMBC matched well
1D and 2D NMR spectra of finasteride and its degradation
products were recorded in DMSO-d6 on 400 MHz Bruker
1
13
NMR spectrometer. Chemical shift values of H and C are
measured on δ scale in ppm comparative to tetra methyl silane
(TMS) as internal standard. The spectra were referenced to
δ 0.00 ppm in 1H NMR (TMS) and δ 39.50 ppm in 13C NMR
(DMSO-d6).
Stress methods: Compound finasteride was subjected to
stress conditions as per ICH guideline Q1A (R2)5 to identify
the potential degradants that could be formed during the long-
term storage. The details of the stress methods are shown in
Table-1.
TABLE-1
STRESS CONDITIONS FOR OPTIMUM DEGRADATION
Stress
condition
Concentration of
stressor
Exposure
condition (°C)
Duration (h)
Acid 0.5 N HCl
Base 0.5 N NaOH
5 % H2O2
60
60
60
8
8
8
Hydrolysis
Oxidation
RESULTS AND DISCUSSION
After 8 h of heating, degradation products were analyzed.
1 mL of the resultant acid degradation solution was dissolved
in mobile phase and a sample of 10 µL was subjected to LC-
MS analysis. Drug solution treated with acid showed three
degradants and no degradation of the drug was observed in
peroxide as well as base treated drug solution as shown in
Fig. 1. Acid treated solutions were taken up for isolation of all
the three detected degradants.